American Health Information Community
Electronic Health Records Workgroup #20
Tuesday, December 4, 2007
Disclaimer
The views expressed in written conference materials or publications and by speakers and moderators at HHS-sponsored conferences do not necessarily reflect the official policies of HHS; nor does mention of trade names, commercial practices, or organizations imply endorsement by the U.S. Government.
>> Judy Sparrow:
Good afternoon. Welcome, everybody, to the 20th meeting of the Electronic Health Records Workgroup. Again, a reminder: This is Federal Advisory Committee, and we are being broadcast over the Web. For Workgroup members on the telephone, please remember to identify yourselves and speak clearly and distinctly, and please mute your telephones when you’re not using them. There will be a period at the end of the meeting for the public to make comments.
With that, I’ll read the list of the members on the telephone. We have Jonathan Perlin, one of the Co-chairs from HCA Health Care; Bonnie Anton from the University of Pittsburgh; Robert Juhasz from the American Osteopathic Administration; Jack Price from HIMSS; Mike Kappel from the Vet, from McKesson; Howard Isenstein from the Federation of American Hospitals; and in the room, we have...?
>> Nhan Do:
Nhan Do. I represent the Department of Defense.
>> Richard Hays:
Rick Hays from the American College of Cardiology.
>> Jason Kreuter:
Jason Kreuter from the American College of Cardiology.
>> Karen Bell:
Karen Bell, ONC.
>> Judy Sparrow:
Did I miss anybody on the telephone?
>> Reed Gelzer:
Reed Gelzer from Advocates for Documentation Integrity.
>> Judy Sparrow:
Okay, anybody else?
>> Robert Tennant:
Rob Tennant from the Medical Group Management Association.
>> Judy Sparrow:
All right. Okay. With that, I will turn it over to Dr. Perlin.
>> Jonathan Perlin:
Well, good morning to those west of the central part of the country. Good afternoon to those in the Eastern Time Zone. It’s hard to believe we’re at our 20th meeting, but let me start off by really thanking each and every one of you for your terrific work, not just over the long haul. But I had the opportunity to chat with a few colleagues before the call, and I think everyone involved should feel rightly proud of a terrific collaborative discussion that led to a draft a very good draft of recommendations for Secretary Leavitt regarding e-prescribing. I think there’s a great deal of enthusiasm around the work, because in the process open process, as per the FACA process itself, it really was a very collaborative and consensus-building discussion that allowed all actors across the spectrum of interests, from provider to payer to physician to allied health professional to vendor to pharmacy to pharmaceutical community, etc., really to come together around the set of recommendations that I think everyone can feel are fair, appropriate, and serve as an important anchor in terms of really stabilizing the ability to make progress in reducing the error-prone process of prescribing, as it exists today, but also serve as a catalyst to building on the broader goal, which Secretary Leavitt really set out for us in his initial charge to AHIC and the Workgroup, which was to advance the implementation of electronic health records. So on behalf of Lillee and myself, let me thank you very, very much for all of the work that went into that.
Today we have a very ambitious agenda. I just note that we will initiate, you know, after our usual business with the discussion of recommendations for workforce, an area that we’ve taken testimony, and with the great work from ONC staff, have actually a series of recommendations. And let’s come to similar consensus around those recommendations, and let’s sharpen those if we think they need sharpening so they can be actionable to as recommendations from AHIC to the Secretary, and ultimately for the Secretary or other actors that need to respond to these recommendations.
We then set out to review the next batch of work that refines some of the use cases and areas that we’ve addressed. We started, as you may recall, looking at how we might advance electronic health records through laboratory as the first use case. We’ll look back at some of that work for the next round of activity, as well as build and expand for the next bunch of activity.
And then we move to really an area that I think is equally important. It has many different dimensions. Most important dimension that it has is to improve the administrative efficiencies and validity and reliability of health information. It has many sponsors and interests, including not just data integrity from a clinical standpoint, but data integrity in a standpoint of fraud and abuse as well. But the main focus of that important area has to do with the improving the efficiencies and the integrity of electronic information that’s captured.
Let me first, as first order of business, ask if there are any comments on the minutes from the October 23 meeting. (Pause) Okay, hearing none, then we will accept those minutes.
Before we go to the recommendations for workforce, I want to both acknowledge her terrific leadership but invite to comment on the current status of the recommendations from the Electronic Health Records Workgroup to AHIC, Dr. Karen Bell. And thank you, Karen, so much for the tremendous coordination of really an incredibly robust national discussion on this topic that culminated with a concise set of recommendations. Perhaps you might update us on the process and any other comments you might want to share on the recommendations.
>> Karen Bell:
Thank you so much, Jonathan. This was indeed a yeoman’s effort by everyone on the Workgroup here as well as the AHIC. And it was very well-represented by Lillee at the AHIC and with your support. There is a letter that has now gone from the AHIC to the Secretary, and copies of that letter have been distributed as of noon today to all of the Workgroup members so that everyone can see how the final recommendations were crafted.
What I can say is that the first recommendation, which was to request that the Secretary move forward with seeking authority from Congress to mandate e-prescribing, was in fact accepted. The wording was changed a bit so that there was not so much emphasis on Medicare Part B, but that the it was broad enough basically to cover e-prescribing in general.
Second, recommendation 2.0 reads, “Prior to exercising authority to mandate e-prescribing, the following requirements should be met.” And literally all of the requirements that were articulated by the Workgroup are included in this letter. There is one interesting change, however. After much discussion and it went the full hour it was clear that the flexibility recommendation was the one that I think everyone came to agreement on, with respect to allowing movement in certain circumstances. So we moved the recommendation about “Flexibility must be maintained, since mandated e-prescribing may not be applicable to all patients, all prescriptions, and all circumstances” up front as Recommendation 2.1, with the others following.
So again, I can’t underline how grateful I think everyone is for all the work this Group did to craft that set of recommendations. And the Secretary in particular read the letter very carefully, was very interested in it, and was quite clear that he felt that there was enough information here that he could the AHIC could come to some agreement and recommendations could be made. And they were. So again, many thanks to all of that you contributed all your time and effort on this letter and this effort. And thank you, Jonathan.
>> Jonathan Perlin:
Well, thank you, Karen. And really, I think a debt of gratitude is particular to Lillee Gelinas, who also did absolutely heroic work in bringing this together, and to each of you.
Lest people think that this work doesn’t gain traction just in these quarters, these are discussions that are going on in parallel. And I shared with Karen that literally at the same time that our work was occurring, the President’s council, the advisors of science and technology, were discussing similar activities. And interestingly, their entirely independent conclusion with respect to advancing the EHR agenda included e-prescribing as one of the central things that should and could occur initially. So just that lest you feel that sometimes these things don’t change turn into reality, I feel completely confident that we will in fact have created a document and a set of recommendations that in the near future, in the life of health information technology, we will see being translated into reality in the country. And I think all of us should feel good about that, particularly those of us who feel passionately about the ability to improve the safety, effectiveness, efficiency, compassion of health care.
With that, we still have this other hurdle. And let’s move to our first order of business, and that is to try to wrap up and finalize as cogently as the group process allows for e-prescribing, Let’s see if we can’t turn that to the recommendations for workforce. I need to tell you that as Alicia tees up the slide the first slide in your package, I know the package was very robust this month, but there is a four-page PowerPoint that’s entitled “AHIC EHR Workgroup Workforce Recommendation Concept for Workgroup Discussion.” So that is the document that we’ll be working from at this point.
I need to tell you that this discussion is, for me, not theoretical. We are in the process of trying to advance our electronic health record and health information technology infrastructure across our large organization, and we simply cannot recruit the peak types of people that we need to have to really accelerate and implement. And I imagine for each of you, these you may be facing similar challenges. And it’s not surprising when we heard the magnitude from the testimony, from Linda Kloss and Don Detmer and other colleagues who talked at some length about deficiencies in each of the respective disciplines, from individuals who come from pure informatics backgrounds, individuals who span clinical and informatics background, to individuals who can support software/hardware infrastructure, clinical informatics, etc.
And so let’s go through these recommendations, not with a view toward wordsmithing, but importantly, I think there are two outcomes that I’d hope we can achieve. One is that this list is appropriate in terms of the recommendations it provides, that it captures the recommendations that the Group us, the Workgroup, as well as those individuals who testified to us, offered, and that those recommendations be actionable, because (1) they’re clear in terms of what the outcome is; (2) they’re based on quantified or quantifiable data. They’re not conjectured; they’re in general, but either extrapolate from evidence of similar professional development activities that have worked or extrapolate from surveys of numbers in particular fields.
Let’s start with the we’ll let’s there may be other ways to thematically group, but for the sake of organization, let’s start with Recommendation A on research. And I’ll read that in case anyone is not seeing it. I know my computer isn’t quite caught up with the PowerPoint. But on recommendation A, it reads, “Additional research should be funded for expansion of health IT career pathways within all health care disciplines.” That’s a pretty broad recommendation. Is there any degree of specificity or examples that you would ask be included or would like to instruct ONC staff to add in the discussion around this when it, too, turns into a letter to AHIC ultimately for the Secretary’s action?
>> Karen Bell:
And this is Karen Bell. If anyone feels we I think we have nine different options here. If anyone feels that perhaps this is too broad or is not one that the particular Workgroup would like to move forward, please also state your reasons for that, and we’ll consider whether or not this should even be there. Thank you.
>> Linda Fischetti:
Hi, this is Linda Fischetti. I’ve just joined. And first of all, my apologies for being a little late today, but then also I do agree that this is very broad. I think that if we were to continue to keep this language in here, we would need to give it a bit more context in terms of are we talking about, you know, which career pathway for an informaticist do we really want to approach? So what is unique and distinct and different and is a different skill for an applied informaticist? What is unique and different and requires a different skill set for a research informaticist versus somebody who is at an executive level running informatics programs? And I think that unless we give this some categorization or chunking that it would be too broad to be actionable, and whatever came back might be interpreted it would require interpretation, but then that interpretation could take it far afield from what we originally intended.
>> Jonathan Perlin:
I think that’s a terrific comment, Linda. Are there you alluded to some categories. Do you want to name any specifically?
>> Linda Fischetti:
For me specifically, the applied informatics skill set is, in my mind, interpreted as someone who is actually working in the clinical environment, doing either configuration or implementation of the systems. I think that right now, that domain knowledge is best represented in the vendor community, because they’re the ones that are out doing the actual applied informatics at this point in time. But now that applied informaticist also needs to be supported by all of the other informaticists that are working in the environment. Specifically, research has come online biomedical research as well as more health outcomes research. And then, of course, what we’re looking for right now are people with more of that executive skill set, so somebody who, yes, is an informaticist but then can function at more of the executive level related to strategic planning, running of programs, managing the managers, that sort of thing; and then of course, you know, dealing with the development of the systems, you know, the increasing body of knowledge of being able to put evidence-based practice into systems to be able to draw the outcomes that you’re looking for in your clinical environment. That’s a little higher than applied informaticist. So from the way I look at it, there are people we need to have well-rounded professionals who can either cross domains or can specialize within the different domains.
>> Blackford Middleton:
Linda, could I offer an observation, too? Good afternoon. It’s Blackford Middleton. Can you hear me okay?
>> Jonathan Perlin:
Yes, we can. Thanks, Blackford. Welcome.
>> Blackford Middleton:
Okay. You know, I guess along Linda’s spectrum of categories, I might offer the other thought. That is that, in addition to the IT professional, whether it’s the development or the implementation side, and in addition to the applied clinical informatician, I’d offer two other thoughts. One is that informatics training and the effective use of information technology should be part of the training curriculum of every physician in the country, and perhaps every physician and nurse as well. So that’s a different slant, more across the horizontal thought that all physicians need to know and nurses need to know how to use these technologies. So that’s a different approach. The other thought, though, is that there’s a basic scientist in addition to the basic science clinical informatician, so that we need to have the basic scientist as well in training so that we can continue to advance and develop the field in informatics.
>> Jonathan Perlin:
Blackford, those are terrific. And let’s add basic science to this list, and I can capture that there. I think we can at the risk of some duplication, we can allude to this under this category. And you’ll be pleased to know that Recommendation D that we come to right now, it’s reading, “Have informatics competencies identified and developed for health profession disciplines. Have these incorporated into academic programs and health professional-licensing activities.” So I think your point bears amplification in this document, perhaps subordinate to this, but also as a free-standing recommendation. It’s just hard to imagine developing successful and appropriately educated health professionals without basic competencies in the future. So we absolutely captured that probably in its own recommendation.
>> Blackford Middleton:
Thank you.
>> Jonathan Perlin:
Other recommendations? I think we can envision that this would then, by extrapolation, be applied to individuals who come to health informatics from an informatics background and people who come to health informatics from a clinical background, sort of a convergence there. And I also hear that as a theme that we want to incorporate.
>> Karen Bell:
This is Karen Bell. I recall from the discussions we’ve had that I know that and Blackford, perhaps you could jump in here, or perhaps others on the call that AMIA and AHIMA have done a lot of work on describing the different pathways for development of informaticians. That included the research community, the public health community, the clinical knowledge management community, and the translational bio-informatics community. What I’m wondering is whether or not the Group feels that we could use that as a starting point to define the different approaches that could be used, attending to Blackford’s earlier comment and having a separate recommendation around the educational structure to support informatics, and then pull in some of the ideas that have just been shared and put them under those categories, particularly under the clinical knowledge management one, which is, I think, the one that health care providers would be most interested in.
>> :
I think that’s reasonable.
>> Jonathan Perlin:
Okay, well, hearing no other comments, I’m going to take that as consensus. And Karen, if you could perhaps for staff, if you could restate that in the form of a recommendation...?
>> Karen Bell:
Additional research should be funded for expansion of HIT career pathways within the realms of informatics with respect to research, public health, clinical knowledge management, and translational bio-informatics with particular attention to and we’ll wordsmith this a little bit more some of the things that Linda mentioned system development, so with respect to evidence-based practices. And Alicia has taken the notes, so maybe I should let her add those. What are some of the other things we’ve added here?
>> Alicia Bradford:
She mentioned development and applied informatics. We haven’t really well, we mentioned clinical and research. The other piece was the executive leadership.
>> Karen Bell:
Right.
>> Jonathan Perlin:
Okay. Well, I think that’s terrific, and I think the lack of objections for it indicates consensus. I think that’s (inaudible with Robert Juhasz interrupting) segue to the next.
>> Robert Juhasz:
Before we move, this is Bob Juhasz. I wonder if we should add “and clinical effectiveness,” just because of the emphasis being put on that and, you know, from the standpoint of after evidence-based medicine.
>> Blackford Middleton:
I think that’s a terrific recommendation.
>> Karen Bell:
Excellent suggestion.
>> Michael Kappel:
Before we leave this topic, this is Mike Kappel. I’m a little confused as to the construction of the particular recommendation. Are we really calling for the additional research as required, or are we really calling for the development and expansion of HIT career pathways beyond what’s been done by groups like AMIA and AHIMA?
>> Jonathan Perlin:
Yes. (Laugh)
(Multiple speakers)
>> Michael Kappel:
(Inaudible) call for additional research, it’s sort of a red flag that we don’t have enough knowledge about this. Do we have enough knowledge, or are we calling for the expansion, or are we calling for more research to be done before we create pathways?
>> Jonathan Perlin:
Okay, I appreciate that clarification. I don’t think the I don’t think the intent is to create ambiguity by calling for more research to see if these points are in fact valid. I think what the call is, is to support more research to create career paths and further develop the field.
>> Michael Kappel:
Okay, I would I think that’s a reasonable goal and a recommendation.
>> Jonathan Perlin:
Okay, but thank you very much, because I think that was a nuance that just, previously, we had missed. Thank you.
>> Karen Bell:
And again, after we get these crafted, there will be ample opportunity for more wordsmithing online for the next several weeks.
>> Howard Isenstein:
This is Howard. I hate to be annoying, but we use the “should be funded” I mean, shouldn’t I mean, are we saying the Federal Government should fund additional research? I mean just to be a wordsmith here.
>> Karen Bell:
There are multiple ways of funding research. And at this point, if we leave it fairly broad, then the AHIC can have discussion about who should actually be doing the funding. And that may be our best option at this point, unless someone else has
>> Howard Isenstein:
No, that’s fine. I just can just see Lillee and Jon getting beat up by (laugh) some AHIC commissioners at the next meeting, like, “What is this about? Who’s who should be paying for this?” That’s all. I wanted to spare them that.
>> Carolyn Clancy:
Well, Howard, this is Carolyn Clancy, I appreciate that point a lot, so I’m going to be a little bit provocative for the purposes of discussion here. Are we making an argument for the full-employment act for people with training and expertise in informatics? Or are we, I hope, helping to get to a place where we can talk about what problem we’ll be solving?
>> Jonathan Perlin:
Well, I hope it’s the latter. And the problem that we’re nominally addressing right now is how to create acceleration of workforce in health IT. How much you envision perhaps you don’t agree with the recommendation that support for research in health IT offers an attractor in the form of (inaudible) jobs for future for individuals with academic aspirations with interest in health IT.
>> Carolyn Clancy:
Well, I think the model that I’ve often heard academic referred to geriatrics. We kind of discovered that we had an aging population 20-odd years ago, or discov you know, kind of re-appreciated the importance, and put a whole lot of money into it, and, you know, pretty soon, we had experts and a critical mass of folks at most academics centers. And that was a good thing. And I know that AMIA has thought about this as well. And I’m not saying that we don’t need it, but I think it’s got to be more specifically crafted. And I guess another way of saying this, Jon, is, what I see all the time, by way of thinking about health care quality, safety and so forth, is that we may have a shortage of folks with IT expertise right now, it feels like we do where I work, but that’s a much different story the but what we really have a problem with are people who can talk to each other. I mean, it’s much more almost of an interdisciplinary kind of focus, where people can understand what IT can do for them so that the IT folks actually know what the clinical enterprise needs and vice versa. And somehow, I don’t see these recommendations actually getting to that.
>> Jonathan Perlin:
Yeah. I think that’s a great point. That’s very it’s thematically aligned with Blackford’s earlier point that our clinicians really need to have IT as part of their curriculum, and at least one segment of informaticists with aspirations toward health should have some health exposure. And maybe that is a clearer framing of, or amendment to, that recommendation. Would you agree?
>> Carolyn Clancy:
Yeah. No, I would like that better.
>> Jonathan Perlin:
Okay.
>> Karen Bell:
The one thing is, one of the other pieces here about the research is having a better understanding of what the workforce needs truly are. We have some idea of what the workforce is to support public health informatics, we have some idea of what it takes to get physicians to adopt an EHR, but we really don’t know the extent of the workforce necessary to support widespread use of HIT in addition to assuring that all of the possible users are engaged. So maybe this would be a way of really moving in that direction. And that would that sounds actually, as being pointed out to me, that that may be more like Recommendation G, where we’re looking at we’re doing the research. And maybe it’s an issue of reordering.
>> :
Yeah.
>> Karen Bell:
If we look at Recommendation G, we perhaps make that a little bit clearer that that’s research to establish workforce needs, and then Recommendation A could move more towards funding expansion of career paths based on the information or based on the knowledge learned from the research on the workforce needs.
>> Jonathan Perlin:
Okay, that is a clearer that is clearer.
>> Blackford Middleton:
I think that helps a lot.
>> Karen Bell:
And then adding Carolyn’s comments as well would be very helpful.
>> Jonathan Perlin:
Before we leave just going back and forth to Jay before we leave the research, Carolyn in particular, any other recommendations or discomforts with either Item B or discussion thus far?
>> Carolyn Clancy:
Well, I guess if I were being very focused, I would narrow C to talk about folks who are working in safety net organizations or in underserved areas or something along those lines.
>> Jonathan Perlin:
That’s a terrific suggestion.
>> Carolyn Clancy:
I mean, you know, if this were part of the National Health Service Corps, that’s something that the Secretary could move, the policymakers could act on, and frankly, they probably need it more. One could make an argument about whether we’re just making interesting recommendations for what the market ought to pick up.
>> Jonathan Perlin:
I think that’s a terrific recommendation at two levels: (1) in terms of the social maze, but (2) in terms of attractability of programs that can be affected by the Secretary without other interventions.
Any comments or discussion on that? (Pause) Okay, hearing none, let’s take that as specific recommendation as you stated it.
Okay, Item D Blackford and Carolyn have both emphasized the importance of (inaudible), and I think we want to make sure that the D, as ultimately wordsmithed, is bidirectional, indicating the informatics competencies in health professionals and health competencies in interested IT professionals. And I think Linda Fischetti earlier made the point that “Don’t forget: It’s also an executive management ranks and planning and others, so without diluting or the two clear convergen unnecessary convergences my own addendum there would be that there has to be some academic curriculum available for health professionals across the spectrum of the administration as well to be able to get to Carolyn’s point have a conversation where words have common meaning.
Other comments on Recommendation D?
>> Robert Juhasz:
This is Bob Juhasz. I just think that this is also an opportunity for, you know, just a collaborative educational experience, because clinical informaticists really are a broad spectrum of people. So, you know, it’s also an opportunity to actually, you know, move that agenda forward as far as having people that would be, you know, developing new ways of care as well, so...
>> Linda Fischetti:
We may also, in D, want to reflect some of the work that’s currently taking place in health care discipline so we know, for example, that ANA has a scope and standards of nursing informatics standard practice as well as a certification in nursing informatics. We also know that AMA is beginning to work, I believe, with Robert Wood Johnson or under a grant by them to begin to develop standards of practice. So we may want to show in our language that we’re aware of and reflect existing professional disciplinary activities as well.
>> Jonathan Perlin:
I think that segues well with the next recommendation in particular.
>> Linda Fischetti:
Okay, great.
>> Karen Bell:
I have one question on this one, in that it has to do with identifying not so much as identifying the competencies. I think there are ways that that can be done even through efforts of the Federal Government. The issue about having these incorporated in academic programs and health professional-licensing activities is a little bit more difficult. And so I’m just wondering if anyone has any comments about you know, we’re making these recommendations to the Secretary. How might he move forward with actually doing these things? Or how might he how might we work with folks in the private sector to move forward on this?
>> Robert Juhasz:
It might be this is Bob. It might be, you know, I mean, working with the large academic organizations, whether it’s, you know, the AANC or, you know, the accrediting bodies to look at whether or not these have been incorporated into their educational programs.
>> Karen Bell:
So that perhaps we can at least begin discussions with the academic institutions about how what they’re actually doing and because I think that would be an important part of it: what they’re actually doing right now, like best practices might be, and how we may be able to generalize those best practices across a wider bloc of the academic community.
>> Jonathan Perlin:
Sure. I think that’s a great theme. I heard greater specificity in Bob’s comment, I thought, which was to initiate discussion with the accreditation bodies so that it’s sort of a little more focused on academics institutions and also for the implied traction as an ultimate component of accreditation. I might I want to make sure I’m hearing correctly and stating fairly.
>> Robert Juhasz:
Right.
>> Karen Bell:
Okay.
>> Jonathan Perlin:
Karen, is that agreeable?
>> Karen Bell:
Seems that that may be the way to go.
>> Linda Fischetti:
Karen, this is Linda Fischetti. And of course, I’m being opportunistic here, in that I am a federal employee and have a number of informaticists working for me and am also very interested in growing the federal informatics discipline. So with that, if you’re asking specifically what can the Secretary of Health and Human Services do, possibly just making monies available for training, education of the existing staff, mentorship, fellowship with staff that we have not yet hired but are coming up through the academic ranks, you know, etc., etc., etc. So we could put more of a mentorship/fellowship/training program into the different federal environments that have a need for informatics and are trying to grow their own staff.
>> Karen Bell:
Excellent suggestion. I think we can incorporate both what the Secretary can look for within the Federal Government and then also efforts that can be taken with the private sector through the accreditation body.
>> Jonathan Perlin:
Terrific.
>> Karen Bell:
Okay, great. Thank you so much.
>> Jonathan Perlin:
Okay, I think those are good comments. Items F and G go together. And so we had good discussion about G and moving that to a different portion of the ultimate letter. Item F I think all of us understand the cultural literacy to read between the lines, but to fund a collaborative group to study and develop an approach to champion informatics workforce needs is somewhat vague. Do we want to provide any greater specificity there, and if so, what are your recommendations?
>> Bonnie Anton:
Hi, this is Bonnie Anton. I agree that I think it needs a little bit more specific language here. And what I’d like to see is the Director of Nursing, VP of Nursing, whatever you’re calling your institution to be involved in this particular approach here, and also the CEOs to really be actively involved, not just at the beginning to say how important it is and let’s do it, but to remain actively involved.
>> Jonathan Perlin:
Okay. So I’m hearing two components. One is what goes on locally, but also representation categorically of certain job titles in a collaborative group of study.
>> Bonnie Anton:
Correct.
>> Karen Bell:
Do you think we could do that by combining F and G together? I think you were heading in that direction a little bit earlier, Jonathan.
>> Jonathan Perlin:
That had been my recommendation. Yeah. That would give it a greater specificity with a clear deliverable. Others?
>> Linda Fischetti:
And does A loop back into the F and G center of gravity? Or is that distinct enough that it should remain separate?
>> Karen Bell:
I think what we had decided, Linda, was that F and G would be the research component, and that that would become the new A, and that the additional funding for expansion of career pathways that said all of the pieces that go with it would then be the result of the information gleaned from the research.
>> Linda Fischetti:
Perfect, perfect. Okay, thank you, Karen.
>> Karen Bell:
Thank you.
>> Robert Juhasz:
Jonathan, I just want to clarify the collaborative group, and are we the way that I was looking at this was really more at a macro level. But maybe I’m misunderstanding.
>> Jonathan Perlin:
I was looking at a macro level as well.
>> Robert Juhasz:
So are we talking about really
>> Jonathan Perlin:
National I mean my understanding of what was implied is and is as the Secretary would for the recommendation, some of it to the effect the Secretary would convene a collaborative, interdisciplinary group to give that little greater specificity in terms of quantification with what we ultimately modify from J.
I think that also segues in terms of a very practical interdepartment request, which is Recommendation H, because it’s hard to measure if you don’t have a measurement category. And the occupational classifications for health IT professionals, as we’ve understood from the testimony, simply don’t exist to allow resolution into different components of the workforce. But H would sequentially have to follow J, because until we had recommendations or clarity on which occupational classifications should be developed, I don’t think we could turn to labor and ask them to implement. Any...?
>> Robert Juhasz:
Well, I’m thinking of, you know, vendors, you know, health information technology, you know, users and, you know, I mean, really looking at who are the stakeholders in the collaborative group.
>> Jonathan Perlin:
Okay, so let’s throw in the word “stakeholder” “group of stakeholders” and a parenthetic that includes vendors, users, purchasers, payers, providers?
>> Peter Elkin:
Also this is Peter Elkin health informatics researchers, too, please. The other thing I was going to mention I don’t know I just came on the call because I’m on the hospital service and our rounds lasted long, but are you all aware of the two what I would consider two of the major efforts to define the core competencies for the field? There’s been an effort by AMIA through the education committee to define core competencies for in informatics. And then there’s also an ongoing effort now in IMIA, the International Medical Informatics Association, to define core competencies for the field. They were done differently. The one that was done by the American Medical Informatics Association was one that looked at it from the top down. So it looked at classifying the knowledge, skills, and judgments that were needed with inside the informatics and, where those teaching areas overlapped different disciplines, how that should be constructed at different levels of education. So they had an upper graduate, they had graduate, they had PhD, they had researcher levels that because different competencies and different activities were needed for achieving these different levels of educational accomplishment in terms of their functional job requirements. And then the International Medical Informatics Association’s work has been more on a bottom-up approach to look at the literature and to define the field by what it does. And so there’s been a joint effort led by myself and Graham Wright from the U.K. to try to look at what the concepts are that are used with inside of health informatics, both bioinformatics and clinical informatics, to understand what the kinds of things that people have published about represent. And that in itself is a bottom-up way to sort of meet the AMIA work and define some of the core competencies. This, we hope, will inform curricula that will lead to educational programs to help improve the ability to train the throughput in terms of training new people to enter the health informatics workforce.
I hope that was relevant.
>> Jonathan Perlin:
Great comments. Great comments. Let me break it into two pieces. One, in terms of the Collaborative Workgroup, let’s definitely include, in that parenthetical list of representation, health informatics researchers and clinical and other informaticians. The second really hearkens to Recommendation D, which we might want to be a little more specific in terms of what you just offered, which alludes directly to (inaudible) current health care worker, encourage participation in educational and certification programs, such as the AMIA 10x10 progression and certificate programs, European computer driver’s license equivalent, and others. So I think if we augment that with the greater specificity that you offered, I think we captured that. The other piece I want to list from your comments is that I think another parenthetical is that which may serve as the basis for national or academic program curricula. Is that fair?
>> Peter Elkin:
Yes. And I’m hoping that, somehow, this can link to the specific workforce requirements to inform educators in informatics what they should be training in terms of what portions of their graduates should be targeted at what needs and what sectors within the industry, which would, I think, help inform academia about recruitment processes at the educational level. And once we get that right, the pipeline will start to move in the direction that I think we all want it to.
>> Jonathan Perlin:
Terrific. So let’s put that as a parent as a piece of after following we’ve merged F and G, and I think your point to inform the obvious outcome for use of this is not only federal policy, policy at other levels of government, but also to inform educators for the purposes of recruitment and getting programs in pipeline.
>> Carolyn Clancy:
Yeah, this is Carolyn. The one footnote I would put on all this is, inherently, we’re talking about closely related but separate streams here. One is, how do we build up a sort of workforce in informatics that’s very knowledgeable and is actually going to take us to the next level in many different ways? And a number of people have made suggestions around specific areas like clinical effectiveness and so forth. And that’s very important, but I don’t think we want to lose sight of and I think we’re getting there, but it’s going to be part of a core competence for all people involved in health care.
>> Peter Elkin:
Oh, yeah. Carolyn, if I seemed to intimate that that was not my belief, I was misspoken. My I clearly think that all people involved in health and health care need to have a certain level of informatics competence, and I think that that’s something that we should push. We clearly don’t have the educators to provide the education for that volume of individuals at the present time, and I guess I was seeing the health informatics professionals as one source of people who could provide that educational resource to our larger health care community.
>> Carolyn Clancy:
I think that’s part of it, although, I mean, to be honest, I think the folks who would be really good at this are not necessarily going to be in academic career paths.
>> Jonathan Perlin:
Agreed.
>> Carolyn Clancy:
Okay. That was my only point. So I’m hoping that in merging F and G, we can make sure that we don’t lose that sense. It’s interesting: You can see a lot of studies that correlate, say, CPOE and quality, or CPOE and safety, stuff like that. But it’s very unusual to find a study that actually goes down deeper to say how many docs are using CPOE and how in God’s name did they learn how to do it. (Laugh) And the doctors’ stuff, you know, interconnected the nurses’ stuff and so forth.
>> Jonathan Perlin:
I think those are terrific points. Carolyn, do you want to offer any specificity in terms of other outside of academe, in terms of providing education?
>> Carolyn Clancy:
Well, I think it would be AMIA, but it would also be meeting at institutions that are doing this now. I don’t you know, if the community and the Secretary could at least amplify what’s known so there’s some sense of where to find out. Right? And in a funny way, Wal-Mart may be at the leading edge, not so much for the health care, but because they train everyone who works there in how to use a computer. That’s how you get paid. So everyone knows how to learn how to use a computer and so forth. I don’t know who exactly those institutions are, but my guess is that AMIA and others could probably help identify some of them.
>> Jonathan Perlin:
That, I think, is a terrific suggestion. And I think it actually stands on its own, or at least as a very clear sub-bullet is that AHIC and recommendations Secretary should take note of other industry and industry leaders who have in fact effectively engaged education of workforce. I mean and I pick Wal-Mart as a perfect example. I’ve heard Wal-Mart’s senior leadership describe themselves as a logistics and information company that happens to move product. So that happens to be, I think, just a terrific example.
>> Karen Bell:
Peter, this is Karen Bell. I was just wondering if the work that you were doing internationally and I know about the work that AMIA is doing that you’re engaged in whether that’s been completed or whether I was under the belief, and I could be wrong, that that still is ongoing and there’s a lot more work left to be done.
>> Peter Elkin:
Yes, you’re correct. It is ongoing, and there is work to be done. There are some preliminary results in clinical informatics that at the international level that Graham has presented at the last general assembly meeting. And at that time, it was very clear that there was additional work that needed to be done, and an international consortium was put together to do the work, and the work has begun done, but the results are not available. I would expect that there will be results by at least preliminary results that would be usable by May of 2008 to be presented at the general assembly meeting in association with Medical Informatics Europe and Sweden.
>> Karen Bell:
Is there any need to provide more support for the work that AMIA’s doing right now in this arena?
>> Peter Elkin:
Well, I can’t speak for AMIA, but I think Don it’s safe to say that Don would not be adverse to additional support.
>> Karen Bell:
All right. I just was wondering in terms of the progression and how much work you felt still needed to be done, because it is a very important area sort of the core of the core from which everything else begins to grow.
>> Peter Elkin:
I mean, I think there’s going to be a point in the after the second quarter of at about the end of the second quarter of 2008 where we will have results from the international study and where we can start to merge that with the results of the original AMIA work, which underwent a first round of Delphi evaluation throughout AMIA at about 2003, 2004. But after that point, it probably would be good to revisit the Delphi method and get a broader consensus from the field around the core competencies, making sure that we haven’t missed anything, and then try to get something validated that can be used, one, to accredit take the programs and then, eventually, to accredit the practitioners at different levels, which I think, just like we’re accrediting electronic health records, is a very important part of where we need to be moving as a discipline. And although I agree with Carolyn that we’re going to train people who are not academic to do this work, I honestly believe that the core group of people who understand about informatics in a detailed enough way to teach it are, for the most part, in academics. There are some people in industry as well, but rarely have they had the time to provide the same kind of teaching program that the academic institutions have been able to nurture. Maybe another model is envisioned.
>> Karen Bell:
Do you think there ought to be some sort of a recommendation around that, Peter?
>> Peter Elkin:
Well, I would certainly like to see AHIC endorse a need for the development of core competencies in health informatics and further the idea that was articulated not by me, but by someone else, that and maybe it was Carolyn or you, Karen, who said that all health care practitioners should have some base core competency in health informatics. And I think those two notions are the bedrock of what we’ll be doing from this point of view in our field.
>> Karen Bell:
Okay. And I think we have captured some of that in the previous recommendations, so maybe we’re okay with that.
>> Carolyn Clancy:
So just as a minor wordsmithing thing, I think it’s “health informatics and its application” or something along those lines. The only thing I’m thinking about is, I could imagine some members of the community saying, “Give me a break. The kids all know how to use computers, so what in God’s name are we supposed to be doing for them?” And on one level, they’re completely right. What they don’t necessarily know is how it applies to the health care enterprise or public health or whatever, and that’s actually what they need to get.
>> Peter Elkin:
Right, and also, you know, the idea of what decision support can do for them and what it can’t do for them and what’s expected for them in relationship to those sorts of systems will be, I think, a striking patient safety issue that will be near and dear to all of our hearts as we move this program forward.
>> Carolyn Clancy:
Oh, yeah.
>> Karen Bell:
Uh-huh. Mm-hmm. Absolutely.
>> Blackford Middleton:
Hi. That last point and it may be in preface to this is that it implies very directly that even among the computer literate among the computer illiterate, but as well among the computer literate, an understanding of the benefits of health information technologies does not yet exist. And it’s hard to create a market or pipeline if there’s not an appreciation of what it is that you want to create a pipeline for.
>> Robert Juhasz:
This is Bob Juhasz again. You know, it’s the way and I’m following up on Carolyn’s comments and, you know, as we think about it, we have all these large organizations that have rolled out EMR and IT and really did that without having an educational rigor to the process necessarily. I mean a lot of it was “Okay, we have to get this done by such and such a date, and we have the workforce, and we’re going to go ahead and do that.” So I wonder if part of that you know, this piece that we actually apply research to is actually sort of looking at that and applying some method to it and actually sort of documenting that and how that relates to importance in health information technology education, because it seems to me that it certainly, you know, was our experience here that, you know, it just happened. You know, I mean, there was some methodology, but I’m not sure that there wasn’t necessarily an educational academic methodology to it.
>> Bonnie Anton:
This is Bonnie. I have to agree. When we went live too, we just did it and dealt with the problems as they came along. But since then, we have developed some needs assessments and looking at what kind of computer needs or learning needs our employees need and worked on that in terms of our educational programs. So at one point, yes, when we went live, we just did it, but now we’re kind of pulling back and saying, “Okay, now let’s look and see what do these people need to know. Do they actually know how to use a computer?”, because we do all our documentation online. So sometimes a needs assessment is very important, and we’re finding that out.
>> Jonathan Perlin:
Okay. I think we’ve captured the sense of this, and I’m recognizing we have a full agenda and we’ll have another round to go through. Let’s go to the last recommendation that is offered in this document, and it’s a very specific one: to encourage work with Governors to increase recognition of health IT workforce needs. And obviously, National Governors Association would be one venue. Before that, any other thoughts or augmentations/clarifications that the Group might recommend?
>> Linda Fischetti:
This is Linda. I wonder if we could also look at another method by which we could get this recognized as a professional, again dealing with this from strictly a federal point of view, so let me know if it’s not relevant here. But up to this point of time, we’ve looked at Office of Personnel and Management in terms of what they could bring to support AHIC’s agenda based on their purchasing power for the insurance that they buy for the federal workforce. But also, Office of Personnel Management does include, in fact, a number of career series. There is, for an example, a 5784 riverboat operator or a 7405 bartender or a 3808 boilermaker, but there’s nothing for the health informaticist. There is a 2210 series for an information technology management specialist. Within this series, that information technology management specialist is actually compensated at a higher rate than the regular GS scale, which the rest of, for example, a general schedule grade-12 employee is compensated on. If you’re a 2210, you’re compensated higher. So within my environment, I actually have people who are giving up their clinical career track to move over into the IT management specialist track so that they can be compensated at a higher rate.
So my proposal is this, for the consideration of this Group: Do we also want to include in here that OPM look at the feasibility of creating a specific job series for the health informaticist?
>> Karen Bell:
And you would include that under H primarily, Linda, in addition to working with the Bureau of Labor? We would also work with OPM as a way of moving this agenda forward as you described.
>> Jonathan Perlin:
Can I recommend we make these discrete recommendations, you know, whatever, point 1 or point 2? Because they’re different departments slightly different implications.
>> Karen Bell:
Okay.
>> Jonathan Perlin:
My own reaction is that it’s absolutely a brilliant suggestion. Didn’t know we had a federal bartender series. (Laugh) Maybe in my next life.
>> Linda Fischetti:
(Laugh) In case this whole informatics thing doesn’t work out, we have another place to go, right?
>> Lillee Gelinas:
Jon, this is Lillee, and I’m on, and I listened to Linda, and I’m always amazed at how crisp she is at getting to the point. I think that’s a great recommendation.
>> Jonathan Perlin:
Absolutely terrific. I actually, Linda, did think you were going to go to a slightly different area, which is that is that - not only encourage Governors to increase recognition but encourage other state and federal agencies with health interest.
>> Linda Fischetti:
Ah, very good, very good.
>> Jonathan Perlin:
Obviously referring to the idea of D, but also, you know, there are health programs at other levels of government.
>> Karen Bell:
Building on your earlier recommendation, Linda, we could basically take some of the components that you talked about with respect to fellowships and mentoring programs and include that language in terms of encouraging other state and federal health agencies.
>> Linda Fischetti:
Great idea, Karen.
>> Karen Bell:
That was yours. Thank you, Linda.
>> Linda Fischetti:
Sure.
>> Robert Juhasz:
I’m this is Bob Juhasz. I’m just asking this question, I guess, out of ignorance, but the HISPC’s grantees it seems like that is also a place where this is also coming from I guess, from the states back up to the federal level. Is that also something that we could include as a vehicle for encouraging the Governors?
>> Karen Bell:
Well, we have we can certainly ask them there. We have a contract with the National Governors Association through the State Alliance that deals with all sorts of HIT issues at the state level, and we can also there’s some effort that they could put into that as part of their contract. So we may even specify the State Alliance contract here and the NGA. Okay?
>> Robert Juhasz:
Mm-hmm.
>> Karen Bell:
Thank you.
>> Jonathan Perlin:
Outside of these recommendations, does anyone feel that there are categorically any areas that have been omitted or things that presented to us by previous testimony that we’ve not captured?
>> Karen Bell:
Or anything else that you think we should include that has come more recently since the testimony?
>> Jonathan Perlin:
(Pause) Okay. With that, I think, why don’t we assume consensus on what we’ve together crafted? Thank you in advance, ONC staff, for the work in putting this together and the first draft of a letter from this. And we will revisit that probably both in this process, but we’ll look forward to an email that gets that draft out.
>> Karen Bell:
Absolutely. And again, what will be coming to you will be certainly predecisional, very draft. Your input will be and editing will be looked for and appreciated. So thanks again.
>> Jonathan Perlin:
Okay. Well, many thanks to all for a very participative discussion. Greatly appreciated the input and terrific additional suggestions.
Before we go on to the use case, let me turn to Lillee. Before you joined us and I need to anyone who tuned in late, she was suffering the travails of air transit, but glad you’re able to join us. Lillee, as we acknowledged, just did absolutely heroic work in the with the entire Group and beyond in bringing together the e-prescribing. Any introductory comments that you would want to make before we move on to our next topic?
>> Lillee Gelinas:
Oh, thank you, Jon. And yes, I know many on the phone understand the wars of the air. I’m in the northeast and Hartford, and it was one of the roughest landings. I don’t know what’s going on with the weather up here.
But I would tell you that the e-prescribing work of the Workgroup over the last, oh, 2, 2 ½ weeks since our AHIC meeting in Chicago was not a rough landing. I have to tell you, I have never been more energized, excited about the amount of energy that went into conversations. And I think the #1 outcome of all the conversations related to the discussion was that we were able to give the Secretary a very crisp list of what the real issues are related to e-prescribing; for instance, the whole issue that we have with the DEA regulation that precludes e-prescribing for controlled substances, you know, the fact that a certain percentage of pharmacies don’t have the ability to accept prescriptions electronically, those types of things. So in addition to the recommendation and the consensus of the AHIC was accepted; the Workgroup recommendation was accepted that the Secretary should seek authority from Congress to mandate e-prescribing you know, in addition to that was a very crisp list of the issues.
So I’m hoping as this Workgroup moves forward, that we keep that in mind: that it’s not just the recommendations that we forward to him and to AHIC. It’s also being able to tease out what the true issues are, very similar to what Linda just related, because only we can only fix what we know. We can only fix what we know. And I just really want to commend the entire Workgroup, John, you, Karen, the Office of the National Coordinator it was a tremendous 2 weeks, and I do hope that AHIC and the Secretary were pleased with the outcome. I’ll end with that.
>> Jonathan Perlin:
Thank you very much for that. And I think there’s a great deal of enthusiasm, and I think that enthusiasm charged our first discussion, and I hope it will carry us the rest of the long afternoon of discussions. And we will segue to Karen Bell. Dr. Bell will lead us through this next bunch of work, which really revisits some of the areas you’ve mentioned that we’ve evaluated as use cases, and some new areas as well, in terms of agenda setting and refreshing. So with that, let me turn to Dr. Bell. I just note there is a handout refreshing AHIC Workgroup 2009: Needs and Priorities. And we’ll work from that and what’s on the screen.
>> Karen Bell:
Thank you very much, Jonathan. And also joining me today is John Loonsk, who, as I think everyone knows, is the overseer of the use case process, HITSP standards and NHIN, etc., all this as his work with the Office of Interoperability and Standards. But I’m going to just begin by setting the tone, and then John will chime in as well. And before I do, I also want to acknowledge that we have on this Workgroup people who are very engaged in not only the use case process, but also on certification commission, on HITSP workgroups, and so really are looking forward to your input and discussion today.
We’re going to start off on my first slide, which is basically just our agenda, which is to review the current process for identification of AHIC priorities and use case development; some of what’s happened with the first three rounds of use case development; some of the prioritization criteria as we move into Round 4: Priority Setting; we’ll be talking about the timeline for the activities with respect to use case development; and also, we’ll review some previously designated priorities for 2009 use cases as well, something we had talked about before.
But the bottom line is that in the course of the next hour, we hope to be able to discuss and gain consensus on your recommendations for Round 4 priorities. If we do not get to consensus today, then everyone’s going to have homework over the next couple of days, because the deadline for getting this done is what, John? Friday?
>> :
Friday?
>> John Loonsk:
A little bit of an extension, but...
>> Karen Bell:
Okay.
>> John Loonsk:
But we really do need these in December.
>> Karen Bell:
Right, so we certainly need them very, very, very, very quickly.
So with that, I’ll move to the first cartoon slide, which is a fairly robust one that overviews how the priorities are developed from the AHIC recommendations. A number of them have been made by many workgroups. I think we have seven workgroups that are making priority recommendations. And these are then clustered in such a way across workgroups so that there is coordination and synergy among them. And the ultimate product, then, becomes a use case a series of use case options that are presented to the AHIC. The AHIC then determines which are the highest priorities. And the use case development process continues on and then moves into the HITSP harmonization standard process and down to the certification commission criteria. I think it’s a process that all of you’ve seen before and are familiar with. If there are any questions, we can certainly go back to them later when John is as John goes into more specificity.
You will recall that on Round 1, on the next slide, there were specific charges to three workgroups. And these use cases led not only to HITSP standards but will be leading to incorporation of those standards in the 2008 certification process. From the provider perspective, the use case was around access to historical lab data as a result of recommendations from this Workgroup. From the individuals’ perspective, the electronic clipboard, medical history, and registration data was the second use case. And the third was a third-party use case, and that was biosurveillance and development of the minimum dataset and assurance that that use case then went through the process.
Round 2, as some of you may recall, particularly from the provider perspective, was medical management and emergency responder EHR. This is also important to the Public Health Workgroup as well. The individual perspective was consumer access to their clinical information, and the third-party perspective was quality reporting. Those use cases have been developed. They’re well on their way. They’ve been through HITSP. HITSP has reviewed them, has done their work on them, and has the public comment has been completed, but the actual HITSP standards have not yet been presented.
>> John Loonsk:
They are they’ve gone through a round of public comment, but interoperability specifications will be available first quarter of 2008.
>> Karen Bell:
Thank you. And I realized that I had a typo on this slide. It’s “medication management.” Thank you, Alicia.
Round 3 are the use cases that are currently in development. The prototypes were out for public comment about a month or so ago. We’ve gotten a lot of good robust public comment, got some very good subject matter expertise in on these, and they are now in the development process. Again, there are three different perspectives, but we have here certainly more than three use cases. We have, from the provider perspective, consultation i.e., referral for consultation and transfer of care. From the patient perspective or the individual perspective, we have secure messaging with providers and a second use case for remote device monitoring. From the third-party perspective, there’s public health case reporting, immunizations and response management, and research supported by genomic and family history information. So there are three use cases there.
So we’re now at the point where we need to begin the prioritization process for Round 4. What I can share with you is that a number of workgroup Co-chairs met, oh, several months ago now, in thinking about this particular process and came up with a set of criteria that you may find helpful as we begin to move forward. You will note that on the very last slide of these series and you don’t have to go ahead to this right now; we’ll get to that shortly the AHIC priorities are all outlined for you as well as a list of priorities must be about 200 or so in there; it’s quite a few for 2009 and beyond. Those priorities are really quite diverse. Some of them are very specific i.e., a glucose result from the remote monitoring device and some are very, very broad, looking at very diverse issues.
So rather than unnecessarily think the think through that particular set of standards or that particular set of issues, the Co-chairs basically felt that given all the experience we’ve now had with use cases, everything we’ve learned to date, what would be the particular problem that still needs to be resolved? And that is a particular well-defined problem. It may not be specific to any one workgroup or perspective. It may be broader than that. But if we could describe those particular problems and think of them from various perspectives, assuring that it will support adoption of interoperable HIT, because that’s the charge of this Workgroup by assuring that there will be value that will accrue to providers and to individuals; to realize near-term benefits, recognizing that there are a lot of nontechnical limitations to some of the problems we need to resolve, is there something we can do near-term, and if it does require nontechnical limitation, what can we do to address that? And also, whatever that particular problem is, if it requires incorporating multiple processes, coordinating over multiple settings, we assure that that happens. Just as an example, when the Workgroup decided to move forward with obtaining historical lab results, we had a very good, very strong discussion on that, very good use case development, but we’ve certainly heard from the both lab and provider community that it would have been more helpful if there had been additional pieces to it; i.e. the lab ordering piece. They want to make sure if there are other processes that need to be engaged that we include them all.
And then lastly, we certainly want to leverage existing efforts in support of interoperable health IT. If we already have a use case that takes us a certain amount and we need to go a little bit further, we could leverage that. If we know there are already that there’s already been a significant amount of work done on standards and it may be not an extraordinarily difficult task to move forward in a particular area that may be something we might want to leverage as well.
So we’ll probably come back to this slide later, but I’d like to very quickly move forward to the next two and then move on to John.
Again, this is a very, very busy slide. But it does underline the fact that we had hoped that we would have workgroups submitting priorities for the next round by the end of November, which is why we really do need to move quickly into the process now so we have them for December, because we do need to bring them to AHIC in January, and the AHIC meeting is Tuesday, January 15. So these the priorities need to be in John’s hands very, very shortly so he can get them to AHIC in that time frame.
And then the last slide I have here is the one that I referred to earlier. I think John has it in his set as well. So I’m going to defer any further comment on this one, which we internally call “the eye chart,” to my colleague. Thank you, John. Oh, does any before I do that, does anyone have any questions of anything that I’ve presented so far, or can we move right on to John?
>> Peter Elkin:
Sounds fine.
>> Karen Bell:
All right. And the reason that we’ve asked John to join us is because we’ve heard from a number of Workgroup members that it would be helpful to have a little bit more clarity about the results of the work that’s been done to date. You’ve just seen that we’ve had use cases that have gone to completion. We have use cases in progress. What’s really the deliverable that these use cases bring to the provider at the point of care? Is that again, keep bringing this back to what this Workgroup is about. We really want these use cases to be helpful in bringing information to the clinician at the time that he or she is being patient. So some of these have more semantic interoperability associated with them. Some have more messaging interoperability. So that’s really the question we have for you, John, and thank you for joining us.
>> John Loonsk:
Thank you, Karen, and thank you to the Group for having me here today. If you can move to the second the next slide. Some slides were sent out earlier today as well, so hopefully you all have them. I wanted to set the context for the fact that this overall process is essentially about a 2-year process, from the time which priorities are being identified and the use cases are developed, to the time when there really can be an expectation of implementation. And the Karen mentioned consideration for certification criteria that we know that there a goodly amount of the market of ambulatory EHRs are now being certified. Getting these standards into those systems is important priority for the certification commission. That has impact on Stark and anti-kickback. The premise of Stark and anti-kickback is that they are interoperable systems. The Executive Order points to the fact that federal systems need to implement these as well. And then we are actually prototyping these in the Nationwide Health Information Network trial implementations as well. So the first seven use cases, three of which have been fully taken through to standards interoperability specifications, as HITSP calls them and then the next four, which are still due for that final publishing of those standards in the first quarter of 2008, are all being prototyped in some of the health information exchange areas that have been by the trial implementations, and the announcements of what who’s doing what will come shortly. We’ve been in negotiation as to which of the different seven use cases are being prototyped in which of those health information exchanges. Essentially, each of them is going to do two.
What I’m particularly pointing to is that it’s a 2-year process, and the work that you’ve done before in this Working Group and elsewhere to try to identify these priorities is very helpful in getting to that end goal, but it does take a little while. One of those years is really where it’s just blatant, where the standard is available and people recognize that it’s coming so that it will be indeed, they can work it into the system that they’re developing. And so that year is basically needed by the industry to know what is coming, but it’s a time when that builds in and adds to the time frame of this process. And I just wanted you to be aware of that.
The other milestone that I should reference here is that in December, the Secretary will actually complete the first sort of the first round here by recognizing formally the first round of standards, so the first round of use cases for lab, for medication history, and registration information, and biosurveillance will be recognized in December. There has been one two standards there that have been delayed because of the standard development organization’s activities in that regard. One of them is this critical lab message standard. That message standard was just validated and approved, but we can expect that to be recognized in June of 2008. And once it’s recognized, that means that all federal systems will be expected in new systems and upgrades to implement those standards specifically. So it is a significant event. We haven’t quite gotten to the finish line, but that’s where it is, and December is going to be the first time that the Secretary has recognized these standards to kick off that context.
Next slide, please. So previously, as Karen mentioned, this Working Group and others working groups have then identified priorities, and what we’ve tried to do was get as fine-grained priorities as we can get, because the goal here is to try to, indeed, coordinate them across the different working groups and to see where there are synergies and to try to get as many in scope. Karen mentioned the lab order. You know, it was recognized that having a lab order that’s electronic is valuable, too, but it was there was too much to do in that first round that was not in scope for the activity. That’s an example of something that could be added as a priority, but the more fine-grained the priorities you can produce, the better the most helpful in terms of getting them in and getting them into the process. So it’s always one of the challenges here not to be coarse in what the area is, because it’s not as helpful in terms of moving forward.
Relative to some of the results here, I mentioned the lab standard. It is it’s now been fully validated. It covered the lab result message, which is going to be so critical to those getting lab results into EHRs electronically as well as getting sort of having access to historical lab results, the number of (inaudible) standards (inaudible) that will be helpful moving forward. It is includes SNOMED and LOINC, these terminologies for the result and the test name. It includes a standard for a unit of measure in (inaudible). It includes a what’s called the standard object identifier, which says that indeed which is important, because it allows for the connection between lab results across an entity, and that’s a critical thing moving forward and other things, other codes and terminologies from HL7 and elsewhere. It is, as I said, balloted and now will be done. It covers basically all the intent of the lab result is to cover all clinical results except anatomic pathology, and there’s an exception to blood banking there as well. And the way that it’s orchestrated is, it basically the coding of this result is oriented to the top 95 results that are recorded in HEDIS. And so SNOMED and LOINC codes need to be used for all of those. The next 5 percent can be done either with narrative or with SNOMED and/or with SNOMED and LOINC codes. But this is a broad message useful for a variety of different purposes. It clearly represents a great amount of work by a bunch of standards people to get to this point, and I think is really a seminal event in terms of standards and implementing in that it health IT systems. Lab results represent a critical portion of a variety of different activities. This one is oriented to supporting biosurveillance needs as well as electro clinical needs in the context of reporting into the electronic health records. So there’s a lot here, but there’s value in a number of different respects, and I think that’s an example of how the cross-connection between different working groups can be helpful to facilitate both activities.
Another important activity that should be pointed to is the summary record work and the standardization of a summary record. Summary record this has also been one of the more challenging areas in the standards community. There have been a couple of competing standards. HITSP has harmonized to what’s called the CCD, the Continuity of Care Document, which takes a lot of great work from the HL7 community and the critical document architecture and a lot of great work from the work in the continuity of care record from ASTM and others who have been working in that domain. It’s hard to totally say how important the summary record is going to be in electronic systems. I think many of us feel that this is really a critical enabler of electronic health information exchange. It’s obviously hard to make the full patient’s chart electronic in the near future. There are standards issues. There are legacy information issues and that’s in paper form. But this concept of having a broad-enough summary record that can relay information personal information, medications, problems, allergies and also allow for things like discharge diagnoses and chief complaints and some of the clinical documents that are associated and need to be exchanged at the same time is really, I think, also a significant event. And part of that was identified in the first use case round consumer empowerment as how to get data to a PHR. But a big part of it is being referenced in the emergency responder EHR use case, which came out of this Group as a need, not so much in the context of how to record emergency care, but in the context of how do you present a succinct and yet useful gathering of clinical information to someone who needs emergency care so that you can treat them appropriately given their status and condition. And that was is also a significant driver for the summary record, and this concept of fleshing it out further still is part of the emergency responder EHR use case.
I should say that the summary record is also something that plays prominently in the Nationwide Health Information Network trial implementations that are now kicked off as well. There’s an expectation that every health information exchange report a few very specific functions. Lookup of patient data is an example of a function. But an important one is being able to push data, to be able to route data, to be able to send it to the appropriate receiver. And one of the things that’s specifically cited here is to be able to route and send the summary record in this format a single standardized format to those that need it in the context of health information exchange. So the standards work is very important in that regard as well.
The slide that’s in front of you indicates some of the more some of the priorities that were advanced previously. It’s hard to believe that it was as long ago as September of 2006, but that’s when those priorities were advanced. The left column identifies those that have been put into a use case in one place or another. And that could be in the first seven, or it could be in the current six that are in this next round. The ones that have not been specifically identified as having a home in any of that first 13 are identified as unassigned in this regard as well. As Karen indicated, any priorities are helpful and useful, whether they specifically relate to this Working Group or others. But I just would encourage you to think as specifically as possible. That’s really how it’s easiest and most likely that the priorities can get embodied in the next round of that activities as well.
So with that, maybe I’ll stop, Karen, and if there are questions regarding the process or the content, I’d be happy to try to answer them.
>> Karen Bell:
That would be great. Thank you very much, John. Any comments or questions from anyone?
>> Carolyn Clancy:
Well, I have two questions. And actually, to be honest, one of them has come up and I think you’re aware of it in the Quality Workgroup: Is there a map anywhere? Can I click on the Web or at the ONC Web site or something and know what the status of this is for any of these, understanding it would be not straightforward to me how you would do this? But I mean, how would I know if I suddenly woke up one day, as I did this morning, thinking, “So where is that lab results recommendation?” How could I figure that out?
>> John Loonsk:
Well, so, it’s a great question. I think that if you can go back to my the first slide, the so-called eye chart is the most specific is the clearest identification of the priorities coming out of the Working Group. It doesn’t it’s not the recommendations for the Working Group, which really have had a different trajectory. And so this starts to categorize at a high level where that content is relative to which round and where it is. No doubt we can do better in terms of specifying its status and the different steps as this progresses. But that’s the highest level articulation of that. And that’s on the Web as well.
>> Carolyn Clancy:
Well, that’s helpful. I mean, one of the areas that I’m confused, a thing that I forgot about this recommendation or (inaudible): It’s much more about dangling items that aren’t related to standards, qualities, or anything like that, like CMS needed to get some regulatory clarification around CLIA, for example. I’m trying to figure out where all those pieces fit together (inaudible).
The other question I have was I keep thinking I should be more gifted and able to ask this more intelligently, but I don’t know a better way: How do you get information out of the CCD? You visit any health care place, right? And what they tell you is, “Oh, yeah, we have EHRs. It’s great.” All this stuff goes into a black hole, and then try to get anything out. We can’t get anything out. And I get a little confused personally about how the CCD is going to help us. So I’ll take your word that it’s a huge advance, but it would be on faith.
>> John Loonsk:
Sure. It’s an excellent question. And it the answer to it actually involves about five different things. I mean and so, I the prospect of that question is about what is you know, the standards people like to talk about semantic interoperability, which is the Holy Grail of the fact that one system can share information with another system, and there’s no ambiguity about how it was shared. To do that, you really need to have a code for every single thing. And we’re far from that, realistically. So much of what we do, even when we use electronic systems, is in text, and the content is embedded in narrative text. And using it and getting it out, as you say, at times includes looking at that narrative text and getting it out.
The summary record, as an example, is part of both. At the same time that it identifies some tabular data, numeric data, coded data that says, “This is exactly where you go to find the allergy,” and it’s in coded form, the second system you’re sending it from one system to another using that summary record can process that allergy. So it could do a manipulation. It can process it. It can work with it. If it’s a medication, it can do a drug-drug interaction check with it, because it’s coded. So summary record exchanging one system to another system if it’s in coded form, that medication, that second system can do a drug allergy text or a drug-drug interaction text. If, in that summary record, instead of it being in numeric form, it’s actually in part of a huge narrative note that says, “Oh, and by the way, the patient has allergies that include these, and they’re on these meds,” you can’t you have to read it when it gets to that next system. You can’t process it. You can’t compute it. And so one of the classic issues that relates to your good question is what form, how specifically coded, how structured is the content, and that varies depending upon what you’re talking about. So in the summary record, problems, medications, allergies these are coded entities. And the ability is to process them when you get to the other side.
>> Carolyn Clancy:
Problems to the extent that, as a clinician, I choose to record. (Inaudible.)
>> John Loonsk:
Can’t solve that
>> Carolyn Clancy:
No, no, but I’m
>> John Loonsk:
Right.
>> Peter Elkin:
Actually, John and Carolyn, one of the this you know how this was developed. It was developed originally as the transfer document for the state of Massachusetts when people left hospitals and originally went to nursing homes or other facilities. And it was used for about 50 years in Massachusetts and then got expanded through ASTM and then merged with the technology that developed through version 3 CDA at HL7.
And so there were a couple of other benefits that I think could be expressed. One is that you have brought together information that typically in medicine exists in multiple different systems, and you put it into one document. And the second thing is, by putting up an XML form that has all these data requirements where data has not been documented, it encourages the health care organizations to actually provide that information. And the thought is and this is untested that this conglomerative information will help whoever receives the patient to understand that patient more completely and provide, you know, better, safer care hopefully get better outcomes. So but I think that I think, again, that’s unproven, other than the experience in Massachusetts, where the form itself did bring together information that was helpful. So it would have all the patients’ medicines and their allergies and their problem list very well-documented and front and center on the front part of the form, so that whoever was receiving the patient had that context. And Tom Sullivan actually was the one who pushed this through originally from mass medical. Anyway
>> Carolyn Clancy:
Thanks.
>> :
That’s helpful.
>> John Loonsk:
Yeah. So also, there are other factors in terms of the classic how-do-you-get-it-out issue. I mean, EHRs are not hist are frequently oriented to clinical transactions clinical activities versus reporting per se. So is the software conducive to the use of the data? Are the data representative of population accumulation? Those are all additional questions, and there are more that frankly relate to this classic question of, “Well, the data got in there, but darned if I can get it out,” you know.
>> Carolyn Clancy:
(Laugh) I hear this everywhere. Thank you.
>> Karen Bell:
But I think what that does bring to mind again, from a clinical point of view, if you are using an EHR and you want to, for instance, be able to identify a population of patients that are on a particular medication or a population of patients that have an A1c over a particular number, all of those data elements need to have the semantic interoperability for you to move them around and aggregate them in certain ways. And I think that may be one of the issues that I think a lot of people may or may not understand in terms of the CCD, because there’s a hope that as this information comes in from multiple sources, it can then be added to the information you already have and then can lend itself to this type of better information or access to better information.
>> John Loonsk:
Right, and absolutely. And the CCD is a step down the road
>> Karen Bell:
Towards that.
>> John Loonsk:
towards that. But the classic problem here has been that highly structured data, the kind you can really process well and use for good reporting and good analysis, are the hardest to get into the system. I mean, who wants to pull down a menu five or six menus to get to a structured result versus dictate or have someone dictate it into a chart or type in narrative text? I mean, that’s the problem, in that the issue, like in so many other areas, is that the benefit of getting structured data does not necessarily accrue to the person who is has to put that structured data in and where it is definitely harder to put it in, in structured form, than it is to get it out.
Now there are all kinds of compromises you can make. You can map data. You can do other things in that regard. But that’s fundamentally a problem here, which is that most of the information that’s processed is not structured. It is more narrative than anything else, and though the summary records is a step in that direction, it’s not the full solution.
>> Blackford Middleton:
John, this is Blackford. I missed part of the conversation, but I just wanted to offer my strong endorsement for this kind of information exchange as an intermediate step toward the full semantic and syntactic interoperability we all dream of. I think that most of the EMRs, if given a standard payload message in the CCD format, could figure out how to add it to their own structured data repositories and whatnot underneath. And my hypothesis, which I think is untested still to this date, is that a complete med list decision support and a complete problem list decision support is better than an incomplete of either by way of decision support. Karen’s already talked about the outcomes reporting in population management, but also should be facilitated but, again untested. But I think this is a you know, a right step in the right direction.
>> John Loonsk:
Thank you. Absolutely agree, Blackford.
>> Karen Bell:
Well, I guess at this particular point, we still have 20 minutes left, and we have some work to do, folks. And that kind of (laugh) and that gets us down to what priorities this Workgroup might choose to move forward with. Perhaps we can move back a few slides to the original document that I had on the prioritization process, bearing in mind that there we certainly would like to have as much specificity as possible.
>> John Loonsk:
And just one parting shot. I’m sorry, Karen.
>> Karen Bell:
Yes?
>> John Loonsk:
But, you know, just to follow up on this conversation, to say to talk about a certain data type that you think would be of value to be structured versus in narrative form is the kind of priority that could be very, you know, helpful to get forward. So that’s an example of something that should be very constructive.
>> Karen Bell:
Good. Thank you, John.
>> Blackford Middleton:
So on that point, John, I would respond that if the CCD were to be sure and standardized around allergies, prescription medication administration or medications admini prescribed, problem list, DNR status, patient demographics, and key laboratory data, you know, that would be a subset of the universe of potential interoperable data which would be extremely helpful for decision support in all EMRs.
>> John Loonsk:
And I think we’re on that is within scope of the work that has been done for the emergency responder EHR as an extension to the summary record. So
>> Blackford Middleton:
Okay.
>> John Loonsk:
I you know, I don’t have I’m not sure about DNR status, but everything else you specified is in scope of what has been formulated as this broader summary record, tabular data, standardized data for those purposes, although talking about chief complaint, discharge, diagnoses, and mode of transport of CDA injected into the CCD. But so that’s my understand of the status of that.
>> Blackford Middleton:
Thank you.
>> Karen Bell:
Thanks again, John.
>> John Loonsk:
Thank you.
>> Karen Bell:
So going back to our issues again, bearing in mind, as we were saying when John was walking out the door the first time, we certainly want to get down to a level of specificity in terms of what data points need to be have semantic interoperability, what particular pieces of information are going to be important to have in the EHR, with respect to interoperability.
But I think, again, in order to do that, we really need to think about what problem we’re trying to solve here. Are we trying to assure that the whole laboratory cycle is working well so that we have the ability to order? That we have the ability to receive anatomical pathologies as well? Do we want to complete that to the nth degree? Or do we want to begin to solve other problems? As we were talking about referrals for consultation, for instance and I’m pleased that we have some cardiologists on the Workgroup right now it became quite clear that cardiologists in general rely very heavily on wave forms in terms of their consulting practices. And maybe you could speak to that a little bit, Dr. Hays.
>> Richard Hays:
It’s Rick. I’m not a physician.
>> Karen Bell:
Oh, sorry. (Laugh)
>> Richard Hays:
(Laugh) Yeah, with the American College of Cardiology. Yeah. I think the wave form in getting pictures is more a controlling, cost-control issue. And I think that’s high on our agenda, in that arena in general.
>> Karen Bell:
And I think the this issue of having the ability to actually obtain the EKGs having them in the record, having them in a way where they’re interoperable has come up. It’s also come up in terms of other disciplines that may want to clearly, they’ll want to see the results not the result of an MRI. They’ll want to see the MRI itself. You know, you talk about the neurosurgeon or the orthopod. Simply reading someone else’s interpretation does not work for them.
>> Richard Hays:
That information’s not adequate.
>> Karen Bell:
That’s right. So we would need to find ways to move that forward. And there are a set of standards already in that particular area that I’m sure everyone is aware about. So it could be another approach.
So I just throw those out as a couple of ways to begin to start thinking about what problems you think you might want to be resolved, and then we can start thinking about the level of interoperability of the data points necessary to move forward. So with that, I know I can sense that there are people waiting to jump into this conversation, so whoever wants the mic should take it. (Pause) Don’t be shy. (Pause) Or I’ll start calling on you.
>> Carolyn Clancy:
Well, let me ask a strategic question. One way to think about this is, what brings us closer to a killer app that would bring us over the line? You know, “If I could have this, that’s it. I would figure out how to get to an electronic health record.” Another way, of course, is to think about those who are using some functionalities now. Maybe they’re using epic light, which I hear about a lot, or something along those lines. What gets them to use more of the functionalities? And I’m not sure where we want to focus our efforts. And I don’t mean that that’s a question for you, Karen. I just mean it might be a way to think about what problems are we trying to solve.
>> Karen Bell:
Well, maybe we could start off by even thinking about some of the issues that are happening right now. We were talking about e-prescribing a little earlier, Carolyn, and we recognize that the full set of standards for e-prescribing have not yet been developed. There are no interoperability standards, for instance, for the SIG, the part that tells you how much to take when of any medication. So in the narrative there are a multiple other issues around e-prescribing that we may think are important, because clearly, if that comes anywhere near to being a mandate, we will need to make sure we have good standards in place for that.
So that’s one opportunity. I already mentioned the possibility of adding ordering of labs to the current laboratory one. We mentioned the wave form one. All of these are important to at least some segment of the physician population. And the question, I guess, that the Workgroup has to come up with is, “Are there others, and what others would be important for the practice of medicine?” And it doesn’t have to be internists. Or primary care physicians. It can certainly be any specialist that is engaged as well.
>> Robert Juhasz:
This is Bob Juhasz, and I think one thing that I think of that actually keeps people trying to live in two worlds is the issue of archival data and, you know, entry, whether it’s through scanning or, you know, trying to actually incorporate, you know, not just digital or paper data into the system. And just limit the reasons for anybody to have to go anywhere else, other than a unified record.
>> Blackford Middleton:
This is Blackford. I guess I agree with the last comment and to Karen’s comment to the extent that the physician can sit down and open or a clinician open up a record that is fairly well-populated at the get-go makes it makes the barrier to use just that much lower. While that’s true, I feel, however, even if we had a perfectly exchangeable CD clinical document exchange dataset that it doesn’t imply necessarily alone that alone will get us over the EHR adoption hump or, you know, the change the shape of the curve of the adoption because of all the financial issues.
>> Karen Bell:
So but suppose just fantasize a little bit with me. If we could get over the issue of financing and there was a much more reasonable business case than we’ve heard in the past here, would this be a top priority? This would be something that really would make a huge difference and would move more physicians to move in that direction? Because it’s as difficult as the financial picture is, there are a lot of other difficulties, too, involved in implementing an EHR. So this could this be a top priority, something we would really want to address for use case?
>> Peter Elkin:
Can I ask a question? This is Peter Elkin. Are we looking here for grand challenges, or are we looking for low-hanging fruit? And the reason I say that is because there clearly is a difference in what one would recommend as a strategy for going forward. If you were thinking about what the endgame is, what we really need to do to get to the point where all data in the country’s interoperable and everybody’s using an EHR, and then but there is some low-hanging fruit: as someone brought up medication error and medication knowledge bases before. And probably many are aware, in this audience, that the structured product label has passed ballot at HL7 for the FDA, and they’re starting now to mandate that all of the drug companies submit their product labels in a structured form. They’re still not clear how they’re going to get all the legacy data into it, but let’s say that gets solved. The FDA will provide a database that will have the indications, the contraindications, the usage, and hopefully drug-drug interactions for the 16,000 or so drugs that are licensed within the United States. That seems to me to be low-hanging fruit, if you’re looking for something that would be easy to get done and that’s already in motion that we could capitalize on.
For grand challenges, of course, I think we need to get a lot of that data out of the that’s locked in the textual material and make it available to people for practice education and research such that if a patient goes to hospital B and they’re usually cared for at hospital A that they can have their records shipped over, and that hospital A record will automatically work with the decision support software that exists at hospital B. And to me, that would be the great grand challenge if we’re thinking that broadly.
>> Karen Bell:
Excellent idea. Other comments? (Pause) Are there
>> Carolyn Clancy:
Strategically, I like low-hanging fruit. And I like a moving train, because there will be pushback even at the idea of having the conversation about a mandate, to put it mildly. And to be able to say this recommendation specifically addresses additional requirements here, I think, would be great. We’re already anticipating what’s going to be needed.
>> Karen Bell:
Yeah. No, I think that’s very clear. So one of the things that we have here is a medication management use case that is part of 2007’s medication reconciliation some ambulatory prescriptions and contraindications. We could build on that to include some of these other pieces that, Peter, you just indicated so that we can make this medication management much more than the initial use case as it was written.
>> Peter Elkin:
I think that would be excellent. I mean, one of the things that’s been a barrier that we can actually help break down, which I think would really move this forward quickly, is that the knowl many of the knowledge vendors in drug in the drug-drug interaction environment I won’t name names, but anyway, they’ve been recalcitrant to interoperable data sharing, because they see that as part of their core business. But it’s such an integral part of what needs to happen for patient safety. All of you in the audience are familiar with the many of you have published the core literature on medication error and how much that plays a part in the overall error that we have within health care today. So it’s I think that, you know, just carving that out and drawing attention to it and making it understood that this is an important public good that needs to happen is something that this austere body could really contribute very, very importantly to the ongoing work.
>> Karen Bell:
Alicia, who’s usually two steps ahead of me over here, just made a comment, and that’s that if we’re really going to get into this type of medication management, we really are going to be relying probably quite heavily also, not only on the information that’s coming on the drugs themselves, but also on how they interact with patients. So the ability to have information, lab information, diagnosis information, so you cannot only do drug-drug interaction and drug-lab interaction, but I’m sorry and drug-allergy interaction, but you can also do drug-lab and drug-diagnosis would be very important moving forward. And that’s something that’s not yet been tackled. Would everyone agree that that could be included in this more robust use case?
>> Peter Elkin:
I would like to see that, personally. This is Peter Elkin again. Yeah, I think that would be very important. One of the fears is that the decision support won’t be robust enough to reflect actual care practice. And if it isn’t enough to affect to reflect care practice, then the messages that we give to practitioners will be, in much higher percentage, false positive alerts. And I think that could actually set us back. So the more we include the core data elements and data types that we think will be necessary to make sure that the messages are as relevant as possible will lead to a greater chance of success in the overall effort.
>> Karen Bell:
Thank you, Peter. Any more discussion on medication management and how we might grow that use case?
>> Michael Kappel:
This is Mike Kappel. I guess I’m just a little I won’t say “concerned,” but if you do look at the products that actually exist out in the marketplace and, you know, most of the EHR and e-prescribing products used, one of the recalcitrant products previously mentioned, both for formulary and as a support for decision support I guess I don’t necessarily see the direction this is going relative to products that already exist that have bundled these capabilities in terms of expressing formulary and capturing the kind of interactions that you’ve talked about drug-drug, drug-allergy, drug-lab interactions as part of their functionality. You know, is there a compelling case for this Workgroup to say that development of this use case or these specifications are required and, in essence, obviate licensing existing commercial product?
>> Karen Bell:
Any comments to that? (Pause) Mike, one of the questions I guess I would have for you, at this point, is that, given that this would be a 2009 use case
>> Michael Kappel:
Yeah.
>> Karen Bell:
and given the direction that we are moving in, in terms of possible mandating of e-prescribing in the future, we’re going to need a lot more, I think, support around clinical decision approaches, because I we anticipate that certification commission will be certifying e-prescribing products, whether they’re stand-alone or whether they’re embedded in EHR, and that those e-prescribing products will have three components to them. One is interoperability. A second one is extensibility so that if it is not part of an EHR, it has to be able to extend and grow into one. And then thirdly, it has to have the kind of knowledge management or clinical decision support that will truly improve patient safety.
So this may be the beginning of moving in that general direction in that process. But again, I would point out that if it’s a 2009 use case, then we are not talking we’re not talking about actually including it in the certification process until well into 2010 or 2011. And so I think that would give the vendor community a lot of time to adjust to the changes.
>> Michael Kappel:
Yeah, I guess my only point was, there you know, there are a lot of products out there that kind of do what we were suggesting that this use case do, unless I misunderstood. And as far as the certification, you know, today in the certification in fact, the first year of certification, you had to demonstrate drug-drug decision support as part of the certification of ambulatory EHRs. So, you know, some of that’s already out there. I guess the question is, “How is this different from what already exists? And why is this an advantage that would you know, that would be compelling enough to drive adoption of EHRs, since that’s our stated goal?”
>> Karen Bell:
I think the issue here is interoperability so that all of these all of this information can be shared and the systems can be interoperable with respect to it. And I think that’s really the difference here.
>> Michael Kappel:
So you mean from a the ability from a standard-alone e-prescribing application with a larger system?
>> Karen Bell:
Mm-hmm.
>> Peter Elkin:
So from hospital A to hospital B, let’s say.
>> Karen Bell:
Yeah, exactly.
>> Peter Elkin:
People have one drug knowledge vendor that uses their own proprietary terminology in hospital A and a different drug knowledge vendor that uses their own proprietary terminology in hospital B. They really can’t share decision support information very well.
>> Michael Kappel:
Right, although most of decision support is called upon at the time of prescription, which means it’s embedded in a separate application. But I do think the ability to the concept of linking across multiple enterprises, in terms of having a consistent ability to link categories of medications, for example, would be useful. Does that make any sense?
>> Peter Elkin:
Yeah. I mean, you know, so what you’re envisioning, of course, is that what happens now, which is people get a paper record, they retype in all of the meds where they take a history, and they type in had all the meds they were given, and then they run their own decision support software against it this might enable things to move a little quicker, where electronically transferred summaries or other kinds of electronic health records would automatically be able to run on the decision support software at the receiving hospital or health care organization.
>> Michael Kappel:
Yeah, I guess I could see that. Today, for example, you know, assuming your SureScripts compatible, you can receive a list of medications even if, you know, you went through an e-prescribing application even if you, as a prescriber, did not actually prescribe that drug. So you could have a comprehensive list of meds that were filled through a SureScripts pharmacy. And you’d have that capability today.
>> Peter Elkin:
Yeah, but see, the many of the drug knowledge vendors don’t work off of that standard still, and so they need it transferred into their own terminology or mapped to their own terminology in order to run their software. And I’m thinking of the group that’s the has the larger share of the market when I say that.
>> Michael Kappel:
Yeah.
>> Blackford Middleton:
I guess the this is Blackford. I guess, Karen or John or I’m a little confused, and maybe a little bit lost (laugh). You know, I wonder if we look at the nation as potentially going through the eRX adoption life cycle through whatever mechanism, mandate or otherwise, maybe Carolyn said this early on, “What is the next best thing we can do to keep physicians adopting higher levels of functionality and whatnot in their electronic health records?” My overarching concern is that we may get eRX in some form or another and then stop there for 10, 15, 20 years and never achieve the rest of the value of interoperability information exchange, decision support, or outcomes assessment that will come with more broad-based information portability.
>> Michael Kappel:
I think that’s the right question.
>> Jonathan Perlin:
Yeah. This is Jon Perlin. Let me just tag onto that question. That’s why I have some ambivalence, actually, about the critical summary. On the one hand, the summary would facilitate, as a priority for use case, a kind of set of key and critical information. On the other, there’s this sort of perverse incentive that might suggest to some that we’ve arrived. That would forgo, of course, all the decision support and all of the greater capabilities that the greater depth of information and standards-based interoperability would allow. So I think that’s a fundamentally critical question, Blackford.
>> Jonathan Perlin:
I guess the one question going forward, then, is, you know, “Do you actually look to have other areas of the record that are non-coded data in the future be coded? And is that an opportunity to work towards?”
>> Karen Bell:
Can you say that again?
>> Jonathan Perlin:
You know, I mean, there’s SNOMED conceptual opportunities to code data, whether it’s coming in, in text you know, do you actually work in the future to actually being able to code that so that you can pull the data out at some time in the future? I mean, we can’t do that now, but do we actually work towards a standard of some day being able to do that?
>> Michael Kappel:
As in an NLP approach to text extraction?
>> Jonathan Perlin:
I’m not familiar with NLP, but...
>> Peter Elkin:
This is Peter Elkin. Many of us have spent our lives working on that problem, and there are solutions that are useful and worthwhile and that can be used at the present time, but it’s not something that’s been standards based to this point forward. And perhaps that’s something that should happen over the next several years. But clearly, there’ve been fairly large government initiatives that have taken advantage of that kind of capability.
>> Blackford Middleton:
While we’re all scratching our heads, let me offer another vector to consider. You know, it’s clear that there’s a lot of interest in the PHR as a patient-controlled element of health care information management. And perhaps we need to frame the question in a slightly different way. It may not be as important, in a future world, for there to be EHR-to-EHR information exchange and interoperability so much as there is need for, or will be need for, PHR-to-EHR information exchange and interoperability that is strictly under patient control. I think we’re going to live perhaps in a hybrid world for a long time to come. But I wonder if that helps us think about the types of information that, at a minimum, need to be codified to a degree that allows the patient, as the hub of interoperability with his or her PHR, to determine where information is flowing or not.
>> Jonathan Perlin:
Well, I think that’s a well, you know, we I spoke about this one model that we saw that started with Cerner and now is called Care and Trust, I believe, in Kansas City. But it’s sort of that same model, where it’s a PHR that has a query from an EHR to say, “Can we have this information?” And then the person releases it. But, you know, to have that to be able to be standardized across the system would be significant.
>> Karen Bell:
This is Karen. I just wanted to add that the Consumer Empowerment Workgroup meets tomorrow, and I’ve talked to several members, and I fully suspect that’s a priority that they’re going to set forward. It’s a very important one for them.
>> Blackford Middleton:
And at the same time, I guess this is Blackford again you know, while I think about the PHR and patient empowerment, which I’m all for, I think we on the delivery side, you know, have a reasonable expectation of a standard set of data that allow us to reliably care for the patient, assure there being a bill generated and hopefully some payment coming forward. I just think of what happens every time you use your Visa card: A set of data is exchanged that, you know, identifies who you are, how much credit you have or money in the bank, etc. What are what is that minimal set of health care transaction data that we have to have to viably care for patients as they come across the door?
>> Michael Kappel:
I think that’s an outstanding way to approach the problem. We refer to that as sort of an enhanced eligibility, which contains clinical data as well as the financial and coverage information. But I think if that were universally available and adopted by and the question is who actually provides that, whether it’s a payer or proxy for a payer or a proxy for a consumer it’s a wonderful way to make sure critical information’s available at the point of patient presentation.
>> Carolyn Clancy:
Can you actually, I like the idea a lot as well, but can you talk for a minute about the jurisdiction of the AHIC and Secretary in terms of their recommendations?
>> Karen Bell:
I’m not quite sure what you mean, Carolyn. Could you
>> Carolyn Clancy:
Who are we asking to change their behavior practices, processes, and so forth?
>> Karen Bell:
I think the I’m this is a guess here. What I think we’re trying to do with these recommendations is to have, primarily from this Workgroup, change physician behavior around use of information. And then ultimately, there will be change in patient behavior as well through use of information, and there will be another set of recommendations around that.
>> Carolyn Clancy:
But ultimately, with all the current “momentum” is too strong a word excitement about PHRs
>> Karen Bell:
Yeah.
>> Carolyn Clancy:
this notion of an expanded eligibility set of information, I think, has a lot of appeal. But under what scenario does the Secretary or the AHIC tell the private sector what to do?
>> Michael Kappel:
Well, I think the current mechanism is, if we establish the use case, the use case goes to HITSP; HITSP establishes the standards which go to CCHIT; and if the products that we’re talking about that would use that in CCHIT that there’s a certification methodology for them, that would be the guidance to the private sector to that there are essential requirements that need to be met for certification. Isn’t that the game
>> Jonathan Perlin:
Carolyn, are you asking the question of who swipes the health care Visa card at the point of care and what happens?
>> Carolyn Clancy:
Well, pretty much. But actually, I was just thinking more of the authority. And I guess the other thing, listening to the last comment before yours I think that was you, Peter
>> Michael Kappel:
That was Mike.
>> Carolyn Clancy:
Sorry, Mike. The since AHIP and others announced today that they’re going to be working with HL7 to have to create some standards for PHRs so that they have interoperable PHRs across all their member plans, based on prepopulating with claims data, the timing for this might actually be very good.
>> Michael Kappel:
Yeah, I agree.
>> Linda Fischetti:
I can also add that HL7 just put out a draft standard for trial use to ballot for the personal health record function, which may be present within a personal health record and with the first pass throughout ballot cycle at the DSTU drafting for trial use level of recognition that we’ve it was passed. So I believe that precedence is going to come out tomorrow from HL7.
>> Karen Bell:
Thank you, Linda.
>> Linda Fischetti:
Sure.
>> Karen Bell:
Could we circle back just a little bit? Because it’s now ten past three, and we do need to move on to coming back because there are a number of items that we had talked about. We talked about expanding the medication management priority use case to include interoperability around clinical decision support, to include a more robust set of interoperability standards for e-prescribing and that could be one use case also to include the ability for drug-lab interactions, etc.
Secondly, we’ve now talked about the possibility of a use case around access to eligibility criteria, which certainly, again, is good low-hanging fruit. A lot of work has already been done. We had mentioned a little bit earlier about the possibility of adding the ordering mechanism to laboratories. We’ve not had a chance to discuss that, so I don’t I’d like to just sort of throw that out there for discussion and also would like to throw out for discussion whether or not one would want to include, if we go in the direction of laboratories, the anatomical, pathological result, because has not yet been assigned to any use case.
And then lastly, the concern we’ve had from a number of specialists about needing to have interoperability for wave forms and radiographic information, not just the results, as being a third possibility. And again, that’s mostly from the specialist sector, not the primary care sector.
So those are a couple of possibilities. And I just was hoping that we might be able to hear from each one of you about those four and whether you think any one of them are much more important to move forward with in terms of use case, and some that you feel are just not appropriate right now. (Pause) Blackford? I’m going to call out names. Blackford, can I start with you?
>> Blackford Middleton:
I’m sorry. Could you phrase the question again or restate it?
>> Karen Bell:
Yeah. We’ve got some possibility of four priorities that we’ve discussed. We spent some time on medication management, expanding that to include more standards around e-prescribing, because they’re not all there yet, and interoperability for clinical decision support, as well as the ability to do some drug-lab interactions, etc.
The second use case that we’ve discussed is around eligibility criteria from the provider’s point of view. And I’m saying that because a lot of the PHR discussion, I know, is happening in the Consumer Empowerment Workgroup, so there will definitely be priorities coming from there. But clearly, having access to eligibility criteria is an important aspect for clinicians. So that’s a second possibility.
We had teed up a little bit earlier the possibility of adding the ordering of labs to the lab use case and the anatomical/pathology, because that’s not been included in any of the use cases to date for laboratories. So we could expand on that one.
And then lastly, we heard from the specialty community on a number of occasions about the need for interoperability of wave forms, actually seeing the EKG, actually seeing the radiographic information. And that’s, again, important primarily to specialists, not so much primary care docs. There’s you know, people like me are usually happy with just the results, but not the orthopodist or the cardiologist.
So those are the those are several that we’ve talked about today, a couple that we’ve heard about in other venues, and I’m just hoping that each one of you could perhaps give us some direction in terms of whether any of these would be a strong priority to move forward for the 209 use case development process.
>> Blackford Middleton:
So Karen, let me offer a slight recategorization, if you will, in a way that might help these different proposals be somewhat more internally consistent and coherent.
>> Karen Bell:
Mm-hmm.
>> : Blackford Middleton
I guess my observation is that the #1, the meds management use cases, is pretty well-defined and is self-containable, if you will, with the exception that we might imply that improved meds management necessitates certain laboratory information exchange to facilitate the meds decision support. So let me just make that observation and move on to #2. I guess the eligibility use case is extremely interesting, but I heard there was also interest in sort of the eligibility with clinical extensions, the idea of certain laboratory data or certain patient attributes and characteristics being included in an eligibility information exchange. Number 3, you mentioned, was the ordering of labs and sort of the order management process. I would actually suggest uncoupling the order management or ordering process from laboratory information exchange, because they’re so two entirely different things, if you will, from a pure IT perspective. If you allow that, then I would put into a single bucket laboratory, anatomic pathology, and wave forms as a class of improved laboratory or pathology information exchange and then create this fourth bucket, which would be the order entry or order management.
Let me just stop there and see if people think that makes sense.
>> Peter Elkin:
I think that does make sense this is Peter Elkin for a couple of different reasons. One is that there’s no good terminology out there currently for lab orders, which makes it more difficult to accomplish in the short run. There are good efforts in anatomic pathology, and we have in DICOM a well-tested standard for image exchange image and wave form data exchange, which couples nicely with some of our previous work in terms of the how the report can be linked to the image. So I actually think that that works well. And also, obviously, people who are looking at images that have to do with the physical human body need anatomic classification support.
>> Blackford Middleton:
So Peter, I’m on the same page as you. I guess if this is an acceptable slight mod of the categories, then on my own scratch paper in front of me, I’ve put three stars by the sort of meds-plus use case, and two stars each by the extended-eligibility information exchange use case, and two stars by the laboratory and other information exchange use case. That would be my answer to your how-to-prioritize question. And a negative, actually, by the order management use case.
>> Linda Fischetti:
This is Linda. I think, strategically, we almost have some responsibility to continue to work on the problems related to e-prescribing, based on the recommendation that we just prior we just did. So I agree with Dr. Middleton and Dr. Elkin. In fact, I’ve put medication as #1, because I think we need to (inaudible) that work. And then I, again, put the anatomic pathology and the wave form, accepting the categorization by Dr. Middleton as my second priority.
>> Robert Juhasz:
This is Bob, and I would agree. I think, you know, the medication management is the thing that impacts the most providers, you know, across the board from that perspective. And so from that standpoint, I think that would be, you know, the #1. The others, I would agree, would, you know, decrease, although the ordering of labs and the interoperability of wave forms would probably be next in my estimation.
>> Jonathan Perlin:
This is Jon Perlin. Again, I absolutely support the logic that Linda offers in terms of, as well, the responsibility to the work we’ve just completed and toward Carolyn’s admonition that we think about the killer app. This really begins to be one that enhances safety and efficiency and provides a platform on which to build out.
>> Richard Hays:
This is Rick Hays. I think that our interest would be in the interoperability of wave forms, and I would think that a big part of that is ultimately reducing the cost of medical care and as well as the clinical decision support interoperability would be second. And the medical management, I think, is also an important item, so that would I’d put that as #3. And now that’s kind of ask the Group is that is the is I’m sorry, the medicine management is within the purview of this use case or this consideration to look at the patient side a little more and ask about adherence to medicine management.
>> Jonathan Perlin:
I think that’s a tremendously important question. However, it gets to a level of socialization around how health care is delivered and received and that, in itself, is somewhat ambiguous. So my recommendation would be, as important as I think that is, I think we’re we’ve got enough complexity at the sort of technical level itself, and that bridges not only the technical complexity but also adds onto it a social challenge.
>> Richard Hays:
Okay. I understand that.
>> Robert Smith:
This is Rob Smith. I wanted to kind of ask for a little bit of clarification about one of the use cases we’re proposing. First off, I would support the medication management as being the highest priority for lots and lots of reasons, not only the follow-on to what we’ve been prescribing, but it has so much interrelatedness with other use cases that it really does help, I think, to drive the other process along, both the interrelatedness with laboratory, with diagnoses, etc. in order to make medication management decision support work. But the question really had to do we’ve been talking about interoperability of wave forms and also anatomic pathology. Are we really talking about a use case for image exchange? True interoperability of wave forms may be something that’s pretty technical. It might be difficult, but exchange or interchange of images to support wave forms and anatomic pathology and a lot of other things, probably using DICOM standards and others, would be a very logical use case that, I think, would follow on with a lot of other things and would help to address a lot of the gaps that we’ve been talking about in other ways. And I would certainly endorse that as a use case.
>> Lillee Gelinas:
Karen, this is Lillee. I’ve wanted the Group to weigh in, but I just want to make sure, by way of consensus, that you know that I agree with everyone about the medication management use case being the #1 priority. I think Carolyn gets the prize for the most pithy descriptors, but when we talk about moving trains and low-hanging fruit, we do have a responsibility as a Workgroup to continue this area around e-prescribing. I think for us to do anything other than that as #1 sends a signal that we’re wavering. So certainly that being #1, and I agree with the discussions in the past. What I before we were actually going around voting, I had med management 1, adding ordering labs 2, interoperability of wave forms 3, and access to eligibility criteria 4.
>> Karen Bell:
Thank you, Lillee. Anybody else?
>> :
(Inaudible) lab, are we talking about (inaudible)?
>> Karen Bell:
The ability to order laboratory order lab results in an interoperable way from any laboratory at all, from any EHR, so that’s become the interoperable piece.
>> :
So are we talking about that I can send a lab order of my (inaudible)?
>> Karen Bell:
Yes.
>> :
You know, for us, in thinking about VA and DOD sharing, some of the facility that’s going to be integrated that’s one of the key obstacles we have to tackle, the interoperability level. That’s what I would put mine as first, and then medication, and then eligibility.
>> Karen Bell:
Mm-hmm.
>> Blackford Middleton:
One of the observations I’d like to offer on the order management or order entry use case this is Blackford is that there may be a significant knowledge-based requirement for those practitioners who live in a world where their patients have labs covered at certain laboratory facilities or entities and not at others, depending upon the health care plan and products. You know, a patient may have Quest labs paid for readily but not some others. So that’s a significant issue with that kind of order entry.
>> Jonathan Perlin:
I would just echo that. You know, as we’ve started looking at, you know, the quality reporting initiative and things you know, that was a significant barrier for some of those, you know, cases for reporting, just because, you know, you’re looking for something and it’s somewhere else. You know, it’s not loading up to your system, so...
>> Karen Bell:
That brings up the interesting point, though, of possibly linking order entry laboratory order entry with eligibility, because you’ll need your eligibility information to be able to do this effectively.
>> Blackford Middleton:
From an implementer’s point of view, I would make a plea for one step at a time. (Laugh)
>> Karen Bell:
Okay. (Laugh)
>> Robert Smith:
But that’s absolutely correct, that you do need that information. It’s sort of the lab equivalent of a formulary.
>> Blackford Middleton:
Yeah.
>> Karen Bell:
So if we were to move forward with the ordering of logical use case, the eligibility piece will certainly come in as we develop that use case, because it will be an important criterion to move forward in the ordering process, or at least or some of the eligibility will come in. It won’t necessarily be all of it, and it won’t necessarily be a use case on eligibility (inaudible). It begins to get into that realm.
>> Blackford Middleton:
I would suggest it would be useful to perhaps uncouple the knowledge of what labs are approved for ordering, you know, for a given patient from the electronic communication and, you know, closed-loop order management processes. Even in the most sophisticated environments, you’d be surprised how many orders are still printed through paper and then carried by the patient to the lab.
>> Karen Bell:
Yeah. No, that’s true.
Well, we do need to come to closure. I think there’s no question that medical management has won hands down as our top priority. I think that the ability to move forward with access to more information around wave forms may be our second priority. And I think that the ability to order labs may be our third, except for the VA and the DOD. So we may tee up these three and eligibility as a fourth, if that’s all right with this Group.
>> Robert Smith:
That sounds fine. I would just say the one thing on eligibility that it’s important if we have this discussion that it’s someone said “extended eligibility.” I think it’s very important that we do go beyond the current definition of eligibility of otherwise, people may lose the context that went into that discussion.
>> Karen Bell:
Absolutely. Well, that
>> Blackford Middleton:
I guess, Karen, I’m not sure if I was distracted or heard something slightly differently. I thought that the eligibility was a proxy for or extended-eligibility information exchange was sort of a proxy for the CDS, clinical data exchange, or the C or the clinical data exchange standard effort.
>> Karen Bell:
Oh, I misunderstood. I apologize. So basically, what this is would be looking at the elements within the CDA that I’m sorry the CCD that we would move towards semantic interoperability room. Is that correct?
>> Blackford Middleton:
For that subset of or dataset, whatever we think are the most relevant.
>> Karen Bell:
Yeah. Okay.
>> Blackford Middleton:
And then, if that were the case, then I wasn’t keeping track of the voting, but I thought that actually had a somewhat higher priority.
>> Karen Bell:
Well, I think what we can do is, there it kind of is it’s pretty much on the cusp. So what we can do is move forward with all three of the medical management on top and these two others beyond that and then perhaps the ordering of lab third, because I think that’s a complicated one anyway.
>> Blackford Middleton:
Okay.
>> Peter Elkin:
That sounds fine.
>> Karen Bell:
Okay.
>> Robert Smith:
So does that mean that the extended eligibility will be included or will not be included?
>> Karen Bell:
It will be, but it will be included with the caveat that of the discussion. And I apologize for missing that piece on with the relationship to the CCD.
>> Robert Smith:
Thanks.
>> Karen Bell:
Okay. We’re going to get this up and send it off email and everybody can sign off on it. And revote if you’d like, but I think that’s where we’ll go. And I do apologize to Reed Gelzer for going a little over on this discussion, but I think that we certainly have time. We’re we scheduled until 5:00 this afternoon. And so if it’s all right with Lillee and Jonathan, I’ll just jump right into the next part of the agenda. Is that okay?
>> Jonathan Perlin:
I think that, Karen, it is. I just need to let you know that about two-thirds of the way through, I’m going to jump off the call to welcome (inaudible) executives here, and but thanks to you and all the Group for terrific work. And as we introduced earlier, this is also tremendously important area in terms of ensuring the reliability, validity, and ultimately integrity of data.
>> Karen Bell:
Thanks so much, Jonathan.
>> Linda Fischetti:
Thanks, Jon.
>> Karen Bell:
Thanks to everyone. Okay, moving right along, we’re privileged to have this afternoon with us Dr. Reed Gelzer, who is the Chair of the Prevention Workgroup and who has been working very closely with RTI, which has been the contractor that’s been helping us look at data integrity within the EHR. You may have heard, as Jonathan mentioned a little bit earlier, that some of this has resulted in concerns about fraud and abuse. What I would like to underline is that there is fraud and abuse in the system. We certainly know that. Very little of it is happening in the physician office, so that the EHRs that physicians are using now, and to which these recommendations are appropriate, become the vehicle by which there can be more integrity in the actual data management and data reporting.
So recommendations were made to the American Health Information Community from this report from RTI. Dr. Gelzer will be going through them. But I’d like to particularly call your attention to several of them. There will be Requirement 5 and Requirement 6 that Dr. Gelzer will be addressing. Those are the ones in particular that this Group will be asked to vote on as to whether or not these recommendations are ones that should be supported and accepted by the AHIC.
In addition to this particular last piece of work for you all, there will also be a discussion that Dr. Gelzer’s going to lead regarding some of the issues around the new e-discovery rule and how that is affected or how that affects various EHRs in terms of the basic system, essentially the back-office systems within the EHRs themselves. And so he will be spending some time bringing those issues to light, with the idea of helping us work through whether or not we should be making some recommendations around that as well. But before we get to that, the RTI reports will require some action on your part. So without further ado, Dr. Gelzer, Reed, please take over.
>> Reed Gelzer:
Great. Thank you very much. And I must say, you’re rather brutal to your participants here to not give them an opportunity for a seventh-inning stretch. So I’ll try to be concise and move directly to the support of the discussions that are on your agenda.
If I can have the next slide, please. Great. And this is just overviewing what I’ll be covering today, and so you might make a few notes to areas that you might want to come back to. And next, please.
Particularly, it’s important to understand that this report has, like all reports, a context. It was actually the third in a series commissioned by the ONC, originally from the American Health Information Management Association, initially on computer-assisted coding and how that would facilitate improved data quality in the documentation to coding aspects of health care data keeping.
Then the second report was the one that more focused specifically on health care fraud as a tremendous infrastructure issue in the United States. It was of particular interest because a number of authorities in the HIT community had hoped that EMRs would pose enough of a control or place new types of bounds on health care fraud to actually demonstrate early return on national investment by reducing fraud, therefore justifying federal expenditures to support subsidization of EMR implementations. And unfortunately, the initial report this report did not support that there would be an early-on reduction of health care fraud simply from implementing EMRs.
And that, as you can no doubt understand, logically led to this third report which was, in essence, an answer to the question an attempt to initially answer a question, “Okay, what do we need to do in electronic health records as far as incorporating aspects functionalities that will prevent lower the likelihood of health care fraud and also aid in its management and prevention and control down the information stream?”
So the next slide, please. This is simply an excerpt. We always like reports to have some scary numbers to give us a context for why this is of great concern. The second fraud report estimated that in the range of 10 percent of annual expenditures for health care were overt fraud. And as Karen mentioned, this is not the periodic, at-the-margins upcoding that may or may not be questionable. This is criminal. This is these are criminal organizations that are creating entirely fictitious encounters that are bilking DME vendors. This is overt, conscious, specific fraud. And so clearly, the numbers themselves justify attending to this as a matter of national concern. And the next, please.
And from the second fraud report and there is one among the attached documents that were sent to you yesterday, there is an excerpt from the second report that is the law enforcement community’s priorities. And one of the things I’d like to bring to your attention about this list of 17 items is that they are also very important for overall data validity and integrity for all end uses. And that was in many ways, for those of us participating in this study, a great deal of our motivation came from the concept that, we’re not out to police the any particular subset of the overall health care industry. These are all initiatives these are all priorities that would serve everyone’s interest by making sure that when you get that information through an interoperability portal that it’s not only valid and accurate, but it’s trustworthy. And it’s the trustworthy aspect of health care data that, in many ways, has been unfortunately an assumed consequence that has not necessarily been attended to closely in systems imple design and implementation and support. Next, please.
So overall project purpose, again, is here. RTI was the primary contractor, with AHIMA as secondary. And there was a public comment period. One of the criticisms of this report was, the public comment period occurred a year ago over the holidays and was brief. Nonetheless, as you’ll see in the body of the report itself, the community did respond with quite a large volume of public comment. Next, please.
And two specific outcomes included working with organizations that are consumers of guidance for where recommendations should go, not the least and very important, of course, being both HITSP and CCHIT. And the fact they were specific, intended end users of the recommendations is an important point I’ll come to a little later. Next, please.
And just the next two slides you can go to the next one as well is identifying the leadership and also the wide range of participation we had from a number of organizations, both regional and national in representation. And it was a very invigorating and very fascinating purposeful discussion. And the next, please, and then the next one also.
And the recommendations are listed here, just in general forms. And there is no accident. There is intent to the order that these are listed. The first was functions relating to essentially how do you assure validity and accuracy up front, and how do you incorporate means for evaluating validity and accuracy of health care information retrospectively. And next, please.
Not to say that these latter ones were less important, but we recognized in drawing up this list that the initial ones would both be the most difficult in many respects, but also the most important. Next, please.
This is a point that is one that we highlighted early on in our process. One of the things that we did prior to public comment was to speak with the leadership for HITSP and CCHIT to get their initial impression of the work done to date. And in both cases, an important point was highlighted that in neither case did fraud management or medical legal integrity or medical legal validity none of those were specifically called out in HITSP or CCHIT charters or mandates. And so one of the unwritten outcomes of the of this report is that, in the background, those areas are being elevated within the many other competing priorities for these organizations in their work to advance health information technology. Next, please.
And in general, the universal message, if you will, is that everyone has an interest, of course, in valid, accurate, and trustworthy health care information and that this applies to all end users and all participants in the overall process. And one of the things that was particularly fascinating about this Workgroup is that we did have representation from every stakeholder, and the unanimity of purpose was a very enjoyable part of the discussion. Next, please.
And I wanted to focus on those two areas, also, that were called out in your agenda today that you were to attend to. This first one, Requirement 5, regarding evaluation and management coding this particular discussion was intended primarily to highlight that within the discussion group and within the health information technology community that we drew from, there had been concerns about the overall prospect that information that is prompted that prompting mechanisms always need to be carefully considered when they’re being implemented in systems. Clearly if you are you know, there’s prompting, and then there’s leading the witness, and then there’s filling in the blanks for the data entry person.
And fortunately, the industry in my work as a field evaluator and systems function-testing authority, I have found that, already, the industry has come a long ways from some of the early prompting mechanisms that sometimes gave one concerns about whether or not, in the hands of an inexperienced user that people might inadvertently document in misleading or inappropriate ways. Fortunately, that is not the case. And the greater intention of including these within the recommendation was to simply underline that ongoing process within the industry that when these kinds of system functions are being designed and implemented and used, they have to be used, of course, with care.
And the specific language here was drawn up in recognition that these recommendations would go through a number of filters, including public discussion within the within CCHIT’s internal processes and then public debate, etc. I also had the opportunity to speak with someone from the AAFP who had particular concerns registered to him from his constituents about this. And so it’s an area that’s being watched fairly closely to make sure that accidentally, or as an unintended consequence is that very legitimate documentation and decisionmaking prompts are not thought to be barred by the language here. The next slide, please.
Requirement 6 is one that is much more concrete in the sense that one of the current difficulties with many EMR functions, including those currently certified by CCHIT under current guidelines, is that under the early recommend under the early functional requirements, as CCHIT is developing functional requirements forward, is that, for example, currently, an EMR only is required to accurately record the final author of an encounter entry. And especially in the ambulatory space, where it’s more common the case in fact, it’s nearly 100 percent of the time there’s more than one individual documenting within an encounter. There’s a current significant issue with data quality and data integrity and also the potential for inadvertent ,fraud where documentation by one author can if the system is not used with very, very close attention to how it actually functions, that documentation by one author can be misrepresented as from another. And the for example, as an extreme case and unfortunately this is an extreme case that I have actually observed in the field: A non-credentialed person may be providing services that are outside of their scope of care for example, a medical assistant doing a physical exam and documenting it and then the clinician, nurse practitioner, physician, whomever, only checks pertinent positives and negatives, and the system represents that the physician actually did the entire encounter. That’s, again, an extreme example. But it is the kind of issue that’s intended to be addressed here.
Also, there are many settings in particular, subspecialties, one of the most common being in ophthalmology there’s often a scribe or other intermediary that may be actually executing the documentation, and so means need to be incorporated in the system to accurately represent who documented and what the relative roles of multiple authors may be within a given encounter. So that’s just some specific information about those particular and you’ll see that there are the page numbers and also the PDF page numbers for both 5 and 6 if you want to read those within the report it itself. Next, please.
And at this point, I will pause for discussion and questions. As Karen mentioned, I would like to highlight the 2006 discovery rules that were promulgated by the Department of Justice their modifications to the rules of civil procedure that apply to all federal courts. Essentially, this is in recognition that information systems are going to be have become critical infrastructure for our country in all aspects of business, not just in health care. But we’re learning all the time. For example, in the Enron/WorldCom cases, I’m sure people were will recall that there were concerns about the destruction of business records. And so the new e-discovery rules were intended to codify how parties can come to an agreement what information, including structural information or what’s often referred to as metadata, gets introduced into legal proceedings. And the whole subject matter area of metadata is has become increasingly important. At least one of the AHIC use cases, the biosurveillance use case, includes stipulations regarding the conveyance of both data and supporting metadata. Also within the HL7 EHRS Records Management and Evidentiary Support profile, also known as the legal EHR profile that’s now out for public comment, there’s a great deal of discussion about metadata and discovery processes that are intended to work off from these new e-discovery rules and to assure that systems attend to the growing requirements for business records, in particular, for supporting their source validity and integrity.
>> Karen Bell:
Reed?
>> Reed Gelzer:
Yep, so I’ll pause there.
>> Karen Bell:
If we could pause there for a moment, because we do need to go back and have the Group literally vote on Requirements 5 and 6, whether or not they are acceptable as written or whether or not there might need to be some changes to either of them.
>> Reed Gelzer:
Very good.
>> Karen Bell:
So and then we can go back to the e-discovery rules, which we, I think, all agree are going to be very important.
So with that in mind, could I turn your attention back two slides to Requirement 5? Thank you. And as Reed indicated, these are essentially two recommendations that came forward under the coding piece about the prompting. There were concerns about some of the AHIC members are actually concerned about some upcoding here, but I think those issues were discussed and fairly resolved. But there were also concerns about unintended consequences, as Reed indicated. So hearkening back to his comments and his discussion and perhaps even Blackford, I know you were on the Internet as well, and you might want to comment here. We certainly need to know from everyone how the vote should go with respect to this particular requirement.
Any quest first off, are there any questions about it at all?
>> Robert Juhasz:
Hey, Bob here. I have a question about the second portion. Was there any I mean, has there been any guesstimates by anybody as far as what impact it would have from any cost as far as the if this were implemented as far as E&M coding, either up or down?
>> Reed Gelzer:
In terms of the impact on claims submitted and resulting cost?
>> Robert Juhasz:
Yes, mm-hmm.
>> Reed Gelzer:
No, there was not.
>> Robert Juhasz:
Okay. I guess the other question that I would raise is, on 5.1, it says the system shall be capable of prompting for omitted, necessary administrative data codes, which certainly our system is capable of that but is not switched on so that it’s mandatory. Is the intent really to make this so that it’s not only capable, but it is utilized?
>> Reed Gelzer:
The intent is was to suggest that EMR design incorporate exactly the kind of flexibility you just described. The capability is there. However, there are institutions that, for whatever reason, have decided to take a higher level of caution on issues such as this. And so the implicit suggestion is to do exactly as you’ve described, which is to have this a function that can be disabled.
>> Robert Juhasz:
Okay. Thank you.
>> Lillee Gelinas:
This may this is Lillee Gelinas. It may be a really dumb question, but does this apply to inpatient and outpatient, or is it just outpatient?
>> Reed Gelzer:
The majority of this report structurally ended up directing its attention to outpatient. However, that was not our mandated intent. It was simply a creature, if you will, of the participant pool. And we did attempt to revise and broaden the requirements where they like, this one, for example, I think, is probably one that’s most directed to an outpatient setting, because E&M coding isn’t usually done by the documentation function in inpatient EMRs, as in the limited exposure I’ve had to them.
>> Lillee Gelinas:
It’s just my universe in the last couple of weeks has all been around new CMS IPPS rule, the inpatient perspective payment rules, and present-on-admission requirements coding, that type of thing. So I just needed that explicit clarification. Thanks.
>> Peter Elkin:
This is Peter Elkin. The Part B Medicare rules for the physician billings do get coded with the same sort of E&M coding rules that are used in the outpatient setting. So for that inpatient type of bill, rather than the hospital bill, it may apply. Just a clarification.
>> Reed Gelzer:
Yep, exactly so. Thank you.
>> Mike Kappel:
Well, just one more question on that. On 5.2, that would imply, then, that things like procedure documentation, which would be reimbursed under regular CPT codes, would be not covered by 5.2?
>> Reed Gelzer:
I would agree that that’s a fair interpretation, though I don’t recall that being explicitly discussed.
>> Mike Kappel:
Okay, because one of the issues we frequently run into is that physicians while they’re very good at performing procedures, they are not very good at following the advances in coding, and therefore, you know, their documentation may not be consistent with coding schemes, and it’s not due to any attempt at fraud. It’s simply that their documentation’s based on the way they were trained, not on the code procedures.
>> Reed Gelzer:
For those of you who are lacking for discretionary reading, there’s one of the ASTM documents I’m sure you’ll all rush home and download it having to do with amendments to health care information has a very eloquent paragraph on the difficulties of writing standards for technology that’s in transition. And in many cases, a lot of the issues that I observe are simply the fact that people in the real world have a much more infinite range of possibilities that they will create for us to then try to figure out whether or not they fit or not.
>> Linda Fischetti:
Hi, this is Linda Fischetti. My apologies. I had to step out of the room to deal with an OMB communication. So I’m back in, and just need to do a little bit of reorientation here. I’m looking at Slide #14 and 15. Correct? Requirements 5 and 6?
>> Reed Gelzer:
Yes.
>> Linda Fischetti:
Okay. And so, Karen, as I read these, I assume they need to be rewritten in a way that they would be actionable by the Secretary?
>> Alicia Bradford:
Hi, it’s Alicia. Karen had to step out to attend a meeting with Dr. Kolodner. No, the idea the discussion that happened when these recommendations were presented to the AHIC a few workgroups were called out as to look at the requirements and have a discussion regarding those requirements and, you know, have a Workgroup (inaudible “sold”?) about them. Any additional comments that we have about them, whether we think they are acceptable or we want to make some, you know, amendments to it or just have discussion, and the goal from this the CPS Workgroup already did this recently with Requirement #8. So we were asked to look at 5 and 6. And the goal is that we’ll send a letter to the AHIC at the January meeting, along with CPS, regarding our discussion of these and whether we’d accept them or not and whether we want to make amendments to them or not.
>> Peter Elkin:
This is
>> Linda Fischetti:
So for Requirement #6 to me, that looks like it falls directly within the domain of the health information management professional who does do the interpretation of the local legal envi regulatory environment within which you practice, to determine document attestation and how documents are dwelt with. The I recognize that these come from an activity that was supported by AHIMA. And within my environment, I would go to a health information management professional to make sure that we had the correct, the correct wording for #6.
I guess, in this environment, we assume inheritance because it was an AHIMA activity that, in fact health information management professionals believe that, in fact, this could be implemented throughout the United States as a policy which would then have inheritance.
>> Reed Gelzer:
Yes. Michelle Dougherty was the support staffer from AHIMA on this report. And as you know, she’s also the Workgroup leader for the HL7 Legal EHR now called Records Management and Evidentiary Support Profile. So
>> Linda Fischetti:
I’m one of her biggest fans. Yes, she’s wonderful.
>> Reed Gelzer:
Yep, we love her dearly.
>> Linda Fischetti:
I’m actually a Co-chair of the HL7 EHR PC, but I’m going to voluntarily resign in January.
>> Reed Gelzer:
No!
>> Linda Fischetti:
Well (laugh), I know they’re doing amazing work, but I need somebody else who has the ability to spend more time with them to be there more so than me.
>> Reed Gelzer:
The only other comment that I would like to make is, again, as a field investigator, although some may see these as oddly meticulous, unfortunately, these are not functions that one can currently assume are in an electronic health record system.
>> Linda Fischetti:
You are absolutely right. So and this is Linda. Obviously, I work for the Veteran Health Administration. We worked on the clinical functions first before we started dealing with billing functions. It just was less of a priority for us, which we understand is a disconnect from the private sector. And so our the level of sophistication of our ability to deal with coding is lower, probably, than in the private sector, where in fact the ability to run an office is very much dependent on that. So you’re right: There is not an assist a the assumption should not be made that all systems could just turn this on immediately.
>> Alicia Bradford:
Hi, Reed. This is Alicia Bradford again. I have a clarification on the statement you had made earlier regarding CCHIT and these requirements. Are they the two under 5 and under 6 are they currently in the clarification criteria or not? And if not
>> Reed Gelzer:
No, they are not.
>> Alicia Bradford:
Then none of these are.
>> Reed Gelzer:
Correct.
>> Alicia Bradford:
Okay. And so, as we hopefully move to a vote or have initial discussion that we need on these two items, perhaps if the Workgroup is in agreement with these functions, we could at least note that, you know, they’re not part of the certification criteria and that we’d like to see that become part of them, as we’re in agreement with them in general.
>>
(Inaudible “Accurately”?), the there are things that are like 6 in the current CCHIT criteria, assuming that the proxy actually signs into the system. And in the current test scripts, the way that it was written for example, if an allied health professional entered vital signs, for example, and signed onto the system and did that, left a note tending, there is, as part of the current criteria that that must be daytime and user stamped for all entries they make.
>> :
Well, (inaudible). I wanted to discuss 5, particularly 5.2. The thing is, you know, this is an interesting, you know, assumption, but, you know, I think there’s and I understand the need to try to avoid talking about coding fraud, but there is a reality out there. And the reality is, there are tons of firms selling soft (inaudible) in an attempt to either do outright fraud or to upload, and those tools are out there already. And whatever it is we do, there’ll be, you know, version 8 or 9 or 10.0, whatever, to change it. And it’s a game that people who are intent on fraud play. And I think the point is, you know, if you want to really maximize upcoding, there’s going to be a way now, and there’s going to be a way later, just the way fraud the way all fraud operates in whatever realm.
But, you know, I think the more important thing to talk about first is how you could use audit logs to detect fraud. And one would assume that if you created these audit logs, there would be distributions of times, author, what they did, when they did it, when they changed the coding. And that would provide this electronic trail that would be more telling both, you know, from an initial point of view and as well as in a court later on. If they disproportionately change documentations, then the audit would clearly prompt something to say, you know, “There’s a higher possibility that fraud is going on here. We should pull that file.”
And I’m just wondering if we shouldn’t go in more the direction of creating alerts and to aid in fraud detection versus trying to, in a way, get rid of what could be important information that could be cost effective.
>> Reed Gelzer:
Sure. As I mentioned, the order of the requirements was not accidental, and the #1 requirement does have to do with auditing. And you’re absolutely right that it is not only understood, but it’s also a necessity that underlying whatever one’s intent might be in using a system, you still need the underlying audit functions to validate who did what, when. And unfortunate and so there is a that section is actually quite lengthy and goes into a greater degree of specificity.
The intent on Requirement 5 was to, in essence, convey to CCHIT that there was, within our Workgroup, a majority who’ve believed that the current requirement there is a current requirement much like 5 that is quoted in the report, in the full text of the report, that that was too amenable to abuse as written. And these were essentially recommended modifications of that language to achieve two purposes: one, to dampen an impression that the majority got from the current language; but second, to announce to the marketplace as a whole that these are issues that when you’re looking at how you purchase and implement and use and support an EMR, these are very important functions to make sure that you understand exactly what they do and how they work so that you can assure that they fit in with the documentation guidelines for your specific users and your institution.
So in essence, what I’m saying is that the message is more the context and the somewhat more cautious language rather than the s than the word-to-word word-by-word semantic exactitude of the statement, if hopefully, that’s clear.
>> :
One other question on Requirement 6: When you are talking about retaining the user stamps, I assume that means that if those were retained in the audit log that was created in the encounter, then that’s adequate to meet 6.1 and 6.2?
>> Reed Gelzer:
There’s another part of the requirements that stipulates that there can be two different output versions. In some settings, knowing exactly who documented what is relevant to the end user. For example, if you are Medicare or if you are Aetna or UnitedHealthcare or CIGNA, exactly who did what is important to you, where it may not be from a clinical end user’s point of view.
And so the there’s two levels in a native report that’s specifically for a clinical end user. This greater degree of specificity may or may not be present. But in an output that we designated a business record output, this level of detail would actually be part of the documentation and not just in the audit.
>> :
Okay. So and what would one of those business output reports I mean, like a claim form?
>> Reed Gelzer:
Well, it would be for example, if Medicare asks you for the actual documentation for demonstration of medical necessity
>> :
Okay.
>> Reed Gelzer:
you would send this as the supporting documentation.
>> :
Yeah, but it wouldn’t modify your existing claim forms. It would be as supplemental information.
>> Reed Gelzer:
Correct.
>> :
Okay.
>> Lillee Gelinas
So are we ready to move to a vote here? Just asking the Group. I don’t want to shut down the conversation, but I hear a lot of good feedback
>> Nhan Do:
Right, so the vote is whether do we accept Requirement 5 and 6? If I am concerned with the way the Requirement 5 is written, it implies a certain design. I know the DOD’s implementation of an automatic E&M coding does not prompt in or alert. So if I were to vote, I would ask if that can be rewritten so that way, it doesn’t imply a sort of design to that automation.
>> Robert Smith:
This is Rob Smith. I was going to say something along the same lines, which is, for 5.1, it implies a requirement that the system shall be capable of prompting. Are we sug and that seems a little bit inconsistent with some of the prior discussion. Are we suggested this as a desired feature or a required feature?
>> Carolyn Clancy:
Who can answer that?
>>
I think Dr. Gelzer. Can you answer that?
>> Reed Gelzer:
Sure. The context was more the understanding that these are ubiquitous functions and that they should be amenable to turning them off, for 5.1. And for 5.2, if you have them if your system has functions that are driven specifically by administrative processes, etc., etc. so the sense is that they are desirable and that the requirement is that it’s a capability, with the sense that, as a capability, it can be either enabled or disabled.
>> :
We may want to work on tho 5.2, I think, is fine. We may want to work a little on the language of 5.1. It seems a little more directive than the conversation we’re having.
>> Lillee Gelinas:
Well, we have two options: if someone can wordsmith it now and take a moment and do that, or we can do what we did with a lot of the e-prescribing recommendations and we took a vote by email.
>> :
As to one point on 5.1, in CCHIT, they distinguish between among other words, but “shall” and “must.” So the way I would interpret 5.1 is, a capability that can as pointed out before is, it can be required, has the option of being turned off, but not one that is mandatory in terms of implementation. So I’m getting at 5.1, for those who follow the CCHIT, is certainly consistent with the wording that the criteria that CCHIT typically writes, if that’s any help.
>> :
Right. (Inaudible), you know, for a DOD system, the way our E&M coding works is that, if the physician chooses to use a structured documentation, all that coding is transparent. He can choose to look at it or not, or override it if he you know, there’s a lot of pretexting on it. But it does not prompt him for anything. He can just leave it alone. So that kind of implementation the way that’s read is specifying a way you should implement it, but that’s not how we implement in our system.
>> :
You know, I’m familiar with the way the coding engine works in Alta, and I would agree with you that this is consistent with the way the system already works.
>> Lillee Gelinas
So do we want to go around the who’s got an attendance list to be able to do that? Do you have one, Alicia?
>> Alicia Bradford:
Not I think several members have dropped off, so...
>> Lillee Gelinas
Do we still have a quorum?
>> Alicia Bradford:
I don’t know. They’re checking on that. I would ask that if we wanted to reword this, to take out any language that seems to talk about the structure or design, if DOD and VA can offer that language up now. And if not, maybe wordsmith that and send it to me, and I can circulate the revised language we would propose to AHIC. It sounds like that really would do that with 5.1. I don’t know that if we have
>> Lillee Gelinas
I think I heard consensus on 5.2, though.
>> Alicia Bradford:
Yeah, and I believe is there consensus on 6? Maybe members could just
>> : Lillee Gelinas
Yes. I don’t know if we need a motion, but I would move 5.2 through 6.2.
>> Alicia Bradford:
Okay, well, how about if we if DOD and VA could submit to me, maybe by week’s end, the revised language that discusses structure, such as prompting? And I could circulate that to the groups and several have dropped off along with the acceptance of the other three requirements. We could hopefully wrap this up by week’s end. And the goal would be that we would write a joint letter with CPS regarding our findings and discussions on the that sound
>> Lillee Gelinas
Is that acceptable to everyone?
>> Peter Elkin:
That sounds fine.
>> John Loonsk:
Yes, great. Fine.
>> :
Yes.
>> Alicia Bradford:
I think we can let Dr. Gelzer proceed with his e-discovery and metadata discussion so we can wrap up this long meeting.
>> Lillee Gelinas
Yes. But indeed, if we’ve lost our quorum, we have to be very careful about declaring consensus, because we’re we can’t do that.
>> Robert Juhasz:
Not to belabor this, but one thought about 5.1, the second sentence: This could include the capability to prompt a physician if the selected code is not consistent with the documentation. Maybe that would be better stated in the positive that it could prompt the you know, the physician with or could state the appropriate E&M code for the documentation present. And that way, it doesn’t sound, you know to say not, actually sounds like somebody’s not doing the appropriate thing.
>> Lillee Gelinas
Mm-hmm.
>> Peter Elkin:
So I just bring that up as we’re thinking about retooling it, just to make it more in the affirmative that, you know, this is what you put down, and this is what you can code based on that. And it’s not really prompting you to you know, the word “prompt” suggests that it’s enticing almost. So I just bring that up as a potential.
>> Lillee Gelinas
Okay. So we have our action now, Alicia?
>> Alicia Bradford:
I do.
>> Lillee Gelinas
Okay. Let’s move to the e-discovery piece now.
>> Reed Gelzer:
Very good. I’m sure that all of you read from
>> Lillee Gelinas
The AMIA piece?
>> Reed Gelzer:
the AHIMA piece. I would encourage you to read the article that discusses the anesthesiology case, simply because when I’m working with the subject matter, I almost universally find that having an example of what metadata is, how it serves the best interests of the patient, how it best how it serves the best interest of the clinician and all end users of data is always greatly assisted by having an illustration of that sort. The particular case does involve an anesthesia case with a bad outcome.
And so, the point that I would simply like to bring to your attention I’ve had the opportunity to listen to your discussions throughout this meeting, and in all cases, the somewhere in the discussion the issue of the reliability of information the trustworthiness of information is a recurring theme. And one of the difficulties at this point, unfortunately, is that these are, as I believe at least one other participant mentioned, not necessarily tested of functions. And in my particular work, that is often one of the functions that I serve is to actually test specific functions. And so the e-discovery the new e-discovery rules have been the most useful as a tool to bring to the health information technology discussion that we’re still learning, as we all know, and we’re still gaining new experience in how these systems actually behave in the real world.
One of the most difficult areas in terms of health care fraud management, with regard to electronic medical record systems, is that there is relatively little case law, and that case law that does exist is not filed in some, you know, broad-banner headline, “Look here! This involved an EMR!” And so a lot of the cases that have come up with regard to the importance of the discoverability of information have been it’s a fairly small population of real-world cases, one of which, though, interestingly enough, was does come from the VA system, where there was a case where the system’s copy and paste functions were misused to create fictitious documentation.
And so most of us are working on these issues, often in our spare time, and usually because of our passion and enthusiasm for health care improvement, with the assumption that everybody is in it for as good a purpose as we’re in it for. And so, as you weigh things, like for example, one of the discussions earlier in your meeting was workforce for health care technology. One of the questions that comes up quite often in especially in the large institutional settings, is that often, the person who’s most expert in data integrity is someone who’s not necessarily at the table discussing how the EMR is evaluated prior to purchase. And so one of the areas in workforce development is simply how are we going to credential people with sufficient authority to their position that they can speak sometimes against you know, in an atmosphere where recognition and addressing key problems with system functions may not be a welcome part of the dialog for the organization.
So I’m going to simply leave you with my contact information in case as you discuss these issues, if there are particular aspects of how e-discovery interacts with these requirements. Essentially, e-discovery is about the legal system’s recognition that we’re required to ask new kinds of questions with regard to information integrity as we learn how these new systems work. So are there any particular questions about areas where these issues have popped up so far that might be illustrative and relate to your specific tasks?
>> Lillee Gelinas:
I know I don’t have any. I did read the anesthesiologist’s piece. And unfortunately, having been in depositions myself, I can really see that this is a picky and extremely important issue. I’m just not so sure what we would act on today, other than let’s just make sure we have the issues on the table for the Workgroup, which is an important component here.
>> Reed Gelzer:
I guess the question that I would pose for you to periodically ask in your deliberations, just listening to the discussions, is, “How is this validated in the real world? If I send a prescription to a pharmacy, what tells me that it was received? What tells me it was received as intended, and what tells me it was actually acted upon at the recipient’s end?”
>> Peter Elkin:
So can I ask a question? This is Peter Elkin. Does this mean that we have to or we should be encouraged to require ongoing audit of data accuracy, and not just systems in place to create audit trails, but that there should be a process that is associated with those audit trails that routinely looks at the level of quality of the data?
>> Reed Gelzer:
Can you affirm that you and I have never met and that (laugh)
>> Peter Elkin:
Yes, but my bank account is no, I’m just teasing.
>> Reed Gelzer:
No, I could have not I could not have said it better.
And I would like to also encourage the understanding that these are moving targets in all industries. Where I have where I’ve had some of the most fruitful discussions have actually been talking with people in supply chain management and just-in-time manufacturing environments, for example, or systems controls for complex systems, and sometimes even simple things like for example, there was a very interesting article about a Mercedes Benz model that had trouble with climate control systems, because it had four climate control sensors in the cabin, and one sending wrong information could override the information coming from all other sensors. And for example, it might be telling the system that the cabin temperature is 105 and turning on the air conditioning full blast when the other three sensors said, “No, it’s only 48 degrees.”
>> Peter Elkin:
Yeah. In medicine, when this happens, somebody dies.
>> Reed Gelzer:
Exactly so. That’s one of the things that, in my work, I have found most disturbing is that in some cases, we have become so inured to the errors that we experience every day, not necessarily the errors of knowing we have bad information, but of not necessarily questioning information that’s out of sync with other information at our disposal, the most famous the most common example being, you know, the patient who has a sei has a one-time seizure doing during due to an idiosyncratic event, but somehow it gets recorded in their problem list they have a seizure disorder.
>> Peter Elkin:
It’s actually quite common. You know, what I was just going to suggest I know we don’t have a quorum, and it may need to be sort of shuttled around a bit, but there may need to be a statement to AHIC from our Working Group stating that we would encourage more informatics research to determine the most cost-effective audit schedules for the accuracy of clinically transmitted data, because I’m not aware and I keep up quite well with the bioinformatics literature of a whole lot of work that’s been done along that line.
>>: Lillee Gelinas
That’s a great suggestion.
>> Reed Gelzer:
That is a great suggestion.
>> Lillee Gelinas
Any other comments? (Pause) I think we still Alicia was telling me we still do have a quorum, but I’m not so sure we need a vote on anything, do we?
>> Alicia Bradford:
No, but I think I think that the suggestion sounds really good, and when I’ll circulate the draft response to these, I’ll incorporate that. And you guys can all vote and wordsmith online and help me out with the final language.
>> Lillee Gelinas
That’d be great.
>> Alicia Bradford:
I think we have one question here in the room from an ONC staff member that actually was the Project Officer, I believe, for this r this contract.
>> :
Yeah, I was sort of this is (inaudible) with ONC. I have a question for Dr. Gelzer.
>> Alicia Bradford:
Please.
>> Reed Gelzer:
Yes.
>> :
What do you see as the next logical step from the RTI report? What would you like to see?
>> Reed Gelzer:
There are stipulated, in the report’s recommendations, next steps. I don’t have those at my disposal immediately, but one step was to make sure that there is a successor to this report, in the sense that the report was never intended to be comprehensive that these were initial priorities. The other possibility, of course, would be to take the highest-level priorities and for other entities within the community, this Workgroup being certainly a very valid participant in such a discussion of essentially adding another level of endorsement to prioritization to functions such as auditing.
This and I work I’ve I’m in my second year of service to CCHIT, so I put my time where my mouth is. The CCHIT is a very important and critical function in the overall universe. But by necessity, it has had to start out at somewhat basic levels in some respects. And in some cases, because these detailed, in some ways very resource-hungry functions, such as audit functions, have been great sources of debate within the vendor community. And I believe very strongly that the vendor community deserves a level playing field. And so and some of these requirements do have very, very uneven impact on current industry, simply because some EMR design some EMR designs make these difficult to adhere to. I was I just had a message relayed to me that within the last few days by one of my compatriots who does field auditing, and she was in a discussion where a large institution with a nationally known EMR vendor was instructed by the technical support people to shut the audit functions off except for the last 3 days of the month, because it so slows down the system when the audit functions are on.
>> Carolyn Clancy:
Good grief!
>> Reed Gelzer:
So these are you know, these are real-time, incredibly critical issues to the infrastructure that we’re hoping to build the whole National Health Information Network on.
>> Robert Juhasz:
You know, this is Bob Juhasz again. One of the things that we’ve seen also is that the level of particularly with, like, authorship within a document, there are systems where, you know, you may have one user log on, and it’ll document that that person has been in the record, but it doesn’t really give the level of detail down to the character. And so, from a legal perspective, you know, there is something certain about you know, if you did something, you’d rather have the information that “Yes, I modified this here” versus, you know, just that you were in that record, and it’s up to speculation where it was modified by character, unless it’s you know, behind the you know, behind the scenes. So I think that’s something also, you know, that we need to look at is to get to a level where, you know, whoever is modifying it, there’s pretty clear transparency at some point, at some level to say, you know, “We know who wrote what where,” just as if, you know, you had a different pen on a paper.
>> Reed Gelzer:
Well and the good news is that, again, in my experience as a field auditor, I have seen systems that handle these kinds of functions with great skill and with great depth and, in some cases, I think, are even more attentive to fine detail than even I, as a compliance consultant, would think necessary.
On the other hand though, you’re absolutely right: The biggest fear that I have as a clinician, as far as the growing use of these systems, is that there are going to be physicians out there who do everything exactly right, who are very conscientious in their documentation, but when they are challenged, the underlying functions of the EMR will be challengeable. One of the most common functions I see is that there are a number of EMRs that do not retain the original version of an amended record.
>> Robert Juhasz:
Right. Yeah, I think that this is probably a critical place. If we’re going to effect that, I think that that is probably something that needs to be a requirement, because I don’t think it is a requirement. It is sort of one of those things that I think a lot of the vendors may have said in order to look the other way.
>> Reed Gelzer:
Well, it is on the CCHIT roadmap for 2005.
>> Robert Juhasz:
Okay.
>> Reed Gelzer:
And one of the things that I would also encourage folks who are interested in this area the current CCHIT materials that are available for public comment include quite a significant number of documentation compliance recommendations. And if folks are interested, I have for the HL7 Workgroup, I actually created a list of where exactly those are if people want to look specifically at those and comment on them.
>> Robert Juhasz:
And that would be how would we get to that, Reed?
>> Reed Gelzer:
I will send that to Alicia Bradford, and she can distribute it.
>> Robert Juhasz:
Thank you.
>> Reed Gelzer:
Sure thing. Thank you for your interest.
>> Lillee Gelinas:
Any other comments? (Pause) Okay. Hearing none, do we have public comments we need to bring in, Alicia?
>> Alicia Bradford:
Yeah, we do. Yeah.
>> Lillee Gelinas:
I don’t see that on the agenda here, but I think it is time to do that. I don’t know if there’s other topics we need to discuss.
>> Alicia Bradford:
No, I think we’ve got a lot ahead of us. I’ll be sending out the revised recommendations that we discussed today so we can shore those up and I can begin writing our letter to the AHIC. I have to get to the Group some language, that hopefully, the DOD and VA will provide me regarding 5.1 on the (inaudible) recommendations. I’ll send out the rest along with some discussion that was just added by Dr. Gelzer and Dr. Elkin.
>> Lillee Gelinas:
Okay.
>> Alicia Bradford:
And, oh, the use case prioritization. I’ll send out the four pieces of that that we discussed, give you guys the opportunity to perhaps elaborate on that so that we can get it down to the level of granularity needed to pass that on to the use case team and the AHIC. I’ll get those out before week’s end.
>> Lillee Gelinas:
Okay. Are we ready to bring in public comment?
>> Judy Sparrow:
Yeah, Chris, do we have anybody on the line?
>> Chris Weaver:
Thanks, Judy. If anybody is following along online and they wish to ask a question, you can go ahead and dial the number that you see on your screen now, and then once you’re connected, just press star-1 to ask a question. And for those who have been listening on the phone already, just press star-1 to get in the queue. And if you guys want to make up make your wrap-up comments, we’ll see if we get anybody.
>> Lillee Gelinas:
We have some important process steps yet. I know we’re going into the Christmas holidays, but Alicia and the ONC staff do a fabulous job keeping us on point. We need to make sure we have consensus around what our EHR Workgroup response and recommendation will be for our January 15, 2008 can’t believe it is 2008, but 2008 meeting. And our next Workgroup meeting will be January the 22, from 1 to 4 p.m. Eastern Standard Time. Judy, anything else I needed to reiterate from a process standpoint?
>> Judy Sparrow:
I think that’s it. Maybe just say to everybody, “Happy holidays.”
>> Lillee Gelinas:
Are there any anyone dialing in?
>> Judy Sparrow:
No. Nobody here.
>> Lillee Gelinas:
Well, I want to thank everyone for a robust EHR Workgroup meeting. You all just continue to amaze me with your dedication and your ability to keep us on track. I also want to commend you for not getting a seventh-inning stretch and for remaining so that we did have a quorum the entire time. John Perlin has had to leave us, but he was with us for the majority of the meeting. So thank you again. We’ll be together again on January the 22, 2008. And with that, happy holidays, everyone. Bye-bye.
(General farewells)