Wednesday, September 16, 2009

Health IT Standards Committee - 9/15

Below are the rough draft transcript and meeting materials from the September 15, 2009 HIT Standards Committee meeting.



Meaningful Use Quality Measure Grid Update

Meaningful Use Quality Measure Grid

Report from Privacy & Security Workgroup: Implementation Specifications Recommendations

Implementation Specifications Handout


Good morning everybody and welcome to the fifth meeting of the HIT Standards Committee and this is a pet goat and buys the committee meeting that it held in the public and broadcast over the Internet. -- this is public and the committee meeting is held to the public and broadcast over the Internet. We are asking for the first 15 minutes of public comment on the meeting today and then we will take 20 to 30 minutes for any comment that you might have in the room or on the telephone on the recommendations from the standards committee from last August 28th. And this is for the ARRA, American recovery and reinvestment Act which states that we need to provide for public input on the recommendations for the standards committee. With that I will ask each member in the room to introduce yourself, name and organization. And if you have any conflict on the agenda today.

I will begin with Sharon, no conflict.

Sharon Tirdel, CMS no conflicts.

Judy Murphy, or were health care, no conflict.

Stan from the U of Utah, I have no financial conflict but I am a closely aligned with HL7.

Doug Fritzma, Arizona State, no public.

Linda, and no complex but I'm on the board of directors for HL7.

Jamie Ferguson, Kaiser permanent seat, no complex.

Janet Corrigan, no no conflicts.

John Halamka, and to serve on the board and provide organization.

[ Indiscernible ] at Vanderbilt University and a board member.

I am at the Board of [ Indiscernible ], no conflicts.

David Macaulay, HIT vendor company.

Mark [ Indiscernible ] institutes and how the exchange, no conflicts.

If Ann Castro, no conflicts.

[ Indiscernible ] no complex and I am the chair of CCHIT.

Jodie Daniels, ONC.

Do we have any members on the telephone line?

With that I will turn it over to Dr. Perlin.

Thank you for your participation. As research I want to reflect on some progress today and to acknowledge that we received a great deal of public comments from both letters, a variety of topics and those are much appreciated. Please note that Judy Sparrow assures that does get all to the members for review so that the it become part of the record and consideration so we very much appreciate that imports could we think as all the members of the city and the work group for the continuing effort that has some ways been a very fast summer but a very long summer. A lot has been accomplished in to complete some of the prior work in the transmittal letter to David Blumenthal and his capacity of the national coordinator for health information technology. That memo described simply what we have approved at our last meeting that was discussed in this forum. So that is at the hands of a national coordinator or at their consideration consistent with the direction provided in statute by the HI-TECH portion of the American investment recovery and reinvestment Act.

We have some work around the three work groups, critical operations, quality, privacy and security and kudos to the work done thus far. But this should be a change in tenor from refining what we have done to getting to the second half of the meeting after the break to probing and understanding the way in which we can be most effective to provide guidance and input. One of the things that has been driven home to John Halamka and myself is the guidance is approved from real-world experience that we understand and appreciate and agree that this next phase which and getting information around implementation specifications really has to be based in that framework of what has worked in order to realize the ultimate objectives of this entire effort. Not simply getting health information technology but promulgating health information technology for more safer and higher value health care. So that will be our transition and look for the committee's input and look for testimony and I know we will speak in the future on implementation and the public comments on implementation. So for those who felt the heavy left of the past few months, that is absolutely true your efforts are greatly appreciated but this is not the beginning of the end. This is the end of the beginning. And a transition really to the consideration of how we can collectively be effective, rational, supportive, and motivated in terms of translating only standards but the opportunity of the moment to read and technologies in this improved health care -- to improve technology for this improved health care. And thank you especially to the broader community to provide comments as always adds Judy identified in this meeting and all of the forms that are part of and should be part of this liquor to process. thisdeliverative process.

What we have tried to do with the HIT Standards Committee is to get more and more granular so we have named some be standards. And it's just not sufficient for implementation 20 Kerr and interoperability to occur by adjusting a standard and we will use HL7 HL7, you need the guidance on how did you constrain it, how do you insure there is enough guidance that they can create interoperable living without a huge amount of expense or customization. So you will see each of the work groups diligently be as specific as possible. In the quality workgroup you see there are 29 measures to of which are privacy and security and tenor about meaningful use measures -- ten are about the meaningful use measures and 17 are to be EHR specific and Floyd will provide a taxonomy on how you get the exchange because when you get to implementation, think about this, you might have a provider and there might be a hell of information exchange in the community and there might be -- a health and permission exchange in the community and then there is CMS it's actually the data transfer that might occur and there may be different standards. It may be that they're identified line item data, and then it may be rolled up, numerators and denominators so we need to think about the complexity of that. Privacy and security, Dixie's personal anti virus software has failed so she will not be here today so you will hear some guidance around some of these privacy and security constructs. I think the important thing that was done from your eyes last meeting was to be extremely clear in putting together these matrixes water the -- what are the ands and what are the ors. And do any to do all 271 or just one so there are two matrices to be given and real clarity as to them "and"s and "or"s and we will have a conversation with genie and also Lee Jones -- Jamie and also Lee Jones to talk about how the guidance has been incorporated pat HITSP so you will see their what are the specific functions and what are the main standards and what the implementation guidance and 2011, 13, and 15 milestones. One of the challenges and David Blumenthal has this to us before coming cannot have an escalator that moves too fast because otherwise people will fall off. I think one of the challenges and so much work has been done by the committee to show you how to get on the escalator and have a glidepath that is smooth. Because this does allow some option malady but here is what you can do along the way to get there. And still in foster interoperability has been tough. So what you will see, although the main standards have not changed specifically, the guidance has doubled in volume. So there will be some enhanced clarity there. And then we will talk about the next steps. After our break I think it's important that we keep an agenda of going for that vessel lose momentum. As you said, this is not just, we are done, here is a nice stack and let us Polish we need to go beyond this. There are caps for 2013 and 2015, and we need to keep an eye and all of the things that we want to do next. We will be getting your input after the break so we can keep this very engaged committee. I look for a word to the day -- forward to the day and thanks to Dave for stepping in for Dixie and let's hope that she reboots successfully.

Everybody has a minute and let me think the Office of the National core Nader, any recommendations or changes?

We will consider them approved and we will move into the first and discuss it meaningful use call the measure is anchored update and we want to think Janet Corrigan and Floyd Eisenberg for this topic.

Thank you, John.

The critical quality work group has a relatively short report to did. Since the last meeting we have an opportunity to update the measure grid and we have a new one to Dave that reflects the theory is edits and issues that were raised at the last meeting and as well as a careful double checking back to the policy committee's measures and the Brady of clarifying issues. So what you find there we essentially, the policy committee have a recommended set of 29 measures can we go to the first slide, please? Twenty-nine measures. 17 of those -- thank you, Floyd 17 of those measures are performance measures that are NQF endorsed and need to be retooled for electronic health records and four EHR utilization like the percentage of patients with access to PHRs and there will have to be some form of [ Indiscernible ] during the 2011 period. And there are none to measures that are coming from the privacy and security work group. So that is pretty much where we are. We think that the measure retooling will begin around September 21st we are told once there has been more vetting of these measures we think that they will move the words so that they can be specified by the end of the year. The quality work group also spent some time on the issue that I think bubbled up in Jamie's work group that had to do with the submission workflow and essentially that is what John was referring to in his opening comments, how does the data, whether is individual patient level data or summary data, the data that is needed to calculate the quality measures and to report out on them, how does that flow from the EHR to whenever entities are involved in the process last meeting on their state, we began to kick this issue around a little bit. We made some substantial progress, Doug and Mark were both very in gauged in that discussion along with Floyd and myself and a small group was formed that included individuals from both the operations and quality work group to think through a remarked on how to keep up this issue for discussion on the standards committee today. Floyd will review a diagram that was developed by that group as a way to get everybody on the same page at this point.


Thank you, actually I want to give credit to some of the commentators for the HITSP -- the quality interoperability specifications with the comments indicated that it was not a very clear what the workflow is. And that is where this actually started. On Friday afternoon from 1:00 to 3:00 we went into our small group for this committee and modified, somewhat simplified it. So the object is thinking about what is the flow of data and what is the interoperable component. The first piece is the measure been sent to the EHR. We are not asking for it to be incorporated, but just sent so that they can be read and understood and you'll see each of these arrows, some are feint on the slide, but they are numbered. The next piece that there may be more than one EHR -- it takes that and does something with it to capture data and the clinical work flow. That is then used by a data collection assistant in order to pull together all of the data on each patient to make sure everything is there related to the measure. For that purpose, it might be a registry, it might be a health information exchange and might be in the external database, third-party vendor and there are other options. It also might be incorporated in the EHR. So you see a lot of different boxes on this drawing. The first transaction only occurs if that is a separate entity otherwise it's part of the EHR and, of course, with in the same system. The next step is that some organization specifically CMS and the Joint Commission certified processing entities to verify that data are correct and prepare the data for submission to the receiver. In that entity needs the patient level data to be able to verify it and make sure it's correct and ask questions about it and make sure that is in the proper format to then submit to the receiver. You'll see transaction fee is the data specific to a quality measure is sent to the processing entity and that is formated as required by the receiver. Sometimes at the patient's level, sometimes summary data meeting how many are in the denominator, new leader, what is your overall performance percentage wise and what is the average time? For instance, what is your average wait time in the emergency department. But the processing entity could also be part of the same as the data collection assistant. In that case there is no transaction between them, it's one organization. The same might be true of the EHR. The data collection system and the processing entity. So the reason for trying to draw the different types of architecture is is when multiple components are in one architecture, the transmission standards is not something as center stand it -- when they are separated is so we have to understand why the transport is for each of these different items within the flow. So what our intent was is to test this back to committee and have a discussion since this was a small group. It was not a full quality work report operations work group, but a small number from each. To look at the standard selected and health guide a road map.

Thank you, very much for those comments, John. I am sure we are thinking about in our frames of reference how this would actually work and I appreciate your discussion on that point. So let's, indeed, open this up for discussion and consideration of this table.

Any introductory comments?

Sure, I was here during Jamie's discussion on the detailed standards for some of these but one of the things I think is a challenge is that the standard to do these various detailed in summary submissions are various levels of maturity. So one challenge is that it's clear that 2011 requires quality data submission as part of speak up. Depending on how you architect that -- as part of meaningful use so depending on how you architect that, do you temporized and if you do that that would suggest that benders and hospitals have to go one direction for one year and that is not desirable so I think the committee and its workgroup should --making sure that path that gets us to Nirvana will be as smooth as possible. And thinking about the continuity of standards both for a coherent data flow and quality metrics as well as to project as John suggested a longitudinal projection that allows evolution of architecture.

The other is at a very practical level wondering and I would be interested in my colleagues thought, less the standards than the technology that would be used to do this. You meet the point in that box were one or more EHRs which may have inherent capacity to perform the data collection assistant or there may need to be another technology, perhaps it's a registry or a data analytical warehouse or something of that sort. I know the current state of play may have multiple approaches because registries are one sort of answer usually for the monolithic said question so it may require multiple cases. I appreciate the elegance and as he considered these bits, of the sort of questions that the broader environment will be asking. So my question for consideration of the group is not just with respect to standards but with the standards that we're offering for this process bowed for a process that allows not only the metrics but the longitudinal progression as John identified, 2011, 2013 and forward. Jim Walker has his card up.

Is this a measure reporting workflow? This only addresses measure reporting.

It does address measure reporting and the purpose for drawing it as it is it part of the measure Enterprise and I will go back to the use case that drove the HITSP effort, having implemented functions around the measure, it should have the ability to show some -- not necessarily real time but near real-time performance evaluations to improve performance as care is delivered rather than waiting until it was reported in the end of the year you know what to change. So in order to do that that is why it could be that the EHR assistant -- I purposely did not call the registry, but it's more of the function. It may be one thing to do that be back more internally. So that is why that is in the picture.

That is what I wanted to address then, this could be read if it were to be published, this could be read as saying it sort of optional whether measures are built into the EHR. I think we want to guard against that. We want to make sure that everybody understands that some point down the road, the efficient way to do this is to have a span that prompts the appropriate actions and the debt in that process and report sent automatically and that could be read that it does not matter if it's in their.

That is an excellent point and then realized that there are many colors in that slide. But EHR and Data Collection System, to manage just what you talked about whether its interoperability with a third-party that does it with the EHR or EHR alone should be performing that capability.

We might even try to draw this so it's a clear trajectory so that's why IC9 or Snowmed in 2011 and 2013 and 2015 so if it's clear to people at the end game is to give its in the EHR in almost all cases.

I think the one issue -- there are some instances where we have complex measure systems in some of the registries, guest registry that has an elaborate risk assessment and it's not clear to me that those will ever be in the EHR. It may be that you have to move the data to the registry and then the feedback of information for the comparative data that would do it but I agree with your point that there will probably be some exceptions but the vast majority of time you would like to see it pushed down the EHR.

As you can imagine this is architecturally neutral and we have our registry functions both inside the institution and in the community, they are for different purposes and, for example, I have this challenge, it's not an ownership relationship and we have to co-mingle data for quality measures for diabetics and I have a data warehouse and a community effort to unify such measures. I think the interesting question for the community is the does the EHR to a direct submission to CMS? I propose that as possible. Does it to a registry functions to specialty society and then that goes to CMS? All of the above. Unfortunately they may create a heterogeneous set of architectures that provide a lot of product development.

Let's go around the table -- let's start with Jodie and then to David and Jamie.

Thank you, John.

My question is tied to Jim's comment, how that plays out and what is your view as to what is maybe not necessarily the EHR but part of the technology and what do we actually need to make sure that we are developing the standards and certification for to build into the certified EHR technology consistent with your position on how this would work. In some of it as John Halamka is saying is community-based and that makes it a little bit more challenging for us. Just trying to get some clarity on what your thinking is.

It's a multifaceted answer because I don't think there is one answer to that. It's my own opinion and I cannot see our work group has provided so I will take responsibility but I know there has been talk about advance Quality certification. I don't know if that is where you are headed. But I would see a set of criteria to manage quality reporting and interplays that to the provider. I would see as required whether it is a certification of the EHR proper or the certification that the EHR can and does communicate with that third-party that is playing the role of the data collection assistant. I think because there are different architectures, as long there is appropriate communication back-and-forth, that could be acceptable. I am not pronouncing the decision but I think it makes sense to say that the requirements are that there is reporting in the ability to calculate performance, and feed it back to the provider. It's either certified with in this system EHR product or its certified that the EHR and that data collection assistant together can do that function. It would health solve some of the architecture issues, as long as it can happen. The processing entity is there mostly because of certification that it can assess the data and verify the data is correct and perhaps certification requires that such verification that the data are correct is also needed. Does that also health?

Yes, thank you.

This is the difference between Product and Project certification. Product certification could be a EHR uses standards to report data natively to a receiver. Whereas a project, it emits data to a registry which goes to the quality reporting entity and then some unstated to CMS and that in total is a sweet -- and then submit data to CMS and that is a total suite that needs to be done.

If I can make an analogy when I worked in the managed-care field, we were a penny we did buy NCQA for credential in but we could also contact with a certified credential organization and as long as we had a contract and there was proof of activity with them, that part of the certification or accreditation was not required as long as the credential certification organization was up-to-date and that might work here to allow out if the EHR is working directly with and it's certain criteria with a certified collection assistant. It seems like a good analogy.

Let's go to Wes.

Thanks, I was calling to try to lead the witness with questions but I think I will make a statement.

I am probably not the first person who distinguished between meaningful used and useful use in EHRs. I think we say at best they are tokens or points holding down a broader spread of values of what could happen. This is one of those areas that is particularly important. One of the things that Floyd did a wonderful job of explaining on the call on Friday was the various methods of responding to requirement to submit quality data and I don't mean standards here, I mean IT architecture. Depending on the organization size, there could be a small practice in Wyoming to an EPMC or something. And depending on what you were submitting it for, it's important to recognize that the standards that are being identified in this workflow are identified as this year near-term importance and to support the broad community in the different places that it's been measured and have a uniform way to submit to all of those different places and so forth. I think that this diagram is a great, if someone is A: not color blind so they can tell the difference between the green dots and the green border and so forth but for a lot of people I think it would be helpful to play out in the sequence of diagrams even if its tedious. The comments that Jim made, I think it's important that we not lose something that we said. He wants to see an expectation created through the work of meaningful use that EHRs are participating at the point of care in collecting quality data. Correct me if I misquoted him, but he said he had the expectation that EHRs would be smart enough to correct the right date there and now unspeaking for myself, including country indications for following -- I am speaking for myself including contraindications. That is a functional standard. It should be addressed in functional criteria. Or we need to make the statement that we are not doing functional standards, and we believe that our interoperability standards are sufficiently robust and it cannot happen any other way and meet the needs of the community. I think we should address that issue specifically. Given the variety of ways that meaningful use data is collected even for the measures that we picked out, do we know that an entity seeking incentive money can qualify on its own for meaningful use? In other words, I by the EHR and assemble a EHR and I do everything right. Can I fail to qualify on submission of meaningful use data because somebody else in the community did not do everything right? I think that is an important issue for us to evaluate. I can argued both ways, what a great way to pressure the community to do great work but I think we need to know the answer. And finally, I love number one on the chart. I think it's a great idea. Some people may interpret it to mean midyear, midweek, or midday we could change the criteria and send them out and that afternoon we would be collecting new data. I do not believe that is the intent. I just want to make sure that we are clear on what the intent is.

Just to clarify, if we come and before your comments Wes -- is this a function of the electronic health records or multiple technologies to address this project the outcome. So that is one question. Number two is the concept of architectural neutrality versus specificity and should also include not just architectural standards but also functional standards. And then to Jim Walker's comment that perhaps be the substrate for real time decision support which is the second component of this discussion. And the third is that Wes, to go with your comments, the taxonomy that John introduced, if one goes to project whether one builds or bias, I am assuming one buys components which are meant to interrupt a together and they feel what is done? How does one in sure that if one takes a composite --

You are shaking your head.

That was my question, it's a good question.

I thought that was your question.


Let's suppose that as an institution, as somebody who expects to get paid for an incentive payment, I do everything right. But my path to calculate the quality data involved, you know, not that John's institution would fail but there's another group of there that is dependent on John in order to make meaningful use. Are they at risk for John not doing his job in order to get their incentive money?

Let's hold some of the questions for discussion and I know you said you wanted to wait and and clarify some of this because I think this will is terrific and I don't want to lose the sense that this is a fabulous work. But there are some pieces that would require additional consideration, and the concept of singular EHR function versus composite function.

In regards to meaningful use what we're talking about is the ecosystem of many stakeholders. The wording from the policy committee has been that you must submit bio surveillance information so there has been an out, so in Massachusetts there is nobody to recede immunization data and there is nobody ready to recede and that is still space so I think -- So I think by 2011, CMS will have an infrastructure for print watch quality measures can be sent to them but ONC being a rational organization, CMS was not ready by July 2011 and we cannot hold you accountable. I think the policy side of things will address some of those concerns.

An institution can impute these measures without the health of a third-party. So it's not just the case of having a receiver, it's the case of having a collaborator.

Let's take that and see if we cannot work to some of the other questions on the table. And as we go through this discussion, if we are interesting new concepts we could bring that as another full point, Mark?

I think it builds on we are hearing some of the same issues and one is the question that Wes leases and Janet commented on -- raised and Janet commented on and there are organizations that will have the sophistication to create appropriate measures. And what I worry about is that you could have an organization with perfect skill set software and tools but incomplete data and the measures will be less helpful. So in example will be to look at a real patient and ask the question, how much of that patient lives in the particular EMR, and if you are Kaiser, wonderful, but with the VA there is shared care and it's not in the system to use reporting. So while we have the skills and capabilities, the data has to be there so I think that is where the data collection system becomes so important. So what are the holes in the Swiss cheese of the data and how big our day is very hard to know -- how big are they is very hard to know. And the other is, I want to go back to the question about building the measures into EMRs in this has been a hard question for me, I started out thinking that I could use the same measures. And Floyd is smiling because he has made the same error I assume in the past in the run-up the same rule -- and they are not the same role because there's a different level of specificity that you need prettify remind eighth position to produce this -- if I remind a position I better be right at a 95%. They are not idiots. We take the input along with other things and we used to think we could get away only two-thirds of the time but now its you better be right over 90% of the time. Quality measurement is a statistical process. It's not perfection. So I found myself arguing of both sides of my mouth, on the one hand I have this quality measure and don't worry about whether a couple of these patients are not yours, A does not matter at the end of the day because we are taking a thousand patients, we're not nitpicking -- but then I give them a reminder and then we start nitpicking. So those are two different tunings of the same knowledge base and it's important to keep us distinct. I love to see them keep coming closer together but it stuck to shove them in the same logic today and be successful at both enterprises.

I know okay for you want to wait and because you have done a lot of consideration for real time decision support.

I think the point made is a good one. We have seen in recent years those two worlds coming closer together and Diana sure whether it's a good or bad thing to be quite honest -- and I'm not sure whether its a good thing or bad thing to be quite honest and they have extraordinary long list of types of exclusions. That is an attempt to make the measure "perfect" for every single patient. It's very problematic. Because it's the Measurement reporting process very cumbersome and it should be statistical process, it does the have to be 100% rate for every patient. But there is a blurring there that takes place.

If I could comment, the reason that EHR box is multiple and not one is because in many situations as you describe there are many sources for data and it might be a PHR and other systems better behind that. It might be a nursing record that is not the EHR per se, that provides that information's for the assistant to get that data. I do think some of the data elements are identical but decision support is a different process is so agreed.

Can I also make one other comment? I think to me some things that could come out of our discussion today and over the coming year is perhaps a set of guidelines for the corrections that we want to go pick it may be a reasonable guideline maybe that a EHR should be able to provide the maximum amount of decision support possible to promote improvement on the measure that would imply that you clearly want that EHR in in -- any given setting to do as many cattle rations as possible to do real-time feedback. At the same time -- as many calculations as possible to do real-time feedback and at the same time we need to bridge what is a highly fragmented deliveries system in most communities where the data collection assistant is a mechanism. Because we do not have clinically integrated will define systems of care that can -- we just cannot have that many locations. So there you need the data collection system. That is a different situation than if you have a data collection assistant that has a value added because of specialized expertise. So for example, if in the case of some of the registries, that is a little bit different situation where they have a risk adjustments and specialized expertise for analysis of the data that you probably will not have within the delivery system. This seems to me that we will want to fashion guidelines that helps push the delivery system in the direction that we want it to go in terms of reforming itself which would lead us to encourage to whenever policy decisions are made that EHRs become an systems become all more encompassing of that patient.

Thanks, David MCallie are you still in for comment?

This question has been asked but I will ask it in a different way, you made clear that the speech might be rolled up tightly with the EHR -- that the data collection assistant might be rolled up tightly with the EHR. Is that decoupled on purpose or does have to be a separate entity for independent? Is there any reason why that is colored out as a separate color?

I did not make a conscious effort on tweeze of colors. To be honest this was my first foray into Office 2007 PowerPoint [ laughter ]

But, no, there are some EHR vendors who are certified as processing entities in date and the registry's in the processing entities are one. Which is why there is this green box. I understand the color issues that combine them all. So it's a different step in the workflow that is done and it can be a third-party, but it's different. It might be a QIO for instance, quality improvement organization. So it could be separate but it also could be one in this scene and I know that their ventures that have festered the quiche and to all of the above with that industry -- that to all of the above with in that registry.

Thank you, I will pursue the argument here and I guess it's pointed to the question that John had summarized as the project.

Product. And I really want to address the question in the sense on what should be looked at, and my compliments on the diagram. I think it was superb. I would like to address the question of whether these are monolithic and whether they are the unit of consideration. James made an excellent point that you need real time reminders or interaction with the commission if it fails to address the issue but I submit that many organizations, if not most are now in an informational area where you deal with multi systems. For example, Floyd, you referenced nursing systems or other systems and I would submit that there are scores and most institutions of departmental systems, laboratory systems, radiology systems, cardiovascular informations systems. All of which begin to the EHR. And while you to have this stacked little thing, compliments on the artistry for stacking, I would make it more explicit saying that in front of the EHR are stacks of systems and environments. And it really begs what is being certified from a provider perspective this has been an explicit question, I guess I would want to register strong consideration that rather than call it a project I would really use the system work in a sense. We could talk about vocabulary which I love to do, but the issue of a for a particular provider is clearly an interconnected system that is going to fulfill this kind of functionality, where the EHR may be a court or hub but only that. It could be that they choose to make an all in one EHR and that is perfectly acceptable but to put the disproportionate focus on that hub without taking to account the surrounding environment is something that we should avoid.

Thanks, let's go to Doug and then Jim Ferguson.

I just want to comment on the diagram again, again, nice use of the colors. When I take a look at this diagram I look at it any different way in the sense that you have identified the things that to the functions but what you are really talking about is that you collect data on a patient, you acre-feet data on a patient -- aggregate data on a patient and across the organization and in terms of trying to get to that. It could be that that EHR, it works as the aggregate coremac for that organization or it could be that its eighth Data Collection -- that its data collection for Quality reporting. I think the thing that is important is that you identified the entities but there's another diagram which is all of the things that you have to do pretty you could have easily said, you know, you collect the data, aggregate the data and report the data and define those to some sort of example entity, if you will. I think at that level of extraction to talk about the things that you have to do, it allows you to abstract out with the implementation details might be. It does not matter if it's an electronic medical record or assistant, the task that has to be done is to find a way that people can understand with or not from certification perspective for the or not the function has been met. It helps clarify the data that is needed to support that function. This is good and I like that we're moving toward the notion of understanding not only the what of standards but also the howl -- also the how of standards. I encourage you to think of that not only in terms of the technology entity but what is the function that they are trying to support.

Thanks, Doug. That is a great sick way. Jamie, what are your comments?

-- that is a great segue.

In the critical operations were group and the standards that we recommended, we started with the recommendations for the standards for submission of a calculated measures for CMS which are flows 4 and 5 on this diagram. It focused on the requirements of the meaningful user to perform that submission. So we started out saying that everything is with EHR, right? So that comprises all of the boxes at the top row, EHR, data collection, and processing. As our discussions and involved jointly with the quality work group, we understood that they are peaking out these functions that they have all want to get there -- that we have lumped together, so I would relabel because the data collection is a built-in function because it's filtering from all of the data, those that relate to the particular measure. And then the processing, quality report processing entity, we talked about as a aggregating and a calculating function because that this patient the data and putting it together and actually calculating whether there is a denominator for reporting purposes we talked about all of those things as being functions that are required of the Ming -- of the meaningful user. And if there are other reasons were these other flues are necessary, two and three in particular, in terms of the standards that we are talking about for Exchange of clinical data using CCD or CDA of HL7 2.5.1, all of that can be expressed in the exchange between these other entities using these other standards if, indeed, that is required. So I to think in terms of the standards that we are recommending we have recommended the primary standards for the recording which is flows 4 and 5, and all the other did exchange standards can be used when that is needed.

Thank you for those comments, I think there's a theme that is emerging which is really the segue that we are taking to this discussion and the implementation guidelines. I think there remains some questions about the sector of how this is created, the specificity for which parents must be given -- for guidance which must be given and Mark gave us great guidance to make sure that you're keeping up good recommendations etc., etc., but this is a segue of the conversation in terms of definition of certain standards. In the traffic core that has been done to some of the real world, messy and ambiguity around implementation and that is why we will have to wrestle with with the clinical quality work groups specifically to get the development. John, do you want to add anything?

These are all great comments and when I think of my own meaningful use, I have a home built EHR and commercial EHR rolled out to my positions and we have the New England Health Care exchange network which is eighth pelting entity -- a filtering entity and a calculator and aggregator. So we go from the EHRs and use them with HL7 2.5.1 to send them to the filtering entity which will then report them to CMS. And we hope with you call that a system or a project that will meet the criteria for achieving stimulus fund payment for all of the providers that to read this way. With that in mind, I would like to propose --

With that in mind, I would like to propose and I have heard a consensus around this general flow -- we have heard for need 40 to around implementation in the intermediate plead so let's take one last comment from Jim Walker and then move to consideration of accepting this with an understanding that we need to further work in terms of implementation specifications or guidance offered.


I want to emphasize what Wes said. Are a lot of organizations that our resource constrained and will not live in a wonderful resource environment like John's. We need to do everything we can to simplify the minimum of image in terms of project and product and systems to do good meaningful use and to report it.

Let me just a quick -- exclamation point on that and it dovetails directly into the guidance, and the metaphor of the escalator, that is difficult with increasing complexity. So we need to not only be able to bring along environment that may be less will be sourced in terms of technologies, funding, or IT capacity. And, in fact, let's move everybody forward.

Joe is not here and I wanted to ask about the Cinco practice in Wyoming -- the single practice in Wyoming and we are talking about quality in a major enterprise and it's important and complex. I just want to understand how a position in an independent practice will qualify for meaningful use with these measures. And if we think that the low end of this is the EHR handling everything inside of the green box with the green border, then I am fine with it. If we think that there's always a more complex relationship, then I'd like to understand it.

Great point, Jim, can you offer comment on that?

Just a question, what do we need to specify in terms of standards for standards? What do we need to recommend in terms of standards for 2011 and 2013 for this data submission to take place? I am unclear, will we be addressing that with Jamie a little bit later or if its QRDA or PQRI? What standards need to be specified?

I could address both comments and it specifically four and five. If you have the complex architecture, there is enough specificity to do it that way, but that is not specifically required review could have the big green box as the EHR and is using data flows four and five to get to CMS directly without any intermediaries.


I think we are in agreement on the directionality of this. I think the implementation work is required and that is a good segue. Do I hear a motion for adoption of this flow, acceptance of the report from the clinical quality work group with the probe be so that implementation -- with proviso that the implementation discussion continues.

Any further discussion on this?

Okay. Let's accept that with the clear understanding that we have additional work to do both in terms of implementation, and we hope, John, that your presentation and Jamie's crater clarification of four and five that you identified and the clarity and on the standards that Janet identified.

I would like to the knowledge that we have been joined by Dr. Shopra. Thank you for your continuing participation in this process.

Let us move now to Jamie Ferguson and John Halamka.

David MCallie will be playing the role of Dixie Baker today, she was stricken by flu, so thank you for agreeing to step in and leave this discussion.

And EDS David MCallie is coming to the podium let me think Floyd and Janet and the entire clinical quality work Group for your work.

Thank you, John and Jon. I may be a bit taller than Dixie but I do not think I can her shoes. I will do my best to -- fill her shoes, but I will do my best to cover what we have in our phone conversations. I will depend on John Halamka to bail me out if I get stuck. The first to acknowledge the work of the members of the committee in particular -- actually the slide has been corrected. I was going to see that Jon Mocehrke was listed twice and he gave us a lot of input on this perspective and we appreciate that input from him.

After the last meeting, this was identified here on this slide, in two points to reformat the recommendations accepted at that meeting and to identify specific recommendations on implementation guidance as John Halamka introduced earlier this morning. On the task of reformatting, basically we split -- We actually focused particularly on two questions that were asked one of which was to clarify the timing of some of the standards and the second was to clarify "and" vs. "or" option nullity and I will point out where we did that -- optionality and I will point out where we did that. There are two handouts and in number one we added specific suggestions for certification and this is sort of an English language suggestion on how this might proceed and we made a few changes which I will call to your attention in the second and then we created a New handout which is the guidelines for implementation which we were asked to specifically focus on. This is an attempt to move from our first output which was renamed the lower level standards to a higher more constrained implementation approach, again as John pointed out.

So handout number one, I will call out here a couple of things. The header for the standards column includes information from a number of sources. First and foremost a good Tory standards such as HIPAA and standards that have been developed by a specific standard development organizations which is HL7 and enforcement organizations such as IEG. In addition, we clarified in the columns that have the three date headers, 2011, 2013, and 2015. I will call your attention to the specific ones but we stumbled a couple times on the question that it's okay -- we want to make the point that it's okay to implement in later standard in an earlier and year. And the absence of listing a leader standard in an earlier year does not mean that it's prohibited, it means that it's optional and is only required in the standard that first shows up in the column. So just a clarification that tripped up a few of us. We have broken this up the best that we could, in a little bit along a fuzzy divide between those standards that are focused on the product itself and those that are focused on the infrastructure necessary to support a product in a setting. Of course, one man's and the structure is another man's product so this is an arbitrary distinction and should not be taken to mean too much other than to organize a thinking approach. Let me focus on a few of the notable changes that we have either clarified or differed slightly from our previously accepted recommendations first we decided that due to the ARRA requirements for accounting disclosures it made sense to specify that ATNA standards be required for 2011 that had been left off in previous versions. We did not think this proposed a particular carrier and is a requirement to be capturing this kind of information so ATNA is the preferred way to do that. Second we had a pointed out to us that federal systems will cease to allow the use of Kerberos after I believe 2012 so we decided to change our diagram or our matrix to show this as an allowed choice only in the 2011 column and after 2013, SUA would be the preferred standard. The third line, this was an "and" and "or" question. We were not specific what happened in 2011. So we made slight clarification that 2011 should be governed by the advice to implementation Guide of the Service coordination, HITSP 112. And then in 2013 onward, the choice for documented change with respect to appropriate security in such would be chosen from any of these listed profiles which is essentially XDS, XDR, XCA, and XDM. We could go into questions about what that means but depending on the specific quality, one or more of these would be appropriate and acceptable. You are not required to use them all, only if it requires the appropriate standards so XDM would be media base and XDR would be point to point and XDS would be registry based or depository based transfers. Finally the last line we added a small clarification that an architecture that uses a restful technology model instead of aid SOAP model would need a profile because since last fall approaches are less constraining then soap approaches one could not use it without constraints to specify how one handles the transfer of a security assertion as to who's the user is. Because rest by itself does not specify that and they have to constrain how to do that. It's somewhat of an irrelevant point because there are not a lot of -- and they would need to address to the constraints the questions of integrity of data.

The implementation guidance selection. We spent a good deal of time with John Halamka and came up with this list to prioritize our thinking on where to get implementation guidance from. It should not be unsurprising list to any of us because I think we are drawing from a similar approach. We started with anything that is the output of the HITSP tiger teams HITSP use PS cased constructs -- use-case based constructs was Number two on our list and we look to IHE profiles for Number three and anything else that we have found from an accredited standards to college organization is Number four. The handout, we will not go through the details of it but just to highlight the columns here we put the implementation guidance for 2011 in a separate column and then lumped 2013 and 2015 and optional 2011 in the second column so there's more specificity in the 2011 column and a little bit more ambiguity in the 2013, 2015. And we called out specific notes will we felt the tension needed to be drawn to particular gaps that existed that would be after the 20 luck -- the 20 luck and timing.

He are the specific guidelines that we pulled in. This headline is a little bit misleading. This says elected guidelines in these other ones that we selected rather than a selected list of guidelines. This led pulls those out and I will not read them all off to you. You are familiar with these, HITSP components in specific transactions. And then finally there were a number of IHE specific integration profiles that were called out and given a place on our spreadsheet.

And that I believe covers the information that Dixie wanted me to present. John, you have anything that you wanted me to add to that Summary View?

Dixie is passionate about insuring confidentiality of patient data is, and the balance of these cities, cost, and bulletproof protection. So what the work group is trying to do is to give this rational timeframe that gets us to increasingly constrained security that recognizes that Joe and the fellow practitioner will have to do this and there will be various demand so for example, the VA may have much stricter authority but we will be as strong as our weakest link so you have to set these minimums in 2011 and 2013 to make sure that these are protected. So what we tried to do is to give the standards and it's clear what will be used and with the ands and ors, so you except both SOAP and [ Indiscernible ] in various mechanisms that might be used in DoD and VA and they try to incorporate as granular guidance as possible. So they want to start with the most specific that we could, which is the service collaboration and the kind of things that the HITSP tiger teams have done. So I look for to your discussion that you feel that we have done a reasonable job of balancing these patients' privacy rights to protect confidentiality but also make it implemented over years, testable because we have been enough specificity on the information side.

I would add additional tag to what John just said. We had a number of discussions where we were on the cusp of trying to decide whether the debate was a policy question or a standard question. Obviously when it comes to privacy and security many of the issues are policy questions so we identified a number of those which will need to get flushed out as we progress further especially around HIE and information sharing in the community but the standards are neutral and flexible enough to support the policy issues that we envisioned might come up. So for example one of the debates that we had as a policy question was what John just mentioned. If you have a network of systems that are connected to each other and they have differing levels of user authentication certainty, one might require two factor with a hardware token. In those systems communicate with each other and all of the systems have to rise to the most stringent requirement to the most stringent member or to the lower themselves to the lowest common denominator? The answer is not in this spreadsheet but the technology tools but that need to list would make it possible for the system to understand what the other systems level of that the indication was. So they could make that decision based on their own policy. So we try to make sure that the systems -- make sure the low-level technologies are capable of expressing sufficient information to make rational decisions once the policies have been clarified. And a second

-- consider consumer consent after 2011, we get into fairly murky territory in terms of the way the current standards are written. Again, the technology is all there. The vocabulary necessary to express with sufficient cranberry tea at consumers' wishes to control the spread of their health information are not yet well developed. At that work needs to continue with the standards committee. John, would you agree with that?

Well, absolutely. In fact, when we talk about our next steps as a committee, I think that focusing on granular patient consent is very important to do. On the content management portion of the spreadsheet, we used it profile called the basic patient privacy consent. It doesn't get to the specificity of, I want to share this data with this individual in this circumstance. There are other are profiles of the standards. There really isn't a widely deployed in the church set of constructs that gets to that level of consent management. I think it will be dashed for the policy question and technology question for us all to work on for 2013 and 2015.

I think this is an opportunity to note that appreciate that clarification. David, you are lobbying for that. The committee takes great not of all the input -- takes a great note of all of the input.

Should we do questions? I am finished with the summary.


Go easy on me.

[ laughter ]

I just have a question about [ indiscernible ]. First of all, I want to say that I am a big fan. I think it is done more to create an effective enforcement of privacy concerns than anything else so far in health care. Is that noise in my head? Okay. Cynic is in our head.

[ laughter ]

The comment I have is I don't think it is necessary or sufficient to meet the disclosure requirements, because they are about disclosure that can come out of some existing. It is conditioned on whether the patient has paid or not for that service. I do want to understand what we are committing the industry to for this standard. Are we committing to use that at the interface with another entity or internally within their systems?

It is my understanding, and I will preface by saying I am certainly not an area of deep expertise, that we are focusing here on the certifiable EMR within itself, its own audit trail. It is not the transfer of information --

Okay. The requirement would be that EHR be able to produce a lot of?


Want it provides a couple of things. There is a standard audit format so you could imagining although, yes, absolutely audit team is to have a standard on a format that could be disclosed. I could imagine as disclosures are made between organizations but the same audit format is used.

So in a sense we come up with one standard for generic auditing internal and external for the organization.

I think it is a great idea for both.

Yeah, I think a unified standard makes the most sense. If you think internal disclosure, a practicing clinicians in front of the EMR and he turns around and causes that same data to be published to the HIE, that is just another instance of disclosure. In that case, you don't know will is going to see it in the HIE. The cranberry is different. In the first case, you knew with a provider was. In the second case, it is a system. You just let that data flow out to a consumer.

Yeah. I think in the ideal world, every system would produce that log, coz then we will have all kinds of great tools for analyzing and finding out. But I am concerned that we are not driving health-care institutions to adopt that on every system that they have internally. If we decide at the HIE level to do it there, that is fine, but I think that would be a staff a have to consider -- step we have to consider carefully.

Maybe, Jodie, you can help there. Some of that will depend on how the era statutes are depending on regulations.

Yeah. And we are an ongoing conversations about this, so we are still working through the details. The Office of Civil Rights has required to, with the [ indiscernible ], and they are looking to us to provide standards to assist them in meeting our requirements that they would come out with under flood level HIPAA. We are trying to figure out how to line is up and make sure there is enough built into the technology to support the regulatory requirements. We are still talking that through. My understanding, and I am not a standards expert, is that the atmosphere is fairly flexible and has a lot of what may be necessary for the accounting disclosures requirement, but we are still working through the details.

I just think it is wonderful, two different agencies trying to coordinate the user standards. I think that is great.

[ laughter ]

The work group did debate, and I think we may have mentioned this at a previous meeting. Do you require encryption of all data flows inside the four walls of an organization? This is to the comment about burden. The answer is, no, unless you could imagine there is an open a wireless network were the data is going to float in the clear. Of course, that would be a breach. That would be bad. We are not going to require double encryption. That is incorrect everything that flows over your networks, and then implement the WPA enterprise. The other debate we had is encrypting data at rest. Should we require every data center to have databases sitting at rest and corrected? Again, the economic burden of that would be very significant. However, if you are going to put patient identified data on the thumb drive or a mobile device that could be lost or stolen, should patient identified data being corrected on a mobile device at rest? Our committee recommendation is that, yes, that should be. And it is consistent with the provisions and the state lost. Now, a question asked about standard for breach notification. David, I don't recall that we have had any discussions -- let's just imagine a horrible thing occurs. An employee discloses 501 medical records from [ indiscernible ] Medical Center and I have to notify the "New York Times", or the Boston Globe's. Is there a standard mechanism for me to do data breach identification? I think the answer is maybe this is a policy as opposed to an electronics standard.

I will further refine my question to John. There is a set of data breach activities taken place beyond health care around how the nation addresses issues of security and privacy. But from a technical standards perspective, the question I just sparked in my mind listening to the dialogue was, I would not call it the RSS feed for Data bridge problems, but somehow a we have a comment mechanism to harmonize section data, in theory it should give us a continuous feedback loop about the nature of kinds of things that led to the breach in the first place. Part of the anxiety about the standards discussions about all of the things we are doing -- and that is why I wanted to serve on the five committee. We did not have a clear perspective as who is the actor in the states. I do not have a good perspective of the market share of actors. It was a foreign countries hacking into hospitals for data? Is a rogue employees? Is it a mistake in code that open up something? To the extent that we have a little bit more of a consistent method by which we could report out when things go wrong, we might be able to understand root cause. Therefore, as a committee, get at the issue as to how we go after this. I don't know if there is a role for us to standardize the basic elements of what you want to know for the purposes of feedback to us. The nature of the breach, the scope of the breach, the source of the thing that led to the breach. I am just brainstorming without pollution, but I am mindful as we get into the mechanics of some of these pieces that at least we raise the question.

You could imagine developing [ indiscernible ] that says, what was the development of the breach? I am aware of the standard today that does that. If any discussion about the mechanism by which a breach would be, and notification would occur.

We have been working closely on the breach notification rules. There is a requirement in certain cases of a breach that those be reported to HHS. We are actually working with NCR right now to think through what are the things they would want to have reported to them in those instances. If this committee had some input to that, that would be welcome. It is something we are looking at, because we are hoping we can get some intelligence from those reports that will help us to provide something back to the field on breaches, understand what the vulnerabilities and risks are, etcetera.

I think it is very interesting.

That is my job. I try to come up with interesting ideas.

[ laughter ]

Part of the state to sort of implementation, but in this instance, it also transcends the concept of breach notification as a responsibility to potentially agreed parties to the idea or notion of burning system that helps to amplify the signal to harden more effectively. I really think that is very well taken point for the further consideration.

Any other discussion on this? We are in consensus on the recommendations I here. Do I hear a motion for such? Okay. A motion to a except the recommendation privacy and security work group. Any objections? First question --

I'm sorry.

A very much. My concerns as will adopt these recommendations.Thank you to all the members of the work group add additional considerations for continuing work as well as the segue to implementation guidance. Now, John, we will turn to you and Cheney and move to the granule rarity of discussion in the critical operations in data standards -- clinical operations data standards.

Two presentations for the committee. We will start with a presentation by Jamie looking at how we have refined the implementation guidance, changes that have been made from the clinical operations. Event a presentation from HHS. When you look at the last decade, ten years ago we had the [ indiscernible ] standards that were simply name based standards. If we were satisfied with name based standards, we could have declared ourselves done a decade ago. The work that has really being done is get into this granular level. You will hear from Jamie about that and from Jamie about how the last four years have taught us to these constructs that are much more specific so that vendors could actually follow the guidance. HIPAA thanks, John. This will be truly a brief update, I think.

Our revisions are truly minor. We have details, matrix in a handout which was not distributed at the last meeting. This is an update to the detailed matrix of our recommendations that was handed out the meeting before last. As you'll see, there are actually no changes to the recommended standards. Although, we have made one change to the recommended timing that I will discuss, and we have clarified and changed our language describing our recommendations based on Tom's of input of that discussion -- terms of input. We have clarified that the use of a single standard for each purpose is our basic guideline with a couple of noted exceptions, one for lab results reporting and one for quality measure reporting period we have also clarified that the recommended use of local or proprietary codes has an exclusion where specific codes are required for purposes of the quality reports. We also clarified where we mentioned the use of legacy, the HL7 implementation guidelines, but that is really intended to be only for pre-existing implementations of lab results reporting or other exclusions that are going to be specified in guidance. And then we also clarified in terms of the version being version specific, we are saying that you should just use the most recent approved version of implementation guidance. That is so that we are not tied to a specific version of implementation guidance. In fact, when the rule or guidance is positioned by 0and C, it is simply the most current version that we are referring to.

To give an example, let's imagine [ indiscernible ] as a vocabulary for medications. Many of our organizations use proprietary code sets. We might buy products from first to databank, gold standards, et cetera. So, are we going to require specifically in 2011 that everyone gets rid of proprietary codes and moves to our tax and Norm? That would probably not be achievable. It would be expensive and unrealistic. Are we going to submit to Floyd and Janice, here are the medications and the quality measures in the first data bank format. That would be unreasonable to have the standards and metrics organizations provide full Cavaliers and codes that for every proprietary code set that might indicate that the medication and was administered appropriately. For those limited measures, there are 17 measures defined. For just those 17 data submissions, should a medication [ indiscernible ] this allows several assembly straight for a transition from proprietary today, and in fact I may state proprietary for a long time. The burden on 2011 is very, very limited in regards using an advanced a cavalier.

Just to those particular codes required by each measure. Okay. Just to summarize our recent work. This goes back to this will change in timing that I referred to earlier. Previously, we recommended the PQRI registry access mouth with peaky you are a to be used soon after. Now the change is timing and that we are recommending either for 2011 and 42013 where the long-term direction to move to QRDA when possible. This is another example of saying that you can use either one, but there is a direct statement toward the QRDA standard terms of this is, again, one of the very few cases where we are recommending two to different standards as an option for exactly the same purpose. In general, we are trying to avoid that, but this is one of those exceptions. Also, just referring back to the previous discussion of a quality work group. In the diagram discussed, this is four steps or transactions four and five on the diagram. That is where this applies. We also discussed implementation guidance in general, and this is something that Lee Jones will come up shortly and discuss in much more detail. I just wanted to note that all our implementation guidance uses the HITSP specifications, except for the PQRI Quality reporting options. In terms of some of our next steps, we have talked about establishing a series of meetings. A we have talked about some of the vocabulary gaps that we have talked about. One of the top priorities there would be for an compendium of order zero labs. We also talked about the need to discuss the long-term standards recommendations for procedures. And a number of other issues. We are looking forward to those meetings in the future. We also want to discuss enhanced guidance and the implementation specifications for Quality reporting and for things like patient access to electronic records. We also in the future want to talk about more standard maintenance prophecies and look forward to moving on to the 2013 measures and starting to work on the next round of work.

To the comment Janet made at the end of her presentation, here is our challenge. We do not want to confuse the marketplace by offering two approaches. [ indiscernible ] is getting to one standard for every purpose, but what do you do based on the maturity guidance we have talked about for several meetings? Is a widely implemented and understood? Is the kind of implemented? Is it not implemented? Is a gleam in the I? QRDA is basically a basic standard, requiring it as a single standard is not realistic in 2011. But yet for meaningful use, we need to get that four and five, that transaction reporting to see asked done. An interesting towers, do you choose and force an amateur standard because it is required for meaningful use, or do you allow a migration paths that says we will use something well understood and the than direction remove us to QRDA?

There are approximately 70 registry's currently operating, so using the PQRI specification for measures with CMS, which is accepting them. So, that is working. QRDA, not so much.

Also, just as a comment on vocabulary. You notice that as part of meaningful use, exchanging lab results is required for 2011, but ordering is not required until 2013. From the HITSP perspective, you will have that by January. The order vocabulary is still in process, but we will be able to get the 2013 standards. When you see all this, this falls into the category of important tax steps to make all this work for 2013 and 2015.


Comments or discussions are welcome.

One question, Jamie, we discussed this a little last time. I am just curious to know what your reasoning was all around the script ten decision, which is not yet implemented anywhere.

We did come back and discuss that again in the work group. I don't know, John, if you want to comment on some of the other discussions on that, but we did feel that it was feasible to move to script tenth by 2011, recognizing that if there are not changes in regulations for an M.B.A. and other things, that may inhibit it. We did feel it was feasible.

So, a technology question and policy question. Kevin Hutchinson said, I just want to let you know that actually all the guidance, all the implementation that is necessary, all the testing before the ten. Accept actually will be complete by the end of 2009. Hitting 2011 shouldn't be a problem. I know that [ indiscernible ] has a working with federal agencies to make sure that that 10.x policy [ indiscernible ]. Yes, we would run into the problem that if the policy was not changed in the technology was available, that is an issue. Let's hope the policy changes and therefore will converge on 10.x.

I would just like to follow that up. C M S is putting into HHS clearance an interim final rule that would adopt ten. Six as a voluntary additional standard to eight. 13 our backwards compatibility process.

Correct. Very, very helpful contribution.

Can you talk a little bit about your thinking [ inaudible ] in standardization in that perspective?

Well, that is the lob order compendium. We recognize that that is something that is likely to be needed for lab orders standardization. That is not a 2011 requirement, so we are just listing Labatt I guess on our list -- just listing got on our I guess wish list.

The committee is going to be looking at that soon?

We are anticipating setting up a series of a variety of a vocabulary topics. We will invite members of the other interested parties from other groups as well.

What Jamie has requested is the notion of getting world experts on vocabulary to focus on these issues of black border compendium, snowmen, little different than the focus of clinical operations. It would probably require some different external experts, so we hope to move forward with that action. I think David have another comment.

A technical detail question reflecting some confusion in our company and interpreting HITSP capability 124 moving documents. The question would be is, in 2011, the simplest possible scenario for interchange of structured summary document from provider to patient, would that require access D M or XDR formatting transfer of the document? Does capability 120 allow for just a six year channel and unstructured, if you would, in terms of directory formats?


That is great detail.

Absolutely. To explain this capability, 119 is the unstructured document 3120 is structured documents to CCC, CT eight construct. -- CDA constructs. They both use XDR, which is basically a push of information. Accidie as is pushing it to a registry. In terms of the simplest, a push using XDR to a personal health record or giving the patient in an extra mouth form on a DVD, CD, or Thumb Drive using XDM is the current guidance.

That is however the capability. I just want to make sure.

That is all done not our intention -- that is our intent. And just to clarify, 120 is unstructured and 119 is structured.

It is backwards.

Just a quick question. I am not familiar with QRDA. Could you just give a little flavor of what that means for the average Joe?

It is a CDA like construct. It is a standard for quality measure exchange.

A relative to where we are, is it easier, simpler, more widely adopted?

It is generally not used yet. It is a newer standard.

Can I just make a comment? Behalfs actually passed it actually has been tested. I believe [ indiscernible ] has done some testing in New York city. Some other areas have. It allows reporting of multiplications data or symbol. It is also using CDA similar to C CD and other similar constructs. It is consistent management of data, which is the reason for looking at that.

For multiple patients, that is the new dimension? PQRI doesn't allow that?

They do also.

What is the innovation of this over the PQRI?

I would say it puts quality reporting on the same basic technical architecture as all the other document based exchanges that we are recommending.

Thank you.

The idea is we can use crossed clinical care and clinical quality, individual patient and aggregate data, the same the framework for data transmission and based on a CDA document. It is when to be simpler to implement.

Thank you.

Good questions. Okay. Well, you want to move on adopting this, then we will hear from Lee.

I think we have exhausted questions. I think we are in a consensus. A motion? Any objections? Terrific. We will take this forward and move on into the next section of the agenda. Actually, I will ask John to introduce Lee.

He is a program manager of HITSP. And this has been a for your process of harmonizing standards, hearing from multiple stakes holders. He will give you a sense of how one goes from the base standards to implementation guidance. Why does one want to do it? To today's game is specificity and limitation guidance. He will give you a sense of the process, where it has gone and where it is going.

Good morning. I appreciate the opportunity to speak with you today. I, certainly, want to applaud you for your great work. I was very encouraged to see the recommendations as a camp for word and HITSP was included in a number of them, so I am particularly pleased. I was asked to come here and talk a little bit about implementation guidance. As much as you are now at the point where you kind of have to wrestle with how to select the proper appropriate implementation guidance. I want to give you some additional bolstering of your free market as well as try to give you an insight as to what you actually already get when you selected and make recommendations are around HITSP constructs. A hopefully, I'll be at will to accomplish those two goals. The first question really is, what we mean when we talk about implementation guidance? I do think that there is a standard definition that one could find that says that this is implementation guidance and that is not. In fact, you will see a number of different organizations publishing things that they may call implementation guidance or served as such even though you may not called at that. It may be as simple as the definition of a standard itself, which to some extent is instructed on how one would use the standard and gives guidance on implementation. Or it could be as complex as a reference implementation where someone builds an actual system and says, this is how something is done. In fact, you could take that to the extreme and someone could say come it is not just a reference implementation. It is the implementation. Everyone uses the same technology. An extreme case of the items on implementing a solution. In addition to that, the implementation guidance can span different kinds of areas of implementation, so we can talk about standardizing around Davidoff, standardizing around technology platforms, architecture, the content of the data, how that content is included. When we talk about implementation guidance, we are really talking about a broad set of things that you have to consider in order to be able to construct some one on how to accomplish implementation. As a working definition, I would like to put forward implementation guidance should really be linked closely to testing or how you are going to recognize that an implementation has done -- has been done successfully. That is what we try to think about in HITSP as we move to produce implementation guidance, and that is what I am going to talk a little bit about today. So, as is often the case with many things, upon subsequent thought, I have greatly improved this slide. It is on my computer, greatly improved, but it is not greatly improved on your piece of paper and on the screen here. I am going to try to take you through a little bit of a visualization. So, I have some access here, one is the degree of implementation specificity verse is the degree of automatic interoperable Libby without solution. Mean that people can implement something with a set of instructions and about collaborating with one another can arrive at a solution. You could imagine that this could be a scatter plot of a number different points that are unrelated, each one corresponding to a specific standard. It could be instructed to some particular degree, and it could insurer interoperability to some degree. One is how you define interoperability, and another is the context in which you would measure these things. Our challenge at HITSP is if we have that scatter plot of different standards not have all kinds of implementation guidance and have different degrees and levels of specificity around that, how is it that a plot a course through all those different points and arrive at some consistent way to describe how you implement something and to insure particular level of interoperability? This blueline the you see here is really what I would call the implementation guidance functions, if you will, or curve around harmonize standards. Not anyone standard, but a set of standards that have all competing dimensions and requirements. We try to take those and make it consistent way to implement interoperability with them. With that kind of a curve, you could argue that the curve may not look like this in practice. It could be -- the slope could be different. It could be tilted one way or another, moved up or down one way or another. What makes your choices around implementation guidance not arbitrary? I think in order for it not to be arbitrary, this is a little bit of art versus Finance, we have to have some sort of dampening function that controls us and put a check on how we select that guidance. So, this red line is trying to depict plodding on the same set of taxes, but introducing a degree of freedom that you have in implementation. The fact that the more implementation specificity that you provide, the less -- of the fewer degrees of freedom that the implement has. If you are very descriptive about how something could be done, then the implement your does not have a lot of choice on how to do it. Conversely, if they have a lot of choices, they may do it in different ways and you did not have a lot of specificity. I would call this sort of our proprietary innovation and unconstrained on interoperable witty. And I say proprietary innovation, because there is often this argument that is made that the more descriptive we are, we are limiting innovation. It really what we are limiting innovation around is be interoperable portions of the application. We do not want those portions to be proprietary to the extent that we can avoid it, but we want them to be, and and available to all -- want them to be, and available to all. And when we look at these to curbs next to each other, the thing that keeps HITSP in check is that we invite reporters to the table. They keep us from going up too far. Especially, when you start to talk about constraining hardware architecture or software architecture or particular technologies or platforms. That is sort of a natural tension that we have. What will wind up really been specific or more descriptive about our constraints around data and information, the flow of that, the representation of it, format and content, and even the rules of from the information flow. However, there are some boundaries that sort of put in natural check on what we would specify as interoperability -- I mean as implementation guidance. So, with that in mind, the other portion that really is important when we consider implementation guidance is how we are going to recognize that one has been confined to that implementation guidance. So, testing is really a very important part of trying to select what implementation guidance you want to have. So, you could imagine if we have this curve around implementation guidance for harmonized standards that if you pick a particular goal for interoperability or level like you have done for multiple years, 2011, 13, 15, et cetera, then there is some particular level of implementation guidance you want to select and order to ensure that that goal can be recognized. So, you don't want to say that people have to do things by 2011 and you don't have a way to tell if they have done it. The only way to really tell or make sure that you test for how they have done is to consider the instructions you have given them or the implementation guidance you have given them. When you do that, that test or that line that you draw around interoperability -- no matter how specific you are all around your interoperability, people don't tend to go beyond a particular level, because that is where the test is set. You have to sort of recognize that there is a natural relationship between the guidance you have and it's utility and the testing or criteria you use for testing to recognize it. Similarly, if you try to move further and, along, you can describe the different levels -- moved to further along, you can describe different levels. We have worked to break closely with [ indiscernible ] and with IHE to ensure that we have a channel to actual implementation and feedback in testing and a body of work and built up over the four years that allows us to a couple of our interoperability specifications with specific tests used by implementing. That continues today and has been evolving over a number of years. So, HITSP really harmonizes standards by identifying the appropriate set of standards for a given context or problem that is trying to be salted, and then producing guidance of around how to implement those standards in concert. There are implementation guides that may be offered by a given Standards Development Organization of around their standard, but when we start to enjoy in multiple standards from multiple organizations, we really have to be prescriptive about how those things work together. Our goal is to produce a system of documentation to that facility is different implementing people to come to a new solution. We try to make sure that our guidance is testable. When you have selected the HITSP constructs and capabilities, et cetera, you really are accepting and Dominic selecting more than just a set of standards -- selecting more than just a set of standards. So, in this next slide just tries to enumerate some of the components of what HITSP specifies. At the bottom, we have the universe of all health care IT Standards. HITSP does not create standards or try to change standards. It tries to reference the standards across different organizations. Our documentation and, by design, is in Iraq and preferential. Some people levee that as a criticism. However, we'll -- we want to make use of the good work that they do. In our body of work, which is represented here by these circles, we do a poll for word excerpt from the standards so we can give enough of a concept on what standards are being used and how they should be used so it is understandable. In addition to pulling for were those and having a balance between that and referencing the external standards, we have constraints and now you sets that we Jews around the content of the standards -- that we use of on the content of the standards. And we really have done a lot of work to implement a framework or semantic framework around what interoperability really means. That is really important, because we could use the same words, and we can use the same phrases, but we could be talking about different things. We try to be very explicit about what we mean when we talk about the exchange of information and who is participating and how those factors are described and what they are actually doing relative to one another. That is part of the system of documentation we have. More recently, as you have all made use of in your work, we have specified capabilities and service collaborations' which have now tried to provide the right level of extraction so you can tie all these technical details to the real world problem that is not described in a technical way. And when you give us a use case or you give us some other problem statement, we want to be ample to select from a library of capabilities that are close to that level of abstraction, yet they carry with it all the technical details that you probably don't want to think about at this level but are necessary for us to achieve real interoperability. Our next slide, I provided this chart, which just tries to give you a sense of what HITSP does versus what you may get from a given a choice of a base standard, like just selecting HL7. You may certainly get the data format or some prescriptions around a transport, the basic transport of how you move data around and the definition of what you are talking about, which are all critical for implementation. If that SDO provides implementation itself, you get those things as well as may be some additional instructions around coding. You may also get some option around different context that they know that the standard is used. What HITSP does is try to pull for all of those the laws that are offered by the standards development organizations so you get all the guidance they get themselves. When we put together that system of documentation I already described, you get a number of other things such as specific value set within a large vocabulary. These are the items that are of particular concern for a context. You also get universally defined [ indiscernible ]. We want to be able to say, what I am talking about it performing is the same actor as I am talking about in this other kind of transaction. When an implementor, they can't see where they play across the spectrum. We also brought along any required standards that goes bond will anyone -- goes beyond anyone SDO. Anything but the problem based dictates, we tried to account for a post. And then we orchestrate how those transactions actually will happen so that the implement jurors -- the implementors know the rules of how to put these deals together. In conclusion, it is really important in our view that you tighter definition and of how to read the press compliance or testing and certification to the implementation guidance that you select. So, we don't want to just grab finger is that have a particular label of. We want to be thoughtful about how it connects to your testing. The HITSP implementation guidance, as I have said, tries to unify multiple sources of implementation guidance, whether it is from SDO or IHE. To the extent that HIT work is referenced in your recommendations, you get the benefit of all of those different things. You have some attendant implementation guidance built in, so that was really my presentation. Hopefully, it has added a little clarity on your thinking around implementation guidance and also what HITSP provides.

I want to thank you for a spectacular presentation. You have the heart of an economist. I don't know if you are an economist. What real scrutiny is that it's sort of has a parable of history. Can you imagine ship Building Co before the 20th-century? Every ship has manufactured some bold independently. You quickly see where this is going. All the sudden you of multiple [ indiscernible ] that improve ship building. On the other hand, it required a market to accelerate it even further. To that was created in part by the First World War were ships were unnecessary. There are two pieces. Without that degree of specificity, your craft of tighter specificity and the ability to have pieces would help collaboration, which really was achieved in not metaphor. On the other hand, to really pull forward and accelerate and amplify -- here we have both opportunities. There is the opportunity for us to converge as we are doing, but what I think is so spectacular about the moment is that the market has accelerated dramatically by the incentives that the White House has brought to bear and have been approved to support this process. So, I think that is a terrific presentation in terms of context settings, because it also addresses the the attention and that we have each and every time with specificity and innovation. I think it helps us understand the two hour, in fact, very complementary. Let's take some comments. We will start with [ indiscernible ] and work our way our roundtable.

I think it helps clarify HITSP in a way that I think is useful. It just a question, the last bullet on the slide about the values we bring to the table. Orchestrated participation in a transaction. Could you just give me a few more words about what that means from a provider standpoint?

Sure. When we are talking about interoperability among different kinds of entities and they are using and a different kind of enterprise and they are using different kinds of tools, et cetera, it is important not just to say what information is being exchanged, but what are the rules of the transaction? I sent something to you, you acknowledge that you of got it. You may have to log something about it. Army have to give you some other information in response. It make you -- It may be a complicated transaction. So, we want to say, what are all the different components of a transaction above level of granular D?

We think about interoperability, there is technical interoperability. I sent you a fax. The semantic interoperability, I send you a structure document. You can interpret that in that information can be used for support. And then there is process interoperability. We agree that we are partners. I am going to send you a request, you send me a response. We have orchestrated a relationship that is a process based transaction, so that is really what you are referring to.

Thank you.

Thanks very much.

I am going to go ahead and use your metaphor. You are a customer of the SDOs. It appears to me that what you have done at this point is you have gone and taken the existing standards off the shelf. You are doing this at the exact same time at the other national programs doing this. All the sudden we have multiple national programs with new customers such as yourself. Knowing that what you are taking off the shelf will not sustain all of the divisions we have an through the next six years. What is your [ inaudible ] and what needs to be done to make sure that these be standards are meeting HITSP needs to meet the goal of this committee?

Were you asking me that or John?


I could not tell you were looking at.

Well, I think, you know, ever since HITSP has come into being, there have been parallel efforts. HITSP has always been, where do we [ inaudible ] and collaborate to bring people to the same table? We proactively is solicit people to participate. We only know about what we know about. Some things have been parallel. We do have a philosophy of not invented here is exactly what we want. We want to go and get what other people have done and take advantage of it. In terms of ongoing great, clients, which sort of put a work together -- in terms of ongoing maintenance, we sort of put a work together. We try to balance that with the work that is coming in. Sometimes, as you can imagine being government contract, we are held to the newer things obverses addressing the older things. I think that is probably going to change as people actually start to implement things as a result of your work, were the focus on meeting to keep the body of work up-to-date inappropriate evolving is going to be very important.

Thinks. Linda, I also want to reflect that before 2011, we are in pretty good shape. We were able to find HITSP implementation guidance, and HITSP was able to find standards that relief fill the substantially all of the 2011 proposed a meaningful use measures. For 2013, for the proposed measures for 2013, we are going to have some gaps. HITSP is going to find some gaps. Not only do we not necessarily have at this point in time the implementation guidance that would be needed for the proposed 2013 measures, but there are some basic gaps in the standards and in the implementation guidance for the standards that is going to have to be addressed some hall for 2013 and beyond. For 2011, we are in pretty good shape, but I see a lot of work ahead of us.

[ inaudible ] to help identify where the base standards me acceleration and figure out ways to accelerate. So, we have been working closely with HITSP, as has been said. We will continue to use this new capability of funding to figure out what the priorities are and where we need to move the quickest to help this along.

Although the supply chain of standards has been adequate today, you might mention that there may be commissioning of standards where there are gaps that might be identified.

We encourage the [ inaudible ] process.

Thank you. First of all, I want to say that this was a tremendous presentation. I will hope -- I hope you would post a more advanced slide so we can see it. You can hear me? God forbid. [ laughter ] Just to show that there is no metaphor that cannot be overused in a series. The supply chain metaphor here I think a grocery a store. HITSP comes and takes several things off of a shelf, but this will create a meal but I have to come up with the recipe. Like any supply chain, there are time lines in the process. When Standard groups set out to create a product, it is a long time from when it starts to wind its on the shelf. I am going to make dinner tonight. I don't have any choice of going back to the farm and saying, why didn't you grow the race differently? I take whatever race is on the shelf -- take whatever rice is on the shelf. I am concerned we have a way to identify two SDOs. I think there is time to do it. I think when it gets to $38 billion, you are talking about real money. I want to see that process go forward three and a lot to ask a question -- see that process go forward. I want to ask you a question. What could we do better in terms of how this is a package on the shelf? In other words, you have this diagram with circles that starts with a big arrow down to a circle that is the reservoir of all standards. That would look better as a camouflage colored thing, because they are all different shapes and sizes. Is anything that the standards organizations can do individually or together that would make the process more efficient and work product more a accessible to non standards experts?

I will start with that. We'll come together as a multiple group to say things like, if we are all going to reference gender in our various standards, clinical, administrative policy or whatever, how about we all agree on what of value set is. That pre harmonizes. Certainly, that kind of harmonization is helpful.

I would add that in talking to the OCO, they have had discussions about who is there customer, their collective customer. What does the customer want? I think that the beauty of what is happening with the national coordinator and groups like this is that it is making that more clear. They all serve multiple stakeholders and worldwide things. With respect to health IT and health information exchange, it is clear who the customer is. When you can identify or put a face to the customer, you can understand better what they want. I think what is true of HITSP is going to be true of them. For example, I look at this group as HITSP customers. I look and see what activities you are doing and I notice that your roadmap in the standards. You are trying to say that there is a progression of adoption that is important for you to capture. Well, for me that would mean maybe HITSP ought to do something to make that job for you easier, so we should start to think about that road map in activity that you are going to do in production of our documents. Similarly, for SDOs who participate in HITSP. We have been encouraging them to produce implementation guides. In fact, these are the kinds that we are advancing to the government and they are using. Maybe your implementation Guide can have similar [ indiscernible ] and take on some of the characteristics. As it progresses down the chain, people have less of burden to try to [ indiscernible ].

I would like to drill down a specific point you just made, which is backed you have to deal with a very heterogeneous set of inputs. Is there work going on so that your ability to meld the standards will be easier? The second question is, what about standards groups licensing requirements for -- I don't mean licensing. Army intellectual property protection for the standards. Are there things that should be done? Would it facilitate a great deal if they were able to come to a common way of dealing with intellectual property issues?

Well, let me take that last one. I think that is very important. One challenge that I have in describing HITSP's work is what I will call in direction. We start off at the top level with the most granular guidance, and that refers underneath it -- let's say we go from capability to service clobbers into a transaction practice to a standard. When you go to click on the basic standard, it says, you are not on the age of seven website, please deposit of $400 including debt this guidance. Wouldn't it be wonderful if you have a Web based? You wouldn't have as much interruption. You could get directly to what you needed, because they have a business model. They need to charge intellectual property to survive. That means they sometimes have an obligation towns Cantwell the necessary -- obligational get into all the necessary so pains to make supper.

Not only that, but that one bank, if we were able to automate the presentation of our body of work -- not only does it go to you are now an age of seven website and $400, but then HITSP may be talking about this sliver of what each of seven is doing. You are going to get this huge thing of information, it is hard to weave through. If we can make it easier for them to integrate their work with these downstream uses, I think that that would be good.

So, what I hear is that one of the waste they are not standard is in the automated standards. They use very different formats and conceptual models for that.

They also have different business models associated.

Right. The other thing is any time we could avoid a licensing issue consistent with the SDOs still being able to do work, that would be helpful. And I think I heard John hinting at sort of a Web solutions and/or something that said a user authorized to access HL7 standards could start at a the HITSP website and sort of link down into a very specific sliver of HL7 without ever knowing that there was a transition because they were already authorized. That is interesting.

I want to make sure that before I go to break, we have some opportunity to set up the discussion of adoption and implementation. This is a terrific discussion.

Thank you for doing this presentation where you have shown the comparison of implementation guidance sources. I was looking through the various attributes of that, and I think John and lots of people know the various kinds of intellect and needed to help produce information on folks who want to help determine a format or the content verses' the sequencing in the processing is pretty varied across the health care organizations. And I think what I was trying to go with with Linda's question is we have done a lot of the content of this work from like 2003 to 2009. John, it seems like a decade, but it was really only six years ago. I think there is a concern, and I have told folks in my organization for health care trying to recruit people saying, if we can get something out there within two to three years, it will be used. It will be a likely contender for national adoption, because we need folks who can understand a clinical and health care processes and help create standards and content in that area. They don't quite -- it doesn't seem quite as sexy to them as saying, I have just put in a new phone and put all my mobile devices together and have all of this great information. I think we are trying to figure out how we can stay ahead of this curve for 13 and 15, because there is more and more rapid consumption of standards. I don't think that we have quite figured out how to produce standards faster. You know, there is more -- which he said, there are more grocery stores that want to open up and use what is on the shelf, but we have to figure of how to keep that content going. I don't know if there were ways that we could help you and various others figure out better ways to communicate with those needs are in terms of the right people to help within organizations and with all. Do you have any thoughts on not at this point?

Certainly, I expect this committee to come up with a series of next steps. Today's meeting is implementation guidance. The next step is adoption. That is point to create feedback loops. I would have to imagine that this committee plus HITSP ONC is going to accelerate adoption. I think this is absolutely the next focus.

I want to leverage some of what Wes said. Before I make anything that comes across as criticism or complain, let me congratulate you on an excellent presentation. I wish I had seen earlier in the process. It would have helped explain some of the terminology and approaches. John, with the work that HITSP has done, it is remarkable. The output has been extremely useful. Take what I question or suggest in the context of a deep appreciation. But I would say someone hold in my career at HIT was sort involved with standards aggressively. As a virtue of being on this committee, coming back to the process. Coming at it with somewhat fresh eyes, I am astonished at how complex we have made it. And not the least of which is just -- navigational complexity to figure out what the heck the standard is. And then to run into, I will call them picky barriers that even though we are spending $30 billion on stimulus for HIT, we have run into license issues that we cannot have a direct link into the HL7 Web sites. It is just crazy and frustrating. A there is a competition all their where there is a group of people who believe that there is a simpler way to do this, and nothing will stop them from going forward with that. This approach that we have all invested have a in is going to be the right way -- invested so heavily in, is going to be the right way. Can we remove some of the levels of intellectual? Can we have to the links [ indiscernible ]? Do we need to consider things analogous to the connection and? People who do their stimulus can get together and test their stuff out in a context outside of maybe circulation environment. Should there be open source [ indiscernible ] endorsed by HITSP that would speed the adoption of some of these more complex protocols? What is it going to take to make it so that we make the pass easier to slide?

These are all excellent topics for the adoption workgroup and we are going to be talking about. One of the thing that HITSP is trying to do, there is a tool called [ indiscernible ]. This is a U.S. health information knowledge base. You might imagine you could type in problem list and get a list of the constructs that are necessary to transmit a problem list. This is still a work in process. We have to ensure that this tool is the front end to all of the work and that this whole process is much easier to navigate through did I think we have heard from the committee, and I think it is worth exploring. How might we remove some of these intellectual property barriers? But they could do as a front end is one click access to the artifacts and not having to go through multiple layers of interaction. That is not yet a body of work that I am aware of an intellectual property side, but at least a web based navigation framework for much easier access is in process.

I guess the one thing -- I think that that point can't be overstated, because I think a lot of what people perceive as the complexity and difficulty is in the presentation. We are bound right now to word documents. They have to be decomposed and strolled together in different ways. Like an automobile. If automobiles didn't have the outside, you just see the engine and all of the mechanical parts and those sorts of things. It would be much more difficult to select one or another without having to select those things. When you put the cover on the automobile, it helps you make a selection. You can tell that this is a sports car and is probably faster than that Saddam. You can tell that this is more powerful than that. I think we are trying to get at that with the capabilities and that sort of thing. Putting that cover all the web based interface on top of things would dramatically improve the feasibility.

This is a fabulous analogy. A car cannot run without a carburetor. A car operator has 270 parts, each of which have their own standards. When you buy a car, you buy a car. You don't care about the carburetor. The idea of these capabilities and service collaborations' are trying to extract us away from the carburetor and get us to construct. You don't worry about the complexity behind-the-scenes in making the products easier to navigate would give the sense -- let's give it Jim Walker an opportunity for questions. Please limit these two very brief points.

Great presentation. Thank you. I want to follow up on Lynn and other comments and maybe just put a little more point on it. As we get to the first few years of this in which the work is pretty clear and there is a ton of it and it is critically important, it is pretty obvious what the gaps are, we are going to get to a point where we really need to rematch and health care if we are going to make fundamental improvements in quality and efficiency. Things like interoperable care processes. It seems to me that this committee is probably one good group to release somewhere down the road and not too far start trying to think about what would really break through transformed health care looks like and what kind of standards that are even in view right now would be required to support that? Is no, start getting that commission process early.

I think that is a terrific question. That in a sense is the essence of adopt some implementation. I personally hope is for that improved health care -- before that improved health care is transformed. A quick response, because I think there is a lot of interest in this topic. What I recommend is that after we come back from break, this is really the meat. I want to have some discussion actually setting that up. This is a good segue to that. I want to leave some time for that, because we have a couple time schedules today. Wes, a quick point on this?

Are you saying the next agenda item will be about the implementation committee?


Want to go to break?

The thing that I really think is important is we spent several minutes talking about the problems of penetrating our standards implementation guides, et cetera. We have created that mess. We have added complexity and layers and made them [ indiscernible ] in a lot of cases. We have a complex enterprise that we are trying to engineer Lee described how we increase option morality and so forth. The trade-off is that I think we end up with a lot of complexity, and frankly differences in how implementation gets done as a result of the complexity. HITSP I think has four different ways to send a radiology report, because we have dealt with very specific use cases. I know there is work to bring out together. I guess this is a plea for us to think about at least sort of the Unix tools approach. Give me three simple ways to do things, and let's reviews and we use and reuse those from the vocabulary standpoint. People get very confused and find it hard to work all the way through, including those who spend their lives trying to sort through this stuff. I guess this is a plea for thinking about, how can we accelerate adoption -- not just by removing option morality and adding complexity, but by saying, how do we take one or two approaches and use them again and again and again and for every use case that we possibly can.

The body language for those online, there is a lot of consensus of the economy of mechanisms to achieving its implementation. Thank you.

Thank you for that point, and that is what collaboration and capabilities are all about. He is absolutely right. When we received these use cases, they were very constrained. Therefore, here is a complex set of interoperability for just this particular use case. Across the 13, you might see different approaches. What we did with these Tigard teams is say, let's take what we learn and try to do one way for every function, you know, try to reuse components. Here is one way to do secure transmission. Document Exchange, here is the way we are going to do that. Use that over and over and over. So, getting to person in the is always our goal.

I want to compliment HITSP in terms of exactly what you just described, John. Appearance of components and the capabilities and a light has been a huge simplification. There is yet more money to be achieved, but I will be if nothing consistent in that one of the functionality is that I still think is grossly under specified and starts to impact of, how do you lose and gauge these kinds of standards, is not surprisingly coming from me of the vocabulary space. There has been much spilled over the notion of the U.S. realm and of having a consistent simplified access method for valuing assets and terminologies. The U.S. health information knowledge base is clearly a step forward. I think I and others would argue that there is more loops to be close to with respect to having a national terminology resources that would house, manage, and maintain so that these capabilities and components could reference them transparently.

And I think you'll see in some of the work group activities, he is talking about organizing with other stakeholders. That is exactly a goal to achieve.

I really appreciate the discussion. I think it's just ratifies the importance of the adoption implementation. I know that Doctor Shepard wanted to respond to some of these comments, but for those of you that do not know his history, it is interesting. If I might share some of your personal story, he had an implementation challenged in Virginia. There we're all sorts of resources that were available, but many of the barriers that we have just enumerated were barriers to citizens of Virginia, the public servants of Virginia of getting access. I think one of the reasons [ indiscernible ] reached out was because he has had some real world experience. I also think it is instructive sometimes when we look outside of our box of health care to people who live overcome challenges in other areas. I know I am in Paris in you, because you are a very humble individual, but these are the achievements. With that in mind, I would like to do two things. First, I would like to share that he is the national coordinator. We have asked if he would consider sharing the implementation work group. Second, if he would offer some framing comments for what translates from another sector that you think we should be thinking about in terms of implementation and adoption. So, with that, let me turn it to Dr. Shepard.

I appreciate that very much and your kind remarks. I want to say this. When the president was inaugurated, he had aligned in his speech that I think is sort of the basis for this conversation, what motivates me in service to him. That is that we of a government that works. As we think about a government that works, we think about a standards body that works. I kept thinking about the question. If I knocked on your door tomorrow and said that we would love to have a copy, how easy will it be for you to actually do this? All the way through the conversations with the hospital CIO plus a where we would say, we have no money, we have no investment capabilities, but could you just give the veterans a copy of the record? Could you make it available in a way that would be accessible, because we're faxing thing and doing silly things. The answers were largely, well, we have a road map. I haven't put the resources together to do this in a of the current budget. Maybe when I do the update, I will get around to this. It is really important we do this. It is right for the country, but I am swamped and don't have the ability. My on the ground sort of experience is such that when you knock on the folks doors and say, can we start playing in the sandbox today, it is not the easiest conversation to have. I am assuming all of us share this challenge. So, as I think about the President's goal of a government that works and that we are successful in the work that has been done, I basically have three basic principles that come to mind. I would hope you all engage in debate on this as the afternoon proceeds. I, unfortunately, have to attend some stem educational work for the president later today. But my question -- my three principles I would suggest are as follows. First, it would deny some of a little bit of a measuring about where we are on the standards today. There's a lot that we've talked about that is novel. There's a lot that we talked about that has been around for awhile. It would be nice if we could find a way to get our arms around our 1% of the nation providers today doing X20, 15, three? Would be nice to have a little bit of, where are we in this space and how we get that information? If we cannot get it through formal channels, is there a way through collaborative technology that we could engage folks to give us technology? Second principle deaf by the way, you did a great job, John, summarizing these three principles. Can we start listening to those folks that have to listen to the tough calls? My presumption is that the CEOs of the health hospitals are aware of this. There may be group practice administrators who might engage on these discussions on the physician's side. The MIB extension centers, as you know, we have a funding announcement on the street for extension centers. There may be a network of those that are growing up that might weigh in. Do we not on their doors and ask, when can you get a copy of the patient summary? When can report the following a quality measure or some subset to just ask? I love John's framing of commissioning new standards to close the gap. I'd love Paul Lee harmonize the [ indiscernible ]. What would just back at the source that might be a modification as opposed to new. I don't know what the answer to that is. Kind of like this QRDA thing. I don't know where that came up. Who was the creator of that standard? Following that story would be an interesting one, since it is a new term for me. Just being able to listen to folks, not on their doors to some capability to get that. All in the spirit of pulling forward. We have this beautiful mouth that even talking about 111315 that doesn't mean we don't want to get folks to start sharing in now. While we want to have this broader framework, and you will see regulations to how to do that, my presumption is -- and again I lean on my friends at the DOD and the VA. These folks are hungry now. They want that data

they want bad data now. I would imagine the same for the 40% at VA -- they want that data now. So how do we take the lessons learned? If the president's budget passes with the incentives around 30 days preignition reads my guess is that we would want to save -- readmission rates my guess is that we would want to get feedback from the folks that want to increase those data needs now so we can listen to get feedback. I welcome your collective input over the next couple of hours and I think you have been delivering a phenomenal work, and will likely give great confidence. So that is my final word of advice.

Thank you for that introduction and I saw a lot of body language around the table. You describe the theory that they cannot measure it, you can not manage it so we heard the charge to build on some of the terrific work that David Blumenthal and [ Indiscernible ] have provided.

She is pretty good, huh?

She is.

And is there is an effort, another level of the annuity to support it. And the input on with the barriers are from the real world. And I think you gave a huge traffic conversation in product versus project there is a tremendous amount of consideration to expose how that is realized effectively. And the third which I will be played back a little differently, not only the honker from those entities that are ready to move this minute but something that I take it from your experience which is, okay, what about in part from those that achieve certain levels of activity or interoperability. I hate to reinvent the wheel and I still think we have the time so I would hope a large part includes the identification of barriers and Best practices that we can emulate.


Before we break to lunch, that is a marker, are there any clarifying questions amongst members of the committee on this one?

Okay. I think there will be a need for a bio break and hunger .

Thank you in advance for the long hours.

We will reconvene in --

I'm sorry, at 12:30 p.m.


Committe on lunch break until 12:30 p.m. EST.

The meeting will reconvene at approximately 12:40 ET.

Excuse me. Can you take your seats, please. We are ready to begin.

[ no audio ]

The meeting will begin in a moment. Please standby.


Many thanks to everybody for reconvening. I tried to give the group an hour for lunch recognizing that a half hour is not possible but many thanks to everybody for sticking pretty close to that half hour. We will regroup, a terrific discussion and I hope that is a prelude to a continuing discussion this afternoon about food for thought on adoption and implementation. Before we get to that I want to remind people as we discussed and deliberate the process, both of the Federal Advisory Committee and the HITECH statute itself that we make recommendations and I just want for the record to say that as we discuss things if we have a degree of the exceptions or definitive -- exuberance and definitiveness and that our work product is ultimately expressed as a serious recommendation to the national coordinator. So I want to state that for the record as a reminder. With that said let's move on into discussion of the upcoming standards committee agendas and the work around adoption and implementation. I think the discussion that we had before got to the salience of this committee and a number of people have expressed interest in participating on deep workgroup. That would be terrific if others have interests, and to express that interested to John or me or HIPAA and we will work with John Glaser and David Blumenthal and the federal advisory committees and there were work groups -- and their work groups and we have a good note from the last -- so John, any opening comments.

As we think of corrigendum of going forward we need to be bold about it, not -- of our agenda going forward, there is 2013 and 2015 criterium and we think of some of these bold challenges that we talked about this morning. How do we handle consent decree management and segmenting the medical record with mental health, it HIP, and other aspects get segmented. How do we achieve the adoption and accelerate through guidance and defining standards. So I think it's important that we want to have a discussion with the group not to limit your thinking to oh, I have read through all of the materials today, but what other. Productivity is should we in Cajun? Because we have the 2013 and -- whould we engage in, because we have the 2013 and 2015 and posturing adoption.

Let me say something that falls in the same continuum with the specification and obviously we have to recommend to the office of the national coordinator of the standards and you may recall that it refers that we have implementation guidance. That goes to the process of being ultimately considered, the Office of the national coordinator, the final rulemaking process. All of that said there is a degree of specificity that can occur in the rate the Tory process but it's unlikely to be productive to specify every last aspect of implementations of some discussion on how we provide adequate support to the real world that will seek to make use of these standards to achieve meaningful use as effectively as possible with completion specificity that is appropriate to be suggested and put into the Reagan toward process but not so overly burdensome in that process -- in the regulatory process. So that will be some of the guidance that we will be seeking as well in terms of both the discretion of adoption and implementation and insight into managing the process itself. So let me stop there in terms of initial comments and we want to link each in a dialogue with respect -- we want to link each in a dialogue with respect to does that and suggestions and degrees of adoption that builds upon the work that David Blumenthal and others and put in the world in terms of barriers. Wes of which we might learn in support of actually adopting -- ways of which we might lend in support of actually adopting these standards. Wes, you had a thought from last session?

Yes, thank you.

First of all, it was inspiring to see the energy, new thinking, somebody coming back and saying, we got to measure this. That was refreshing did I do not mean measure quality, I mean measures interoperability. As he talked through the user communities that he dealt with in the Virginia, he was focused on the government, non-government interface and that continues to be a significant challenge to everything from standards for security inside the federal government and what are acceptable standards outside and what is the interface like and so forth. If at that high level there is one thing that I would like to have heard mentioned that he did not was the economic impact of cooperating, of the interoperating. The ARRA has done something to create an economic benefit for people in the civilian community investing in interoperability that is by government terms, "a flash in the pan." There is at least one other significant economic opportunity that we should consider using to drive achieving interoperability. That was mentioned in another context which is doing business with VA and DoD by civilian Health Care organizations. There is work going on by Kaizer. And if that works, it's not a technical problem it's a policy problem, if we solve those, that should create such pressure from Kaiser like organizations that stand to make more the more they work with DoD and VA that we should see a run towards interoperability for those organizations. I would like to see as be able to support that rather than have happened on another playing field.

The goes well with the third concept that was offered with what we've learned and I don't know if Clinton or Nancy wants to offer any initial comments.

I agree that working out issues like the Federal Information Security is a very critical because it is not only Kaiser like but the state's working with federal agencies, too. That has always been an issue and that has been part of the problem between the federal and state level so I would say that I think one key thing on this as you said it's important that leaders in all of these areas know how this work is not whether the technology is advanced enough. It will have to be clear policy and risk mitigation policy. I believe and Linda may correct me, with CMS we have been working on a draft policy for release of information to the federal and other entities which is basically to set a guideline for risk mitigation in case of breach of data. But that is still being worked I think at this point. But that is one aspect of dealing with this so there is a risk mitigation policy so that they understand what risk mitigation would be in the case of security. I think that we need to keep working that.


Let me just -- I agree with that. I heard more beyond, as sort of the economic drivers --

I did not mean to highlight problems a policy which are not our job anyways.But to recognize, I had a conversation with somebody at lunch that I talked to [ Indiscernible ] and they said, hey, we have enough money to invest in adding more support for our nurses or in becoming interoperable and what should we choose and somebody said, "am I on Candid Camera?" [ laughter ]. It has been my view that when there is an economic driver, Technology and Policy things get solved. And when there is not, you try to push it by solving technology problems and it does not work. When we see a peak one, we ought to pursue it and I think this is a big one.

Linda? Yes, going with what Wes and Nancy said, I think one of the actual items has to do with listening settings related to the nationwide Health Information Network. This is critical to the federal task force before and it would be invaluable to hear what the committee has to say of but NHIN soon -- about NHIN so that we can fix anything well ahead of the glide path.

Jim Ferguson?

It seems to me that there are some things that may be enabled by standards that should be out of scope buyer's and other things that I think should be in scope for us as a committee -- and other things that should be in scope for us as a committee. There are a number of security standards that we require here in our recommendations to data required by FISMA and the cultural aspect for federal agencies with a require separate security agreements for each entity that a deal with and are not willing to use a common agreement form. I think that is more of a cultural issue and we can enable that to have adequate security standards but we cannot give over that cultural bridge without something for us. What I would like to see would be to form the basis for cooperation paid to enable bringing together the standards that are used, NHIN and those with HITSP and meaningful use into one big standard, if you will. That is something that really should be in our scope or clearly it might be.

Thanks, that is helpful. This conversation will be a very helpful in terms of going back to the office of the national coordinator in terms of what should be done. Let's go to Gina Perez first and then Janet.

One of the things that we focus on its electronic health records and standards in regards to those. I would like to see that we really look at the role of various types of organizations so at the health information exchanges in collaboration with them and in the cooperative agreements there is the requirement that the HIE provide quality reporting so we should consider the direction that ONC is taking in regard to all of those programs and the impact of implementation in regard to that.

Thank you very much.

Janet Corrigan.

The other area was the hospital read missions, 30 day readmissions to get that information sooner rather than later and we have laid out measures for 2011, 2013, and 2015 and to be able to look at that and I wonder if that is an area that we should focus particular attention and I don't know offhand which standards of of HITSP's bundle or glob of standards relate to that area but if we could get that between all of the hospitals in the community and then realized that solving the readmission problem involves the ambulatory. To manage patients more carefully but it would be a big step forward if we could celebrate that time frame rather than having the ability to even cuddly and report 30 day readmissions in 2015 -- or even calculate and report and if we could push that up I think it would be helpful. It is eight win/win economically, they will get lower payments for the readmissions and it would pre of substantial evidence that there is a very sizeable proportion of the readmissions that are affordable soap it would free up dollars -- So it would free up dollars.

And Dixie though she could not be here send me an e-mail on what she thought the next tax would involve a did exactly this issue which is even though we're not a policy group, are their best practices, cultural issues that involve security of the exchange that her committee could start noodling through. She views her work as the technology standards, the infrastructure on which they ride and the best practices to enable secure Exchange and the guidance that you offer and we will be working with him closely on some of the cultural aspects. To me, if we get this transmission piece taking care of in the packaging is not so hard there is the technology and infrastructure and the policy and culture and Best practices around transmission that seem to be the hardest not to -- nut to crack at times.

Anything to put on the table about if the implementation does this or something else? Linda and then Liz Johnson.

Again to go back to a comment earlier today, we need to be made aware of our entire ecosystem. And that goes that we have consumed so much from HITSP and we will always put worrying about the supply chain management and I do think --

Could you pull your microphone closer?

So again talking about the supply chain management and I do think it would be worth this implementation group having a listening session with SDOs to see if they have the organizational maturity and constructs while also supporting and being true to their open consensus rules by which they are audited to actually be able to take in commission work and do this on behalf of the U.S. Federal, a national initiative at the same time that they have other national programs that have rammed -- ramped up implementations.

I appreciate those thoughts and reinforced by a luck about the continual of development and to use the technical term, the "globbing" that is done. As well as the broader ecosystem because it's not linear nor is necessarily development or identification at a particular stage guaranteed. The broader adoption, and the broader ecosystem operates not only in terms of the development but also the ultimate expression of the standard.

Liz Johnson?

I think that Janet talked about jumpstart and Linda talked about where we are and what I think we ought to talk about is this current state base line concept because in the past we have had that discussion and we have said we will set some standards. We will clearly eliminate what we want but we have not talked about where people are and I would like to hear from others around the room what the impact of that is, for example, if I said, for a given set of 50 hospitals we are not using many of the standards that have been shown as appropriate and we are looking at edge technology right now so we can transmit in the standards, I agree with that completely. And what are we looking for? How do we define current state because the implementation group -- as the implementation group goes for Levine some informations on what we actually survey for.

By you see -- I see heads nodding.

It's a difficult task to characterize the state adoption of technologies as David Blumenthal said. Ironically it further down the road we get the use can be somewhat self evident, a sort of trace of the use but we are not at that point. I think that is why we have a definite challenge. So we will circle around the room and go to [ Indiscernible ] next.

From what we said earlier about listening and what the standards need to be and which ones are most important. I think that is something this group can health, to identify the priorities, which ones do you need most that is the sticking point sometimes rather than to everything.

Thank you.

We will start with Christ and then David McCallie.

I am repeating myself on this morning but I will give you a ridiculous technical issue and that is shared vocabulary and right now terminology's are not -- I want to emphasize this point -- they are not available and what I would characterize as common, consumable access methods. Obviously I have my pet interests on what this might look like but the requirement to make these things available as needed to vendors, to providers, to the community in formats and structures that are readily usable cannot be under emphasized and I think that has been a huge practical sticking point for the implementation of many of these standards historically.

It is not pulled stomach insurmountable in terms of SnowMed itself in terms of movement of product availability. Once you get physical access to SnowMed it is still left to the user on what to do with it.

Very enough.

And through their repository they have tried to make some of these available and it's not complete at this point certainly between [ Indiscernible ] and Sinbad we have the seed of doing what you suggest.

Terrific. Because it will be difficult to have the in dropper ability that we are talking about.

David MCallie?

We be slightly off topic -- me be slightly off topic, I am struck as a vendor looking at requests coming in from clients in states trying to build channels and the other answered architectural questions of organizing wide scale interchange have been sidestepped by the subcommittee and are not addressed in the existing standards in the run-up Gascony start talking about federation. So I would put on the table in line with solving some of the granular consent management questions focus on the architecture is available and perhaps recommended or wild and acquired skill did sharing is another -- and wide scale Data sharing is another one to address.

We may have to filter over to our colleagues for some guidance.

For example, on the point suppose I want to send information from Massachusetts to the sooner Corporation and I need to have a routing directory. But at the moment there is not a participant directory in a health-care information exchange that is complete the standard space so there has been some talk of using LDAP or DNS or some combination of those so this is a very rich area. Think of components that are necessary, and with the gaps and what are the priorities and how do we solve them.

And I think the fundamental questions of what is the right kind of data to share at what levels of localities. So it maim make sense any state level to share information around state mandate reporting requirements but patient data may not make sense for State boundaries were someone who lives in Kansas City, everything is two States. And thinking of the traces of how we might decouple HIE and to what level of for annuity so I don't think why the patient records have to be regionally based at all because I am not regionally based much of the year, I move around but by local state reporting requirements are hardly regionally based did I don't think we'd tease those apart while yet. Some of them was our policy questions and some of them are standards that do not exist grid some of them are standards that have not been applied.

That point is extremely well taken because moving from that level to the trans form health care that one might imagine here we are in the District of Columbia, 72 miles of the state, and residents of the stereotypically get care into if -- in two if not three. And your points are well taken, a great area to interface with our policy and colleagues on.

Mark -- Nancy, were you up?

I will come back later.


I am is trying to health you with the transition here but I always give the hair bristles on the back of my neck when we talk about better standards, because at the end of the tea cannot mandate a standard, they have to be adopted. So this implementation discussion that we have is incredibly critical because we can do all of the buying to me and we talked before about complexity and all of those things are absolutely true. At the end of the day if we can just take some of what we have and put into use we will be a lot further down the road than a week continued to worry about -- if we can see to worry about how things will work 12 years from now.

Thank you. I think there is a resonance and the economy of standards but also the reality that it has to be adopted by popular demand, not by anointment. On the other hand, I think the economic traps are very instructive -- drafts and the lack of interoperability. So I guess the fundamental question for the workgroup is how one fosters the economic incentives for the market views, the shipbuilding metaphor. And I did not even complete the story entirely, even though a GATT standardized in 2009 there is still the English thread system and metric thread system. So we have the case we can build product.


When it comes to metaphors, you can beat a dead horse but you can't make it drink. Going back to supply chain and what I hear Mark talking about, let's not put more products on the shelf, let's make use of the products on the shelf. And what that means in effect is let's make those things work as opposed to wait for the perfect solution downstream. I would support entirely the scene that that has to keep our number one goal. I am still worried about the packaging of those things, is there ways that we can take them off the shelf. And the long-term supply, I mean, CDA seems to have been one family of standards that is coming to being by been created by standards organization. Over a mere ten or 12 years they have become actually very useful. We need to either decide that it's an entirely open market and we want every innovator to put together a little thing and put it together on the shelf or we want some kind of systematic approach to what is on the shelf and I think that has to be part of the things that we do this year to support the longer term mission.

I think they're looking at the work of the last ten years, six, seven, eight. There has been this conversion of going from a multitude of truces and taking those products of the shelf and making them as useful as they can be so I hope we hear from the implementation and adoption workgroup that we went to do what you suggested and it was hard and therefore go change CDA, the XML is overly complex in this regard can you take all of this or that and make it easier for us to use? I think it's important feedback for the ecosystem to hear and I really think that the comment of asking the users for their experience and feeding it backs of the products that we have are easier to use in the implementation and educational materials become clearer.

You know, I helped write the application that started HITSP. I was involved in discussion before that framed that paid what I had in mind, IHE with governance and authority. The part that we have not achieved is the feedback loop. We are looking for the implementation committee to will least look at ways to piece the feedback loop. Also think we need to talk seriously about the connection. I was involved in a standard called CCOW that was an individual group and we used trade shows as our connectphones and they were helping one another because if any of them did that work, one did not work and the like to health each other at that level. And somehow and the request for governance we have lost that and I hope it can find a way to get that tool.

Okay. I appreciate that discussion. I think there's a lot for the office of the national coordinator to consider and a charter for the implementation work group. I think this strikes exactly how difficult this is paid every element of the discussion, there is a tension between over specification and adequate specification in terms of, and to Mark's point to adequate support.

Going off and trying to fill all holes and getting some experience before someone modify is in perfect a standards. The cultural issues, connectivity issues in terms of broader interoperability if I could sort of synthesize my hopes for this group in part that there is a market that has been created. Wes describe the as potentially somewhat [ Indiscernible ] and some degree of connectivity with the value of the health product is increase by Pat HITSP and the economic model has changed. That is my personal best, how to tap into that city can get that experience and there's a desire for interoperability that requires the sort of incentives put forward right now to jump-start this.

Jim Walker, do you want to jump in on this?

We talked about the need of prioritizing and you mentioned, do we fix what we got, do we create new things that is important, I think we could put those two things into the same priority stream, and maybe if we put that all in one continuum I think we would provide a real service, it would not be final but at least it would be a discussion starter for us to start to organize how we do this.

Again, lots of agreement from the body language around the table. I think there is some prioritization in terms of what has been put forward in terms of meaningful use which is a natural pathway. This will be, I'm sure, and ongoing discussion. It's difficult, some of us are very practical and operational and sometimes it feels a little bit theoretical but absence without the -- I am reminded of the famous philosopher, Yogi Berra who talks about. In practice, theory and practice are the same but in practice, they are not.

I am excited about this work group in that we will take the theory and intellect around the table and also couple it with the experience, the barriers as well as good, those exemplars that can provide insight into where to set the balance on some of these things that do exist on the scale. And hopefully be pragmatic, practical, and Jim, I think that was a good summary comment, prioritized.

I believe that we have completed the planned agenda for today's.

Let me ask around the table if there's anything that anybody wants to offer.


I think the discussion that we had on next steps is fantastic and I know that there is polished that needs to be done because we have five different transactions that Floyd outlined in his charts and the vocabulary issues that we have to work on and some of the quality standards. We are working on the security side and Dixie has said that she wants to work with NITS on the aspects of best practices and what might be practical for 2013 and 2015 and we all talked about the need to work on implementation implementation, communication, education, prior position so it will be a. a very rich agenda going forward.

Any comments from anybody at the office of the national coordinator? John Glaser?

I should encourage you to use the microphone for the on-line discussion.

I would like to join the rest of you in commending the work that has been done that is visible over the last couple meetings which has been extraordinary important to was. So unmeasurable and invaluable contributions on the wall and I think John and Jon what we will do is follow up with you along with some ideas what we have and cream a preliminary agenda for the workgroup and -- and frame a preliminary agenda for the workgroup.

Thank you so much and thanks to all of the ONC staff and many thanks to members of the Committee for all that you have been doing. Again, I cannot over state to the public the decree of dedication that every member of not only this committee but all of the Workgroups has contributed hours and hours daily since the inception of the committee paid and that goes to the members of the public who have committed with great insight, a great references and we appreciate that insight and comment. It helps us not only honor the intent of the it buys three committee process but it sharpens our thinking -- the advisory committee process but it sharpens our thinking and many thanks.

Toward that end we will conclude this component of the meeting and move towards the public comment session which is also one of the components of the process. As Judy mentioned, it will be divided into two parts and I will ask Judy Sparrow how to divide this session.

We will take ten or 50 minutes from the audience and on the phone and we have had three people signup -- 10 or 15 minutes, Alice and Beth and Lawerence from the American Hospital association. If the operator could give instructions for dialing in on the phone, we will do that for about ten or 15 minutes and then take comments from the recommendations from the August Standards Committee meeting and we will do that for about 20 minutes.



So the people who signed up, if you could step toward and the operator would tell us how to dial in -- you dial 877-705-6006 and if you are connect you push *1 to speak.

My name is Alison from HIMA and I am serving as a representative of 54,000 professionals who would be required to execute the decision's been made and many of these same have been parked in the area of research, terminologies, and classification. Things that have been discussed will have a lasting impact to providers, payers, vendors, patients, and Health Information Management professionals for years and HIMA is a component in uniform data standards and have been active in promoting the use of SnowMed as well as standard EHR systems and want to ensure inappropriate implementation and use of the ICD10 for 2013. This is just a sample of the classifications and terminology that comprise health-care data and we encourage you to consider the concerns that we have raised regarding the coordination and integration of all of the terminologies and classifications. We work with you to take a pragmatic look at how these classifications and terminologies can be used appropriately and we stand ready and look forward to being called upon to support your efforts as it continued to tackle such challenging issues. Thank you.

Hi, good afternoon, Beth from the American Hospital association. I would like to state that the overall goal is to improve patient care and meaningful use should be defined by the ability of the IT system to provide a path that moves hospitals and doctors forward in improving care. When assessing whether providers have been filled the need to use objective we should focus on metrics that assess when hospitals are using HIT to support activities that have a direct and meaningful impact on patient care. Activities to solve these goals will include items like, checking to insure the summary care records are transmitted to the next provider in a timely fashion and testing to ensure that the it mimics system can health prevent unintended harm such as long dosage or medication errors. Many -- they do not meet this criterion. Using criteria unrelated to Corps patient care function would be distracting and is likely to impede hospitals ability. On the measures themselves, we strongly believe that all measures should be endorsed and adopted by a hospital quality alliance. We know that the quality work group has identified many physician low and QS endorsed measures but has largely ignored the fact that concurrent hospital measures do not exist. It would be nearly impossible for these measures to be developed, tested, specified, and implemented and endorsed by the beginning of fiscal year 2011. We are most concerned about the listed reimbursement measure which would require hospitals to be measured on whether the decrease their admission rate by 10%. This is an unrealistic expectation in the absence of any scientific research or testing to show that such a large production and readmission rate can be achieved within one year. Will stem partly, this is wholly unrelated to HIT adoption and is inappropriate for inclusion in a meaningful use definition. Currently, hospitals [ indiscernible ]. We are confused as to why the quality work group has listed a different measure cannot currently in use in its meaningful use measure grid. The question not asking hospitals to report on two similar but different measures on the same topic. Thank you.

Thank you very much.

I am the assistant general counsel at the American Hospital association. I will confine my remarks specifically to the privacy and security standards that the community has developed and keep my remarks at a very high level so we can follow up with additional details if necessary. We, certainly, support the statute -- as a statute indicates, the requirements that the meaningful use definition taking into account privacy and a definition taking account into a HIPAA. And they have a particular warning that suggests that the work of this committee should not change or alter the authority of the secretary under the privacy and security standards. We are afraid that some of the standards, in fact, don't adhere to this warning and, in fact, will change the nature of what it means to comply with privacy and consider -- security standards. If I might take a moment to use the example of security and the security rule requirements to a lesser my point. Under the Security rule, they are both required specifications and addressable specifications. Some of the standards that I think the group has developed a sort of change is addressable specifications into something that is mandatory in uniform in terms of what it means to comply with the security role. I think that should be avoided. HHS has certainly avoided that dilemma in dealing with the breach notification requirements and the guidance on securing PHI, making it quite clear while the guidance remains a voluntary to use, if you voluntary use them, you would be exempt. It is quite clear under the interim final rule that HHS is clear that nothing about those guidance standards changes your obligations under the security role. We believe that it is important for the group to be very careful about going beyond the requirements, and thereby introducing new particular written notice of compliance and changing the nature of what it means to comply with the security and complains roll.

Thank you very much.

We have one caller on the phone. Let's see if the person like to come on.

The caller on the phone is Charles.

Can you hear me?

Yes, you can.

[ inaudible ] group of the self director association. We would like to commend the great work. [ speaker unclear due to low audio ] The documents created for this meeting are a major improvement over the output of the August 20 meeting. [ speaker unclear due to low audio ] We would support implementation in an effective way. In particular, we believe that the exception of our list of documents Nista of a better link to the specific [ inaudible ]. Sedalia a list of exceptions and no direct communication. In many cases, there are many, many relationships, but not all relationships make sense. We are providing a sense of complexity and a sense of risk of confusion that can be quite simply corrected, clarify, and we think that this can happen within a few days. We will provide reasoning in making a couple examples for this interpretation. We would like to also commented on quality reporting and welcome the significant progress and clarity and that hasn't brought there. However, we are not yet there. At that are a few words [ inaudible ] that has to charter new territory, new water. But standards are not complete. They are missing alignments in terms of [ inaudible ]. We see the need for a slightly longer timeline to make the recommendation for Quality reporting of the same thoroughness and effectiveness of the one for security, privacy, and clinical operations. We trust the committee to make those minor improvements and to continue the planning and implementation of this to start a for the regulation is issued. The Third kind, we believe that the ongoing work in HITSP in this area is making tremendous progress and a standard committee should continue to support the work and further refine its standards. Thank you very much for your attention.

Thank you very much for your comments. We will go to the next question here in the room.

Thank you. High and David. I want to commend the committee for the additional level of specificity provided this month. In particular, the reference to all the hard work done by those hundreds of volunteers -- volunteers in producing the capability, so it is great to have those mapping of capabilities. I learned from my 30 plus years in the industry, no matter how good a job you do writing down things and specifying them, there is always of questions of interpretations. I would strongly urge that either the committee or perhaps ONC establish a help desk, help line with hundreds of thousands of people that will have questions from all sorts can get an answer perhaps in a consistent way and where they will not have to -- You don't have a thousand people asking the same question because they will see it has already been answered. I think it would be a great efficiency in clarifying the meeting of the various standards. Thank you very specific brief points. I believe that in the glide path of Snow Med, there is I CD nine and ten, but there is a current [ indiscernible ] since. It seems more logical to allow the ICD nine option, if you call it that, to extend until the ten kicks in and October rather than nine to 12 months earlier, so I would suggest looking at that. Finally, I really appreciate the fact that HITSP has limited its focus on purpose to the boundaries between organizations, because that is where the standard position really pays off where you do not have lots of proprietary customer interfaces. So, for instance, in the security area as mentioned earlier, I do think there is one exception where I would question the selection of a particular standard, which is the enterprise user authentication, which is selected for 2011 and then disallow in 2013. I don't see how it is really a glidepath to XUA. It seems like there could be a lot of expense put into organizations who are already there migrating to this only to see that it is a short-term investment. In two years is not even what they are supposed to be using any more. That one seems to stick out to me like a sore thumb, like why is that there? What is the economic impact? Is this really worth that?


Thank you very much. My name is Tim. I am a technology vendor. I just want to talk about for new technologies that people may not be too interested with. I want to reinforce what he set about setting up some sort a collaborative website that people can participate in. We are involved in efforts to take the existing financial systems at each agency and department and the federal government and link them together so you can track and appropriation of two signature and then to 0and before a portion and then back to the department and so on. The key to us is we are going to of a very, very large collaborative website so we can get all the people under 30, because they are the only ones who count, to collaborate entellus their best experiences of how to implement this so we can take the experiences we have in this demonstration at the Department of Defense and rolled out to education and HHS and two of our else. I want to reinforce that. Secondly -- maybe Elvis mentioned this one -- maybe I will just mention this one. I was surprised. I got a call last week from my partner in a very large state to talk to the governor and attorney general and they were upset about their Medicaid reimbursements. Apparently, they paid for a wheelchair of a certain model, paid $600 you could buy it on the Internet for $150. They said, is this broad? What is going on? We have some skills in terms of banking, and there are some issues there. The answer is, we can set up a system in about a week for this government to be able to go in and have rules that if it is within 10% or 20% of the reimbursement rate, approve it. If it is not, take it out so someone can look at it. You'll always be able to test the market that way. They simply did not know that that's kind of technology to do what art match ups on the Internet and be able to data mining them were possible. Cheap, easy, fast technology. There is lots of stuff off there to help people implement this. Thank you.

Good afternoon. My name is Clint. I and the CEO of a company called [ indiscernible ]. I came down here because I was curious. I had no intention of getting up and say anything, but I am actually quite encouraged. I like to share with you in about two or three minutes why I am encouraged. First of all, any committee at this level of technology that can quote Yogi Berra and get away with it, that is a sign of great encouragement. [ laughter ] In a post-mortem of the Santa Barbara real failure, and David said that this thing failed basically for two reasons. It was too high tech and there was no recurring revenue model. I'm here to give you some encouragement. There is a revenue model of their four interoperability in the National Health Information exchange. It is called the release of information industry, and you just heard from one of the industry people here. That is about one. $5 billion, very reliable recurring revenue model. And these people are in hospitals and in physician hospitals making on balance paper records, copying them, and mailing them to the request for. That is not what you would call very sophisticated technology, but it is a technology. Working with Mark over shed, we take what we have an move forward on that. Well, this is what you have. It exists today. These people are teching up. We happen to represent our company, someone who is in South Carolina. We have to deal with them. We do all their stuff electronic and we deal with it all electronically. If the big one hits Charleston enable all, our people from Minnesota can go in and get medical records out of their and move the more they need to be. Our customers pay by a company called E M SI. They are hired by insurance companies to get your medical records with your permission. They underwrite applications for health and life assurance with it. Up until about six months ago, we had seven or eight locations asking us for stuff, and we would try to send it to them electronically. We would have to mail it to them. Now, we electronically request these records. We get them to them within one day, and I am happy to tell you that they pay us the next day. Now, that is a PDF. That can be used creatively by the insurance underwriters to put in various models or rhythms and create and hand to the customer something light a PHR that jury is interested in. You are now engaging the patient, the consumer in terms of managing their own medical records, their own medical information. It is a really good start. I am very encouraged by what I've heard today when it comes down to a practical level where I operate. And we operate fairly broad for a small company. At any rate, my hat is off to you in that regard. I look forward to attending more of these.

Thank you very much for your comments.

Good afternoon. My name is Lindsay. I am a registered dietitian and a member of the diabetic association. We are the largest association of food and nutrition professionals in the world with over 70,000 members. ADA has been working diligently for the last eight years to provide and develop both evidence based nutrition based practice guidelines for our evidence analysis library and a standardized terminology to reflect a nutrition care. The terminology is called International by text and nutrition terminology. 88 is working with other organizations to ensure that a nutritional data center is included in the electronic health records, all the methods, and language. This reflects an approach agreed upon both by ADA along with leading experts. ADA had been supporting the implementation of the terminology in practice in the U.S. for the past five years. Several EHR vendors have been licensed and are building it into their systems. When a published updates to our terminology and are in the process of conducting research into verify its use and the liberty of practice. We have been identified as a Standards Development Organization for nutrition and diet text terminology. The terminology have an adopted in Japan, Korea, Netherlands, and is under review in Canada, Australia, and Israel. In the past, ADA has submitted comments to HITSP and this committee on how to include key nutrition terminology and concepts and meaningful use in certification standards. Our comments have consistently address the needs for an electronic health records system to have the capability to, one, provide key nutrition related data to registered dietitians so they can provide quality care and, two, care provided by dietitians be captured in a meaningful way to communicate with the health-care team with the plan is to enable it data collection and to incorporate data into a larger systemwide [ indiscernible ] as part of an overall treatment and prevention regimen. HIT ADA believes we are a key stickler for the standard certification process of. The initiative is to create [ indiscernible ] and health information exchange. Nutrition is known to be associated with seven of the ten leading causes of death in America. We thank you for this opportunity to, and look forward to working with you to support the use of nutrition related care and data for support.

Thank you try much for your comments.

Hi, American Academy of ophthalmology. On behalf of the academy, we'll like to thank you for the ongoing dialogue we have had with ONC and the opportunity to provide comments. The academy is the world's largest of a -- colored association. 18,000 members in the U.S., 93% of all ophthalmologists in the country. The academy still has concerns with the revised meaningful use matrix and standards, which is not addressed and may hinder electronic EHR adoption of ophthalmologists. Our physicians are in an ambulatory settings and may not order traditional laboratory tests and radiology procedures as other physicians would in a hospital setting. However, the academy has created an eyecare technical framework that provides technical specifications for computerized physicians ordering of procedures within the office. For example, visual fields, photos, biometry. These are all key point of order entries used for eyecare and to satisfy the CPOE criteria. If not, I care would have an issue with the proposed CPOE measure. With regard to specialty relevant measures, we found that priorities, and improve care coronation and improve population for public health, four of the 2011 measures are not to remain too many specialties. One, Report 30 days readmission rate. Thank you, a person and daughters were medication reconciliation was performed. 30, reports up to date status for Delta and immunization. Four, what results submitted electronically. If these are considered mandatory measures for meaningful use, ophthalmologists and many other specialists in and with great care settings would have difficulty qualifying, because they do not apply to their practice patterns. Currently, we are unaware of laboratories providing electronic interface with I care only each our vendors. Rather, only for large enterprise vendors. Ophthalmologist don't order enough lab tests for the laboratories to find it cost beneficial to create interfaces for data exchange. In addition, 2011 objectives are not specialty germane. One, calculate and display the and I. Two, the capability to provide electronic centrum surveillance data to public health agencies and actual transmission according topical law and practice. In regards to the HIT Standards Committee, we are concerned with the presentation on incorporating quality measures into the EHR. A list of measures presented to the Committee for 2011 did not have any eyecare measures listed or any relevant measures ophthalmologists could use for qualifying for meaningful use. If it is not possible to make quality measures electronically compatible for all specialties in 2011, then the requirement needs to be pushed further and the timeline and tell all measures can be electronically collected and physicians should still be able to qualify for the full EHR in incentive of $44,000. The Academy appreciates the opportunity to comment to ONC and looks for to providing ongoing input to ensure that the meaningful use objectives and ranchers are able to be achieved by practicing specialists. We also would be happy to provide real world feedback on implementation from an ophthalmology practice perspective. Thank you. HIE thank you very much for your comments. Are there any more on the phone?

Nobody is on the phone. If nobody in the audience, we should make -- audience which is to make a comment, I think we are complete.

Well, let me thank all of those to make comments. It really goes out to our process. I very much appreciate each and every, entered today. It is most helpful. Thanks to all of my fellow committee members for all of your work, the ONC staff. With that, we stand adjourned until our next meeting.

Thanks very, very much for all the hard work. Many, many hours. Thank you.

No comments:

Post a Comment