[This Transcript is Unedited]
Hubert H. Humphrey Building
Room 425A
200 Independence Avenue, S.W.
Washington, D.C. 20201
Call to order - Mr. Hungate
MR. HUNGATE: The thing that you have in front of you are the specific comments that have come back and I think it is very important and very useful to read through the specific comments because there is real content in them.
I end up with some considerable uncertainty on what the best way to deal with what we have is because I don't feel that we can simply report back here is a scattergram because I don't think that is going to get anywhere.
The other things that concern me are that watching the letters that go forward in the vetting process, that go through our full committee work, an awful lot of our recommendation content isn't as strong as that. Right, Kathy? You know, it is collected from input but it doesn't have the same level of development and that becomes a problem. We say about ICD-10, but it is almost redundant for us to have a recommendation on the ICD-10.
MS. COLTIN: I think part of that is the timing issue. We have been working on this for several months and didn't know that they were that close to putting --
MR. HUNGATE: So, I end up trying to say, well, what do we mean by all this. They are all important, but there are different stages. So, what is it we really need to say?
That is where I have this feeling of uncertainty. I can grapple with some of my uncertainty by going back to the HHS strategic goals and say, okay, pick out from those goals, those things that I think are germane to what we are working, what the Quality Workgroup is -- and one of those is reduced racial and ethnic disparities. We have got several specific recommendations that relate to that.
They are in varying degrees of development, I think, but they are clearly to that. There are -- some of the effects relate to improving the managerial efficiency in the way we go about it and then the other is health care quality, improve the health of Americans.
MR. STEINWACHS: What is helpful to me is just -- I think we have been inclined to cluster things. Maybe we could pick one cluster to move ahead first. So, this racial and ethnic disparities I think fits very much with where the Population Committee is also going. It seems to me that is a very important cluster.
Kathy went -- I think you went to San Diego to talk about what are the kinds of sort of items that fit around pay for performance. So, there are a set of the quality measures that are really tied very much to the capacity to use the system in a pay for performance, pay for quality and because it seems to discuss the next place where sort of demonstration efforts are going. It seems that that is another cluster that has some importance.
Then there are other sort of maybe quality issues. I guess my temptation might be is to say maybe the two priorities are to think where do we go first or can we move sort of the pay for performance side and the racial and ethnic disparity items and then maybe consider the other ones as a next priority after that or do something like that because this list has all sorts of important items that we don't sort of hang together and if we have to defend them as a cluster, do we want to do it --
MS. COLTIN: I would agree with you with one caveat and that is that I think pay for performance is a narrower construct and I think we should take -- I think we should think about changing benefit and reimbursement strategies because another example that depends on this type of measurement is a tiered network for health plans based on quality and cost and when you are tiering providers and saying, you know, if you pick a provider in Tier 1, this is your premium and if you pick a provider in Tier 2, it is that cost, some additional dollar amount.
What is happening is in order to tier on cost, you need to do risk adjustment, you know, taper risk adjustments at least and in order to tier on quality, you need some of this same information. So, both of those things are progressing very rapidly in the field. You are seeing the evolution of these consumer directed health plans, which include a lot of these kinds of compliance, as well as pay for performance.
MS. JAMES: Has any of that been ruled out or is it still under discussion?
MS. COLTIN: Oh, it has been ruled out, I am sure. Consumer directed health plans are out there. In fact, there was a conference in Washington on the 15th of September, I think, and you can actually see all the papers.
MR. HUNGATE: Vickie, go ahead.
DR. MAYS: I also wanted to jump into this cluster issue because it is something that when you really go through this and I thought about, you know, kind of how it is that we have actually been dealing with the quality issue, which has been as a small committee. What I actually thought might be workable is to think about if we could cluster them and then maybe taking this to that particular subcommittee because I think the more that I keep seeing -- I think the subcommittee will have the background that you need, like, for instance, the race and ethnicity stuff. I went through that in terms of things like, well, what the rationale is. You know, we can do it like that and think that maybe it might be helpful to see which of these you could give to the subcommittee and then assign, you know, something from this to work with that subcommittee to get either -- and I don't know how you want to do it.
It would be either to get equipment or to get even one letter that has some of the heart or -- I don't know exactly kind of what the end product is, whether it is really to be a report or whether some things are better done as letters. I would suggest killing clusters and if we do after we look at the particular clusters, what is the most effective thing that will move that because some might be letters. Some may be not.
Then, three, is there a process to work with the subcommittee that can then help get this stuff done? Because I think it actually could get done by March then if you kind of did it that way.
MS. COLTIN: I think that is a good suggestion.
MR. ORTIZ: This is kind of a general comment because I think like everybody, I am kind of still struggling in terms of what we are trying to do here. I think Simon actually did a nice little job on page 8 of his comments saying the exact same thing that Bob is saying, basically I remain unsure of what to do with the quality report and talks about a lot of these issues.
Let me just throw out some things that I am thinking. I mean, to me I am trying to think of -- I am very outcomes oriented -- what are we trying to accomplish? So, on one hand, on one level, there is the report itself. So, I am thinking the report is something concrete, something that has been developed. So, one thing we need to do is do something with this report, either get it somewhere out the door, published, sent our or canned or something because they have to have it. I am not saying that to be mean, but we have to have some kind of conclusion and it has been going on for two years or something like that. We need to do something with this report and we have to figure out what to do with it. Do we make a report? Do we do a publication? Do we say, you know, we are too late? Let's drop our losses and move on, but I think that is something we have to make a decision on. So, that is one piece, I think, of what I think our role is.
Then the No. 2 thing is I think within the report, there are maybe things that can inform us as a committee, as a subgroup. I mean, we have to decide, okay, once we are done with this report, what do we do next. We are supposed to be doing something. So, if we do something, you know, what is the next thing to do?
MR. HUNGATE: We are supposed to at least look like we are doing something.
MR. ORTIZ: So, you know, one, the report itself could be used as a way to inform us of what we might want to do or some next steps. So, this sort of helps us some, although it doesn't give us the answers, but it does help us. I mean, Bob's analysis and, you know, one of the things I tried to do was go down and make these kind of general things, where I tried to look at things that were considered important, which would be on the right side, but, yet, things that were considered feasible, which would be on the left side and then things on a time frame that were also on the left side because it is more of a now versus later.
Now, that may not be what we want to do, but I am trying to just get a way of saying, you know, is there a way to stratify this and if you do that, there are like five things that kind of sort out that way. Now, does that help us? I don't know. I mean, you know, because one of them is like the secondary diagnosis, the provider identification, the patient identification. Does that help us? I don't know because it depends on what we are going to do.
I mean, that could help us because it could help us in terms of deciding what we are going to do in the future or what we are want to focus on next or hearings, et cetera.
Then the third piece, I think, which relates to that is the fact of what are we going to do but it may not be necessarily informed or related to this report. It doesn't have to be, although it sounds like if you did the work that this should be part of it. So, it should be part of it. It may not be the only thing. It may be that we do this and something else. But I am trying to get something like -- you know, like the standards thing that has kind of a mission, you know, to deal with the ICD-9 to ICD-10 thing.
They kind of had it. It was concrete. They did it. They had hearings. They made a recommendation. At some point, I would like to see us --
MR. STEINWACHS: They did start in 1990.
MR. ORTIZ: Exactly. We still have eight more years on this.
I would kind of like to get people's input in terms of helping us determine a direction. What do we do with this report? What are our next steps? And do you see that there is -- I mean, there weren't a lot of responses -- is this useful at all to really go through and use it to stratify something, direction, priorities, et cetera, et cetera?
MR. HUNGATE: We have some recommendations from some of the respondents who made some choices of things they thought we ought to deal with. I think we can pick those up as partly --
MR. ORTIZ: The problem of the choices, the ones that actually gave us recommendations, there were only three. So, this is not enough of a number to really, I think, guide us. If everybody responded, it would be helpful. This plus this and maybe we could get people just to give us -- maybe that is something that the committee decides. Can't we just get the people to do this, the prioritization. You know, like give us your list of top three things or top five things, if that is helpful. Maybe it is. Maybe it isn't.
MS. COLTIN: I think the important thing is -- hope we get it.
MR. ORTIZ: There are only two of them that overlap, like two of them put secondary diagnosis but that is two. You know, we don't know what that means.
MR. HUNGATE: That could mean we asked them about that twice.
MR. ORTIZ: Exactly. So, I don't know what that means.
MR. HUNGATE: John, you had something --
DR. LUMPKIN: We could just look at it -- I think there are four people who did that and we could just take the ones we like.
MS. COLTIN: I think if you take -- there were two very good suggestions, I think, that were made at the full committee meeting in September and some of them also played out in the comments that came in on the survey, as well as the two suggestions that came out today about clustering and about delegating some way to other subcommittees. I think that is a good way to actually put those together.
One of the recommendations that came out in September was that -- or suggestions, I should say that -- our recommendations were too prescriptive. They were too detailed and they needed to be taken out of the trees and more at the forest level. So, to say, for example, that we want to add race and ethnicity or make race and ethnicity, you know, a standard data element on the enrollments but not to say how it should be done or what the wording in the implementation guide should be or to say that we believe race and ethnicity should be reported by providers in the context of submitting information about the care delivered, as opposed to saying it had to be on the claim versus the claim attachment, you know what I mean, and leave those options open.
So, that was one suggestion I heard that I thought was a good one and basically for each recommendation what we should provide is a statement of the data needs, a description of the existing barriers to meeting that need, a broad recommendation, such as the examples I just gave, and the rationale for the recommendation, which we already have, but just in that kind of a format.
I went back and was reading the transcript and I sort of distilled what I heard from the suggestions about how to do this so that it would be better received as opposed to, you know, this is what needs to be done but not necessarily how to do it.
The second, I think, is to do what we talked about today, which is to say let's cluster these and let's deal with those that we think are high priority by whatever means we establish what those are to narrow those down. To me that also means moving out of our current recommendations those things that have already now been addressed . So, you know, you have the ICD-9 letter that has been approved. Let's take that out of our recommendations and put it into Section 5, where we talk about other things that are going on that were supportive of interests in the -- or reductions in the data gap.
MS. GREENBERG: The other thing I would say is that keep in mind what Jim Scanlon said earlier that this -- because this was recommended by the committee, you know, doesn't mean it will happen. So, I certainly would have something supporting that recommendation. If you feel it contributed to the quality of supporting the ICD-10 recommendations, I would --
MS. COLTIN: So, we could support the -- I could see supporting the recommendation of another subcommittee, but we have already actually endorsed it at the full committee.
MS. GREENBERG: Yes. It doesn't need to go before the full committee again as a recommendation but I would not drop it entirely.
MS. COLTIN: So, we could put it into Section 5 of the report and commend it --
MR. HUNGATE: Let me try a model here because I think this inter-committee activity is also a crucial part of what we are grappling with and, in effect, the ICD-10 process worked through the Standards Subcommittee. All the DHR(?) recommendations that are in here, interoperability, basically works for NHII.
MS. GREENBERG: Well, or the subcommittee.
MR. HUNGATE: I think I can take virtually everything we have said and it in one or the other subcommittees. My sense is that the recommendation will be a stronger, better developed recommendation from that place than from this place. Now, we can articulate that it is necessary and essential to the quality that there be action. I am trying to understand whether we are really making a recommendation to HHS in our report or a recommendation internal to our work.
MS. COLTIN: Well, I think there is precedence for doing the letter --
MR. HUNGATE: That is what I am trying to make sure I really have a fix on because I can see some other things that we might be uniquely positioned for. I can think, the personal health dimensions and patient language.
MS. GREENBERG: That is NHII also.
MR. HUNGATE: Yes, but if we say consumer driven quality system -- if you are going to start with the consumer, you have to start with consumer language. So, I think we could sign up for a stronger piece of that if we did it early and had it worked out well with NHII.
DR. MAYS: I think that is kind of why we need to get into the subcommittees, because I think that is where maybe the decisions can get made because there will be part of it that the subcommittee may see something that they know and they would say let's definitely send this to HHS and then there may be something else in which then it is something that it has to do more with internal workings or something and then it gets -- I kind of like do the report after that. I think where you are going to end up after you work with the subcommittee is any different than where you are starting and I think it is also would be easier to get it through the full committee. What has happened is that all the subcommittees have had kind of ownership of working on some part of it and then they may give you suggestions that you would want to put in a report.
MR. HUNGATE: I have one concern about what you said because I don't think we will ever finish because it is a moving train. You know, all the -- everyone of our recommendations has a long time frame. They never reach completion. So, you can't wait to report until you are finished because you never finish.
DR. MAYS: I think you set a deadline. You ask the committee for it. They do it or they don't. And if they don't do it, you take it back. You do it yourself.
MR. ORTIZ: I guess once again it kind of goes back to trying to figure out the whole point of the report, who is it written to, who is it targeted to, who uses it because that makes such a huge difference. I mean if it is really -- and I don't know enough about the way this thing works to really know that, you know, whether as a subcommittee we can put out a report and put it out there and it can support other stuff that other committees do, whether it is something that when we put it out, it really goes to the main committee and really goes out as something from the NCVHS because if it is really going out to the outside world, then I can see this being we did all this work. We are trying to give a report that, hey, these are things that are important to advance health care quality in the United States. It is okay if someone else is dealing with it.
I will give an example. We think that in order to do this, you have to have a patient identifier. Well, we are not going to go into details. We are not going to do all the hearings on it, but we think that that is something that is necessary. So, it is a Quality Workgroup. That is a recommendation. We don't do all the other stuff. Other people can work on that but this is it. You know, health care quality is this. This is some things we have seen and this is what we think. We need a patient identifier. We need this. We need good ways to code data. We need good ways to pick up discharge diagnosis, blah, blah, blah.
These are the things. You put it out there and this is it. This is our report. You run it through the committee. You let them look at it and you get it out the door. If it turns out you have a recommendation that is supportive of some of that other stuff, then out in the world where you are looking for support because things are controversial, maybe that helps because some of the -- hey, the Quality Group thinks that, boy, you really need to do this if you are going to have good quality. We didn't say how you do it. We didn't go through all that stuff. But I don't know if that is our purpose.
So, I am throwing that out as maybe something we should do.
MS. GREENBERG: I really agree with Eduardo. I think we have had I don't know how -- I don't have the report in front of me, but I don't know how many sessions, how many hearings, how many presentations in which the need for certain types of information have been identified repeatedly in order to assess quality. I think the more specific you get as to what the recommendation should be, the harder it is going to be to get that to the committee and even it may be -- and to get it perhaps adopted by X12. I know the people or whoever have to adopt it. I know the people who were trying to arrange an extension at X12 feel that really what -- the kind of presentation that needs to be made to X12 is exactly kind of what Eduardo was saying, that we have heard this testimony. We have studied this issue and we feel that from the point of view of moving ahead the quality agenda, these are gaps or these are needs and they need to be addressed.
And let the people who are experts in working with the transactions or whatever come back and say, well, this is one way you could address it or these are four ways you could address it and then, you know, initiate the dialogue. Whereas, if you go like to X12 and say make this situational or make this mandated, they are going to go I hate that and they are going to react against it. But to just keep this as an internal document and not get out there what it was you heard and given what you have heard, what you feel are the critical deficiencies in the current information infrastructure to address quality problems, I think would be a shirking of responsibility, but also would be unfortunate because, I mean, we have always, you know, had the kind of position now for years, since I have been involved with the committee, if you study an issue for a few years and then you get something out.
Maybe that is not the definitive word. Maybe it says more this work needs to be done, but you do get something out because then, you know, new groups come along as we see this happening and if you haven't done that, there is no kind of record as to at least what people who worked on this or who had the opportunity to synthesize it, what they believed and what they found. It is used well beyond the department. People look to these things. Where they would support them, of course, they point to them. Where they don't, maybe they ignore it.
But I think the committee has that history. So, you may want to back away from some of these very specific recommendations, but to not -- to back away from a report, I think would be a mistake.
MS. COLTIN: I am not suggesting that, in fact. I think we were only talking about responding to the suggestion around the recommendations themselves, not the rest of the report and saying sort of how you set it up. Right now it is very prescriptive and it is very detailed and it doesn't need to be to meet either of your recommendations. So, I think that these -- our recommendations could be brought up a level in terms of the way they are worded and get out of the specificity of how to do it, but rather what needs to be done.
Then, I think, you know, what you are suggesting is not inconsistent with what Vickie is suggesting. I mean, at that point the report is done. This report needs to be done. What this community chooses to do with that report once it is done could be to say we all endorse this. We think this needs to be done and we think there is a role for the Population Subcommittee in furthering this one and getting input from the field for the Standards Subcommittee to try to put in this one and so forth.
But at least you have a report and then in terms of furthering the goals of the report and the recommendations, but I think it is the role of the Executive Committee to say listen Population Subcommittee, this is an important, you know, recommendation that came out of the Quality Workgroup. It needs to be on your work plan for next year and Standards and Security, this needs to be on your work plan for next year.
Once it is on their work plan, it will move forward, but it will move forward in the context of those subcommittees. So, I am not disagreeing with anything that I am hearing. I think it is very consistent, but I do think there are a few of our recommendations that have already been accomplished and can be moved into other sections and we can, you know, endorse, for instance, the ICD-9 recommendation and we can -- I think there were at least two like that, the terminology one, the clinical terminology.
MR. HUNGATE: If we are identifying needs, isn't it different than making recommendations? If we are basically saying our testimony exposed these things --
MS. COLTIN: Well, a recommendation is --
MS. GREENBERG: To address these needs to fill these gaps.
MR. HUNGATE: We are basically saying our testimony exposed these things that need to be fixed.
MS. GREENBERG: Except where you can get through some specific ones --
MS. COLTIN: Where we can be a little more specific, we should.
MR. EDINGER: We probably should be careful about making too specific a requirement. George and I worked on this one many years ago, about the -- on a physician. That was a very sensitive issue that -- didn't want to pay for, submission testing. I would be careful in making it too specific. As you mentioned, even though these are examples of reporting selected tests on admission, if you make a specific recommendation that these are the specific tests, that would be a recommendation that you should do pre-admission testing and recommending, which is probably not what you really want to say.
So, I would be careful about getting too specific on some of these topic areas or these priority codes, which ones, priority health conditions you use. I would probably be more general and not get into the nitty-gritty details. But I agree, we probably should get something.
DR. LUMPKIN: I think I agree with what I think I am hearing as far as which direction to go in, which is to take this up a level and get the report out. Say these are things that need to be fixed. These are things that need to be done, not specifically -- you know, a collection of race and ethnicity needs to be available, needs to be used in making management decisions -- to be used in making individual decisions or whatever.
And then we are spending a fair bit of time looking back on a report and what we need to do is then take that report and decide where we want to focus our efforts in moving forward.
MR. HUNGATE: I agree.
[Multiple discussions.]
In that context, part of it is a grouping and part of it is the moving up of -- I think the groupings are the population-based race/ethnicity issues. The things that are currently labeled under the pay for performance rubric is another grouping, I think.
A third one is the grouping of getting the NHII to follow the quality needs and then the other piece there is the single statement on privacy, which I think goes to the categories that those two fall into. There is some other content that I don't think we have in there in the recommendations.
One of the things that I keep bumping into is people making the statement, people respond to incentives and all the incentives in the system are cost. There aren't any quality incentives. I think that there was some of that in the testimony. But we haven't said anything about change in the incentives in terms of quality versus cost. That is not in our recommendations. So, we don't say anything about review or improve the payment system so it is more quality sensitive. We don't make any recommendations in that direction.
MR. STEINWACHS: But we cluster those like around pay for performance. Then you end up introducing that and identifying a consumer essentially designed health plan as options that need to be looked at. That is again sort of our sense of how consumers are empowered.
MR. HUNGATE: Well, I am a payer in my other hat. I am up at the Massachusetts trying to buy health care for all the employees. Well, we are being prevented by pay for performance bias. But the performances, do they have a minimum number of cases and is their cost below the other's cost and that is the basis of definition. There is no definitional content or what the risk adjusted outcome was of what the risk base that they were managing. So, there is no assurance that you have comparable features.
So, I can see that now just, you know -- them by the inability to answer questions. Now, the other piece of that in a positioning sense that I think relates to our subgroup is that the -- our workgroup, is that we are placing long term reliance on the improvement of information by NHII and getting better information to define a position. But we have to get from here to there.
It seems to me somebody has got to start defining the modules of information that are going to be useful at the point of care. That, to me, is a piece of the quality agenda that could be addressed here as uniquely different from other places, which is where I come back to the risk adjustment piece of it, patient language and other things.
DR. LUMPKIN: I am struggling a little bit with what you said in that I am not sure our role is to say that that is the way to get quality in pay for performance. There are a number of strategies that are currently in place. To get a 10, based upon those strategies, there are data and information needs and those data and information needs cannot be met with the current approach. I mean, that is how you get to wanting to have the switch on pre-existing conditions because if you are going to do things for performance, are you going to do consumer-based, you know, cost stratification, then you need to be able to do some risk adjustment.
So, our conclusions can really be based upon our assessment of where industry is and what are the daily
needs --
MR. STEINWACHS: You also need to know what complications occur.
DR. LUMPKIN: But the other piece that we aren't really looking at is looking at the future with the NHII. What we believe, at least what I believe, is that having -- incorporating decisional support into information systems that are now, you know, everyday you see in the trade press about a new entity that is purchasing a long term health -- how do you get from that -- what are the information implications of that? Are there standards implications?
Maybe that may be how we transition from our report defining sort of broad areas to selecting some and say our next set of hearings, we are going to really begin to delve into those people who have already developed decisional support and what are their data needs and how do they get them or what are the barriers that they have had in trying to implement those.
MR. HUNGATE: Look at those leaders that have done those sorts of things, Intermountain Healthcare --
DR. LUMPKIN: It may be that what we have done today and what is going to be done with the CHI may have solved all the problems. But to ask those questions -- we need to ask the questions of people who are pushing the envelope.
MS. COLTIN: I would agree with that and I -- there will always be sort of poor -- in terms of functionality when you are looking at electronic medical record systems. I have lived through three of them and I have worked where they transitioned them. Not one of them has had really good clinical decision support functionality. None of them came with or established as a minority population-based, search and query capabilities so that a physician could say let me see all the patients in my panel, who are over 50 and haven't had a mammogram in the last two years. This didn't have that capability built in at all.
So, I think one of the roles of this group could be to take some of those -- and try to elevate those types of functions because when you go out and talk with physicians about what really makes an electronic medical record attractive to them. It is a lot of those features, not so much the everyday. A lot of them are okay with the current way of recording notes and things like that.
They are a little intimidated by having to type in their diagnoses and their notes and things like that. The benefit, the payback is often these kinds of functionalities that can be delivered and yet many of the systems don't come with those or they are inadequate. Even just looking at some of the best practices in that area would be useful
MR. STEINWACHS: A companion piece might also be interesting. What information will support the patient or consumer in their decision-making about their own health care. I am not saying it compares with which providers, but what do I need to do. It seemed to me that you could look at hearings potentially in both areas as to how do you think of that decision support, as well as put it in the context of a new clinical setting, how do you provide that physician support.
Stan.
MR. EDINGER: It might be better to go John's direction than to go into the pay for quality because that is more of a political -- it shouldn't be, but a political decision area with the department and the other -- depending on what happens next election. Then these basic issues of the data, which is our role, what do you do with it. How do you get things implemented and how it could be used by different people with different approaches to pay for quality or pay for services.
We might also want to bring -- Eduardo and I have been talking about this separately on another issue, but you might want to bring in the people at Intermountain Health, who are doing the work, but you might to bring in some of the CEOs as some reflection of how the hell they got them to pick up the results because it is hard to put something in place that nobody wants -- it may be a good system but nobody wants to purchase it. That is not going to help you any.
MR. HUNGATE: There is an interesting article in the weekly magazine in The Boston Globe about the acceptance of --
[Multiple discussions.]
And it has not had any commercial success and yet it is a decision support tool. It relates to problems and it is well-documented. So, that would be I would like.
MR. LOCALIO: A couple of points.
It was interesting for me to see my bipolar --
[Multiple discussions.]
-- anybody else who would just use the white out, but it is interesting to see that and there is diversion here of opinion among those who -- these issues that -- because of just that assumption.
MR. HUNGATE: I think that is one of the keys.
MR. LOCALIO: But to follow up on that, just from listening to people, I would recommend avoiding somewhat simplistic or one size fits all recommendations. I know where some of the people here come from and I have moved around a little bit. So, I understand that sometimes what works on Longwood Avenue is never going to work anywhere else for very good reasons. So, we need to be careful about what happens -- what works in new -- or a different kind of place.
MR. HUNGATE: Different climates in Longwood.
MR. LOCALIO: So, we have to be careful about what we say. That doesn't mean we can't propose a certain number of principles that at least -- I think everybody agrees there is a least common denominator. There are a few -- and say we all kind of agree that these things should work.
The other thing is after discussion, outlier opinions tend to regress to the mean.
MR. HUNGATE: Yes, they do.
We won't ever have time to have the full discussion to get all the regression, you know, and so we have got this body and we have to do something with it. I think the only way we can do that is to sort it into categories where we can get the help that will make it work in the longer context.
MR. ORTIZ: I am going to go back to my concrete thinking here. I am going back to this report. It sounds like we want to do this report. We want to change some of these more detailed recommendations to a little bit broader instead of just in the detail of, yes, we think that we need to collect race and ethnicity data; yes, we need to have a patient identifier, et cetera, et cetera. That has to be an action item to be done.
No. 2, I am intrigued by your comment because our things are not in the recommendations that maybe should have been but maybe didn't get included because they just weren't part of the hearings or, you know, like you were saying, the whole issue of -- I wouldn't call it like you said, pay for performance, but it is pretty well-known that we don't -- our systems do not incentivize quality of care, that our incentive systems aren't built on quality of care. So, there is a way of making that kind of statement or making statements that, you know, using, you know, good evidence-based decision support tools on the provider and patient side would be useful, et cetera, et cetera.
So, my other question is -- or comment would be if there are things like that have been identified that we should put in there, but just haven't been in there, if they weren't discussed in the hearings or aren't part of it, should we still be putting them in thinking that this is something that the world may use as a document to kind of point to and say, ah ha, this is what NCHS does.
Then the third thing I was going to say is that if we do this stuff that -- by the time we finish today, come up with where does this go? What are the real steps, like who does it, where does it go, like does it go back to us, does it go to the full committee? I mean, I am trying to get very concrete because I want to get this thing out the door. So, I am just throwing all that out.
MR. HUNGATE: In response to your first one, I would like to put out the position that we make the report focus on what does it take to change from a cost driven to a quality driven system. You know, I would like to make a bold statement and say that if we are going to take care of the population's health the way we would like to. It has got to be a quality driven system and not just --
MR. ORTIZ: I agree. I think without getting us in trouble we should say what we believe, without saying something -- you know, within reason. We have to be politically sensitive, I know.
MS. GREENBERG: And that these are some of the things that need to be addressed. These will not stall -- that could be the context.
MR. HUNGATE: I would like to put a flag up there that says what we are really trying to do and then what is the recommendation in that context. Sure, they are not going to get all the way there, but you got to start someplace.
DR. MAYS: I think that if that is really where you want to start with the report, it is going to require, I think a whole different front end to it. I think it requires a lot more writing and work to it. So, I think that should be a part of it.
PARTICIPANT: [Comment off microphone.]
DR. MAYS: The front end has to be I think the set up for why that is true, what, you know, kind of -- the studies are that point to that, what the hearing has said. I mean, I think you really have about a good three or four pages that have to be written in the beginning --
MR. HUNGATE: There is a lot of testimony in that that talks about that. I read the whole testimony --
DR. MAYS: That has to be done and I think, you know, to give it a legitimacy.
The other point I will make, as I went through, just as we were talking now and looked at each of the recommendation numbers and looked -- you know, if you want to cluster and to say which ones, I think, at least are in the population. I started doing --
MR. HUNGATE: Great.
DR. MAYS: I am not good. I can't do the other committees but 1, 2, 5, 6 and 11 look like they are population based and probably need more work but -- and those are the ones that really do seem to be within our area. Having said that, if it comes to populations, I think what is really necessary is that it comes with a request very specifically of what you want.
I mean, we can work on the recommendations, but if you have something that you are going to do in the report and populations can help you with that information, we should know so that we do the work well. So, it may be that we really do need to struggle with what we want the report to be or if you just want, you know, each of these subcommittees to just work on the recommendations of -- you know, make the recommendations kind of, you know, whatever format you want then that is fine. But it is also the part that we want a different kind of way of life.
MR. HUNGATE: If I pick up on that, if I start with the goals and objectives at HHS, then one of those is improve the health of the population and part of that is reduce the disparities by improving the health of those who are not as healthy. I think that then fits to the population piece. That is what we want to say in that section.
Is that the kind of thing you are talking about?
DR. MAYS: I don't know. It is like whatever you want to tell us that is the goal of where you want to go so that when Population sits down with these recommendations, part of what we do is we understand how to contextualize employers and what to bring up as a rationale. I mean, if what we are doing is dealing with it because it is a strategic plan at HHS, that is one thing. If we are dealing with it because of health plans, that is something -- that is something different.
So, it is just there are many ways to spin the tale and we just want to know, you know, kind of what -- I mean, some direction as to how you see the report or else we can just do a recommendation and then you stick it in and then you do all the contextualization yourself.
MR. HUNGATE: Other comments? Do you have any brilliant conclusions from all this?
MR. STEINWACHS: We do need in the Quality Group I think, just as you are saying, to pick sort of a thematic piece. I guess what you had me thinking about was the objectives are to improve the health. As you say, what we are about, we are really looking about improving health services that serve populations and so it is within that context. So, we probably very quickly moved from that broad to something that says that disparities in health services become an issue as part of the disparities in populations.
You know, I think the other part of the framework is both from an accountability point of view, being able to monitor quality and from the other point of view to be able to provide incentives for --
MR. HUNGATE: See, all my experience is in quality improvement, as opposed to monitoring and tracking. So, I tend to want to think about what are the things that have to be in place for the people who do the work to do the improvement. And then how do you get the offshoots from that that enable you to do the tracking?
MR. STEINWACHS: You have to assume that someone is motivated to these groups. So, I mean, if one of the reasons you talk about incentives is why would you be motivated, then why --
MR. HUNGATE: You come back to accountability.
MR. STEINWACHS: It is a framework. I mean, you are sort of putting together a framework and these are interactive pieces that you have to be able to monitor. You have to have information that helps you intervene and improve -- so, I think maybe what we are all thinking is there needs to be a little bit of a conceptual framework put around this. I don't think too elaborate and we need to draw some links to the population issues, as well as to the issues within a practice.
DR. LUMPKIN: Well, I just had this random thought that we may be in danger. Our report and our recommendations really talk about ways to -- transactions and standards to be able to produce quality, improve quality. Issues such as putting race and ethnicity in -- 37, adding on preexisting conditions at time of admission, all of these sort of things.
The conceptual model is that we have got these other things going through. Why don't we piggyback the data and then what we will do is we cherry pick the data out of the individual transactions and create this quality database.
That is one approach. One could argue that another approach is to create a quality transaction and the way all these transactions are created is that at some point the data is input. There is the EDI engine that maps each one of these data elements to a field in the transaction standard and then it populates them. That is the same process instead of doing it -- you know, having it pass into all these different things and transmitted to where it comes in and collecting all of them, extracting, why not where at the stage in which you have the EDI engine, populate a quality transaction that then goes to the folks who are really concerned about it.
Even within the same health plans, there are folks who are concerned about paying the bills and folks who are concerned about quality. When you deal with very large institutes, their databases don't even mix.
[Multiple discussions.]
So then we could present a couple of alternative ways. One way is we could do this and if the folks feel -- you are going to get pushed back from the 837 folks, but here is another way.
MS. GREENBERG: Well, yes, because -- public health came to that conclusion with these hospital discharge systems, saying, you know, should we -- how much do we like hitting our head against the wall, trying to get some of these things that we need for our public health reporting into the claim or should we just come up with kind of a parallel transaction, a public health transaction that is populated by a lot of the same stuff that goes into the claim, but then there are some other things that are in the standard. You do have to get them into the standard in order to -- if there is going to be an X12 transaction at all, but that will just go into the public health line and then we are not fighting -- what we did with the race/ethnicity. I mean, we just -- you know, we took -- so now there is a transaction, which -- in fact, I have talked to the consumer disclosure people about this, but that public health transaction already has in it several of the things that they want, the second -- they indicated for secondary diagnoses, race/ethnicity and has the capacity through external code sets at NUBC, et cetera, to capture other of them as well.
It might be that some of those then eventually would migrate to the claim and everybody wanted them there, but you are not -- I mean, a lot, 80 percent or whatever is already in the claims and, so, it is just as you said, you just populate this other thing and then you get some other things that still have to be collected. I mean, it has to be available in order to populate the quality transaction or the public health transactions. If they are not collected, there is no way to populate it.
But you are not fighting as much about whether this should be in the claim and it is being designed not by people, whose major concern is paying the bills and at this point not paying on quality, but it is actually being designed by people whose major concern is quality.
MS. COLTIN: But, Marjorie, are people actually using it?
MS. GREENBERG: New York State is implementing it. Several other states are.
[Multiple discussions.]
MR. HUNGATE: To me that is not quality improvement.
MS. GREENBERG: If it isn't on the claim?
MR. HUNGATE: Well, what you are talking about, collecting separate public health information, which then goes off into a place and then reports back.
MS. GREENBERG: No, but I am just saying this is a -- would be a parallel process, what John is suggesting, that you would have a separate transaction. You are not trying to get this stuff so much into the HIPAA claim as a parallel transaction that is designed -- that goes right to the people who have to be responsible for quality. I am not saying public health should handle it. I am saying it is a parallel process.
MR. HUNGATE: The one that is responsible for quality is the one that delivers the care to the patient. That is the one who is responsible for quality.
MR. STEINWACHS: But they don't necessarily --
MR. HUNGATE: I am trying to live the quality improvement model, which is decision support.
MR. STEINWACHS: That could still be supported. This is a structural issue. That could still be supported by quality transactions where the billing transaction goes to the billing clerk.
MR. HUNGATE: It is a clinical transaction. If it is added information that doesn't help somebody get their work done, it is not valuable because it won't be done --
MS. GREENBERG: We are on another stream here. What we are saying is if we want -- if there is agreement that we should enhance administrative data with -- now, I mean, maybe you want to throw that out but I thought that is what we were talking about, enhancing administrative data for the quality assessment purpose.
One way as John said is to -- general stuff into the claim and then you pick out what you need. There is a million stuff in the claim that you don't need, frankly. It is just all sorts of stuff that has nothing to do really with quality. I mean, that is not -- you know, but it is needed because of our complex reimbursement system.
But another approach is to have a parallel transaction I was just saying we have already done that in public health. It could be done for quality and it could piggyback on the experience that we had doing it for public health.
MS. JAMES: All this would be based on the X12 standards.
MS. GREENBERG: Yes. Now, this is just -- this is if we are still talking about enhancing administrative data. Now, that is a different issue --
MS. COLTIN: -- that I feel compelled to share as an organization that would necessarily have to request that transaction in order to get that data. It is different when you have state law which -- when you are talking about hospital discharge abstract data, something like 48 of the 50 states have laws requiring hospitals to report this. If we built quality transactions similar to this, health plans would somehow have to now put that on the negotiating table with a whole lot of other things to try to negotiate with providers to submit those transactions.
It is not going to be even across the provider network, but you are not going to win every time.
MS. GREENBERG: That is the downside to it.
MS. COLTIN: That is the downside to it. The other thing is you then create a data stream with a lot of redundant data elements and you have to find a place to store it and create a database when you have already got a database that houses a lot of the claims information and you are talking at the margins of adding a few data elements to that. So, I just want us to be aware of what the up and down sides are to the different --
MR. ORTIZ: When I hear that, though, I don't see that that is the point of their conversation. The point is not that that is what we are going to do or even what they are advocating. This is a brainstorming, idea generating and we have to try and think about what are we going to do in the future. What are some of the topic areas. What is going to be our task for the next year.
I think this is important because that is what we need to do. We need ideas and they could come from the quality report. They could just come from our own creative minds. John has a creative mind. I would like to see something -- and I don't know how we want to do this. I have ideas on how to do it, but I would like to see us generate some ideas for our next topic area that we want to tackle, whether it is free flowing, drink some wine, tangential ideas that bounce off and we say, wow, we get a list of ten ideas or five ideas and John's idea might be a great idea or later on we may say we don't like it and we are not going to go after that.
But I think we need to do that and I am not sure what the best way is. I have ideas, like I said. But I am going to throw that out once again to you, Bob, the leader of the group now to figure out if there is a way that we can do that, whether it is brainstorming here, whether it is letting people go back to their places and send in ideas on the Internet and just send us a list, whether we can discuss it at the full committee meeting and let people -- you know, hey, you know, what are some ideas for the Quality Workgroup in terms of things to tackle?
I mean, that might not be something the committee wants to do. I mean, John knows that better than I do, but I would like to find some approach so we can decide and come up with a list of some feasible things that we would like to tackle and just brainstorming.
MS. GREENBERG: There is a time on the Executive Subcommittee agenda, though.
MR. STEINWACHS: It seems to me we are still sort of caught in the -- well, we have a history. Bob and I weren't part of the history. Kathy is. Thank goodness. Otherwise, we wouldn't have a history in the sense of what was done in this committee and we wanted to bring some closure to that and produce something that is part of the memory and where there was a basis for it to make recommendations that might help shape what you are talking about, which is -- I see as somewhat separate, but not totally of where do we go from here. What are the issues?
So, I still think we need to leave this meeting with a sense of how we bring closure to what has been important work, which has been a learning process and that can range, I think, from just the simple documentation and saying these are the needs we identified to something that does make very specific recommendations.
You know, we talked about how to re-tailor the one we have.
DR. LUMPKIN: Well, I actually didn't toss out that idea as a way to -- but it is a way to sort of organize our recommendations. We would have, let's say, two major recommendations, a recommendation that we enhance the current data stream, the current mechanisms, to collect the data that is important to be able to do quality purchasing and quality monitoring. This whole issue of quality improvement is just a totally different thing and I think that that is important and I think it is where the answer is, but that is not where the reality of the system is right now.
So, conceptually, we can say that we either enhance the data stream or we create a new data stream. Enhancement to the data stream would be collecting the data in currently existing transactions. The X12, 837 we have talked about. There may be others. Then we would list the alternative ways of doing that.
So, our recommendation isn't necessarily that we change the situational, you know, No. 2 would change from not use to required, just to say add the following data into the current transaction and we would list it. And we wouldn't be as proscriptive as we were and we wouldn't have as much pushback from the other members of the other committees.
Then we pose the second one.
MS. GREENBERG: Or create the capacity to get the system through the current transaction.
MS. COLTIN: I think that is the way to say it. Sometimes they aren't on the transaction, but it is the implementation guide --
DR. LUMPKIN: And then we list what it is, what we want. Then there are some natural conclusions that people are wanting to move that forward. Then the other thing is to create a separate data stream. I think to think about that, which would -- we have an essential structure, but there are some other pieces that we would assume would already be on the transaction. So, we would probably want to identify -- diagnose, you know, primary, secondary diagnosis. You know, there are few other demographics that would then help us define it.
We don't necessarily have to spend a lot of time going into detail but saying that there are things that we can use from the old transactions and then talk about how current technology would that accessible. And then what are the things that we believe need to be incorporated in development of new technology? The new technology is the electronic health records.
Then that is where we get to the quality improvement piece.
MR. HUNGATE: I don't think you have to wait for that. We are at odds on that because I think you can change the climate -- approval, by how you do accountability and we do that with the current information.
DR. LUMPKIN: I think where we are at odds because I think we may be --
MR. HUNGATE: I think we are.
DR. LUMPKIN: -- is that I think that -- kind of at odds, at odds in what the committee should do -- because you and I share, I think, a common --
MR. HUNGATE: I think we share the same motivation.
DR. LUMPKIN: -- the vision of going to quality. I don't believe that the purpose of this committee, of this workgroup, of this committee, is to set out an agenda for approval of quality in health care.
MS. GREENBERG: No, it is not.
DR. LUMPKIN: It is really to focus on what are the changes that we can make in data policy.
MR. HUNGATE: In data policy. Precisely.
DR. LUMPKIN: That will tend to enhance the ability to achieve quality.
MR. HUNGATE: That is what I am talking about.
To me, that is risk adjusted outcome for accountability. It is one of the key factors of information that will change the incentives and improve quality. It won't per se, but it changes the dynamics.
DR. LUMPKIN: I am with you there, but I am not sure -- when we talk about changing to the 837, which is in our recommendations, and other ways that data is currently transmitted after the clinical encounter, that can then be compiled into feedback of risk adjusted outcome data.
MR. STEINWACHS: -- so that idea that there is where we would like to get to, that requires interface of collecting information that goes outside the medical care system by itself.
MS. COLTIN: I heard you say that there were three options that we could look at. One was optimize the data stream from current transactions. One is creating new data streams to administrative transactions and the other was create new information sources, electronic information sources, which could be the personal health record, the electronic medical record, where you get the patient to tell you about the outcomes through the personal --
MS. GREENBERG: -- all three and you work on, you know, concurrently on all three of them.
MR. ORTIZ: When you say that, John, just sort of my understanding, are you saying that that is your recommendation and that becomes part of this or are you saying that is a recommendation that we need to explore further and maybe have hearings and do some stuff in the future. It is how you get there.
DR. LUMPKIN: Yes.
MS. JAMES: -- is the point. I am very much in Eduardo's camp and that is I am really focused today on where are we today and where will be tomorrow. How do we get out of this and what I hear you saying and also you saying, Eduardo, I think is take these recommendations that we have now or go forward under this quality banner -- and we talked about leaving it as it stands now or tweaking it somewhat, but basically not major changes, keeping it broad, dropping some of the detail, wrapping it up with some version of what you just said.
These are our recommendations. Now we are going to look forward and we are going to say there are three general ways in which one could implement or achieve these and they are A, B and C. And then leave it at that and then you also will have a stepping stone to subsequent activities.
DR. LUMPKIN: Well, I think that is sort of what I see as the -- okay, we take the report as we have it. We go into the recommendation section and we split it up into those three areas. There will be duplications because --
MS. GREENBERG: You want to actually split the things --
DR. LUMPKIN: Not necessarily. There are three options for which the same recommendations may appear under all three options. That is not true because there are some that talk about electronic health record, which would be in the third option and then we will say that these are the three options. That is the end of the report.
It may be that --
MS. GREENBERG: But you would put specific things under them?
DR. LUMPKIN: Well, no. I think that we would probably be much more specific under the expanded current data stream because that is really where most of our recommendations are. We drop the option, the conceptual model of a new data stream and then we talk about the new technology and we can then as a committee, as a workgroup, say, okay, if anybody is going to work on expanding the current data streams, it really ought to be the Standards and Security Workgroup. They may or may not take it.
But then we can decide which part or how much we want to spend in the future working on the new data stream or the new technology.
The report then becomes a transition point rather than an end point of the workgroup.
MS. GREENBERG: One thing about this new data stream idea or it is really a parallel data stream, because it is not -- it uses a lot of the stuff that is already in the current data stream. You said many of the elements are the same -- is that -- I mean, one thing that I could see that you might want to recommend or that could be useful is this wouldn't have to be forever. This could be like a pilot -- maybe a pilot project. If you could get some people to collect this and actually use it and demonstrate how it can be used, the value added. That might then either get lots of people doing it or make the case for migrating some of that into the current data stream.
But because right now -- I mean, like all of these idea about the vital signs and all that, I mean, it seems like a lot of data collection and people just aren't sure, well, how valuable would this be or would people really use it if they had it or would it just be collecting more information. So, what we really need, I think, is some demonstration projects with using it.
DR. LUMPKIN: I don't disagree with that, but let me go completely out of the box. If what we are talking about occurring, if we are looking at the transition and the effort of going to the National Health Information Infrastructure and all the polls are saying that the number of electronic health records of some sort are dramatically increasing, the biggest problem that we have had in quality, in looking at any of these data, is not the transmission of the data. It is the data collection.
So, if the data collection costs come off the table, thinking about this as a new transaction or moving it to the new -- or to another old transaction really becomes an immaterial argument.
MS. GREENBERG: You are saying if it is all captured in the electronic records.
DR. LUMPKIN: Once it is captured, how it is parsed and fed through to one place and another becomes immaterial.
MS. GREENBERG: But when do you think the majority or the vast majority of providers are going to actually have these electronic records?
DR. LUMPKIN: Well, the stake we have to put in the stand is when they are developing these electronic records. Do they capture or not capture we think is important.
MS. GREENBERG: Well, and the functionality that
-- to use it.
MS. COLTIN: If it is in the record, you can shake the computer but you can't get it out. It doesn't help.
MS. GREENBERG: I mean, that is what I am hearing from -- sure, that people's age is in there. Their gender is in there and even whether or not they have had a mammogram maybe should be -- you know, would be in -- and yet they can't seem to -- functionality to gather that.
DR. LUMPKIN: But the time, the cost of taking the legacy system and developing a new report is dramatically greater than when you are building a new system to say, oh, yeah, we are going to add this report in there. That is really what we were talking about trying to influence. There is a lot of work going on.
One of the grants that we are giving at the Robert Wood Johnson Foundation is the Kaiser to fund them to -- they are building an electronic health record to make sure that the new set of data standards for emerging quality that is the National Quality Form is built in there. It is kind of like a teaser.
Those kind of things -- what we want to do as a committee is influence people who are building systems, who are doing the quality work, to say, okay, what is it that I need to check off to make sure I didn't forget to include. I think it is here. I mean, this is -- that is what we have in those recommendations. If we can tell the story that way, I think it will be easier to get it through the full committee and accept it and it won't be a statement to the 837 committee, the reason why, you know, we have disparities in health care is because you guys won't turn the switch on on collecting race data.
Then they are going to push back and say, oh, no, it is not our fault. It is not our job. We just changed the conversation.
MS. COLTIN: If we go that route, we focused on data gaps and administrative data and, in fact, there may be information that is in the administrative data that isn't right now necessarily in most electronics medical records or captured in the same way so that we may need to take a look at that if we are going to go that way and it is just that it may be that there needs to be a recommendation around the electronic health record to make sure that something that might not be part of every electronic medical record system is captured because it is captured in the administrative.
I am just saying that we get the absolute worst claims data from the parts of our system that have electronics medical records and the reason we get it, the worst data, is that on a visit specific basis, they know all those other comorbidities are in the electronic records. So, they don't record them at that visit. So, the claim comes in with, you know, one diagnosis or possibly two if the patient had two problems that were dealt with on that visit.
It doesn't go back and pull up, you know, other comorbidities; whereas, people who are coding for claims tend to do that. So, your claims are not rich and you can't do good risk adjustment.
MR. HUNGATE: I argue that the claim has to be the document that is the interface between administrative judgment and clinical improvement and that it has to be a document that enables both to work without creating too much redundancy, which is why I come back to risk adjusted outcomes needing to be on the claim all the information necessary to risk adjust the outcome of that event has to be in the claim.
DR. LUMPKIN: I would argue that the issues in that redundancy is double data entry and there is a difference between the two. If you enter the data once and it is in five different transaction, that is redundant, but it is not double data entry. The cost of that is negligible. My argument would be is that if the claim data is bad, it is not -- and how the claims --
PARTICIPANT: [Comment off microphone.]
DR. LUMPKIN: -- of EHR was set up --
MS. COLTIN: But that is a functionality issue and it is not on the table.
DR. LUMPKIN: What we are talking about is functionality. So, I think as we are writing it up, I would think that, again, trying to get this report out, that would be one of the items that would be listed under Option 2, which is the new data screen, of issues that have to be addressed in developing this new one is to assure that data that is currently in the claim is adequately calculated in this new record.
It doesn't have to be any more specific. That will be the task.
MR. HUNGATE: If we could provide some samples of things that are often missing that are needed.
DR. MAYS: I want to ask the question -- I want to go back to population health care. I want to make sure I understand in what you are proposing -- I understand the intent of what you are proposing, enhancement in terms of quality of the data collected. Does it go the next step and present a compelling argument or in any way, you know, kind of outline the urgency of what needs to be done with the data at all?
Because it is like we can build it and they can collect it, but I don't know that know that we are saying anything because part of the quality is not just, okay, well, I collected it, so the quality also has to be something about the reason for using the data at a population level and the way we had it before, I think that that would be there. The way we are going now, I am not sure it is going to be there.
So, I am trying to, you know, be the population hat for now and say, okay, collecting is very important, but we know if you read the report that -- oh, it is a report from your committee, for example, Nicole Lorey's(?) paper, Freeman and Louie's(?) paper, where they talk about people don't want to analyze or look at the data for fear they might get sued. They are even afraid about collecting this data because of that.
So, what is the extra thing that we can do to try and say to collect it and use it and what compelling reasons are there at a population level, not just the evaluation, for example, of, you know, at individual, you know, encounters, but, instead --
MR. HUNGATE: I argue that accountability is a population issue, that it is accountability for the management of all diabetics. How do diabetics do in this environment? How do those with chronic lung disease or congestive heart failure -- you know, when I say accountability I am thinking populations.
MR. LOCALIO: But at what level? Is that a plan or a state or a county or --
MR. HUNGATE: The only way it works is at provider.
MR. LOCALIO: Oh, okay. We are talking about different things.
MR. HUNGATE: You know, the only place that a patient goes for care is at a provider. Plan doesn't really do that.
MS. COLTIN: -- with the fact of how many patients go to different providers and there is no coordination of care among them and then how do you deal with that and, you know, can you deal with that at the level of a practice group or a care system or even a health plan that tries to plug those goals with --
The patient went to all those different providers for their reasons, you know, and they had a reason.
That is what I am -- I focus on the patient in saying we have got to make the quality system work at that point and I think that drives you to population.
DR. LUMPKIN: If I can respond to Vickie's concern, I think we need to, if we are going to do this modification of the report, put that on the shelf and come back to it because as I can customize the report, it says there is -- we are talking about all these needs to improve quality. Are there ways to improve quality, that analyzing data is a way to get to quality and we can't do it because we don't have anything. So, I think the report answers the question because it starts out there.
But we need to look at it from the perspective of your question to make sure that because we all assume that, that we didn't say it and I think that that is one way to -- you know, if we can just sort of remember that and then come back and, okay, did we actually say it enough in the report so that someone won't get to come to the conclusion that, well, just collecting the data is going to --
I have one other piece about that second -- came on my disease registry, which are the thing in care management. How do you populate a disease registry with this kind of transaction? Then we do that on the public health side. The immunization registry, that is how the immunization transaction was developed in order immunization registry. Some states and I know they are starting to look at this for cancer registries and populate those with a particular data transaction. So, we can really draw some analogies.
MR. HUNGATE: The advice I have been given from the people in Massachusetts is to speak in terms of virtual data registries as opposed to true data registries because there is great possessiveness on the part of many people of their data.
You can visualize distributed registries, which become virtual registries, as opposed to a registry off of the public health department.
MR. STEINWACHS: Maybe we should talk about -- virtual sometimes makes --
MR. HUNGATE: Well, virtual memory is a common term in the information world.
MR. STEINWACHS: But registries also require data linkages. So, if it is virtual, you aren't sure that those are occurring.
MS. COLTIN: I think the concept that Bob is referring to is that if the data are all linked at the patient level, if you have a standard identifier, the data may be housed in different sites and you create a record on the slide but you can say I am going to put in this identifier and I want to see their history of X. It goes out to all these separate systems that it is connected with and it constructs the record and brings it up on your screen. That was, I think, the concept.
MS. GREENBERG: Do we have a plan here?
MR. HUNGATE: I think the three categories plan is a plan. I think that is appropriate. I would be willing to say that in the discussions tomorrow.
MS. GREENBERG: What I am not sure about, though, is if you are saying -- the way Gail interpreted it was you raised this, you identify and put the content -- you identify the sufficiencies of the areas where information is needed and then you say these have three ways that they could be addressed and then some combination of -- may need to be working on them concurrently and a committee will be addressing them or something and we challenge the industry to do so.
I thought I heard you say you would actually take these data needs and --
PARTICIPANT: And the three options.
MS. GREENBERG: Yes, and the three options and some of them you would put in all three or would you distribute the data that way or would you just identify the data requirements and then they would say these are three approaches that need to be explored?
DR. LUMPKIN: I think I would sort through these because there are some we need to talk about. I think we need to talk about the ICD-10. There are some that fit, you know, for instance, add the EHR content. That clearly is a new technology approach. There are others which are the genesis of changing and not used. There is a collection of race and ethnicity data and that needs to be in all three.
MS. GREENBERG: Let me just say about ICD-10 that we heard what Jim said and if people feel that it is just fine that we don't work on the 10 for maybe another five or ten years, then I don't think it is going to help the quality agenda. So, I don't think you should say, well, now that one is taken care of. If you feel it is important to quality, then it needs to be given a little -- you know, all the letter said this morning was a minimum of two years.
MS. JAMES: Well, somebody, and maybe it was you, Marjorie, said that with these issues that have been at least at some level addressed, like ICD-10, like the MRI kind of things, that we can handle them in a slightly different way and perhaps pick them out of our explicit laundry list, but then still take this opportunity to say the workgroup supports these.
MS. GREENBERG: And that would, of course, go into the regular transaction because that is what it is going to be recommended for.
MS. COLTIN: That is true about the -- I mean, Simon's comment and the comment thing about the standard provider identifier, you know, that has been done. You know, the standard -- the recommendations went in and the NPRMs and --
MS. GREENBERG: We don't have a final rule.
MS. COLTIN: But it isn't there. It hasn't been billed. So, the same thing can happen with ICD-9 and then the ICD-10 is that --
MS. GREENBERG: -- language.
MS. COLTIN: It can take forever and ever and what we are talking about is the implementation.
MR. HUNGATE: It would not be inappropriate to say that there are these things that can help quality but they are not here.
MS. GREENBERG: Somebody has got to say do we really want to collect data at the exquisite level of granularity that SNOMED allows and then snap us back to ICDM? You know, that is ridiculous if you think about it.
DR. LUMPKIN: I don't think we are going to do that in this report.
MS. GREENBERG: No, but I mean -- if you are talking about administering data, you would want presumably to go to a newer -- you know, there is a disconnect there. There really is.
MS. COLTIN: The one thing that is in here in a couple of places that I am not sure where to put in the three bucket framework that we came up with is some of the survey-based recommendations and one could stretch and put them in the new data streams bucket, either, you know, in connection with the notion of personal health record or, you know, other ways of cost effective means of collecting data, you know, on-line surveys and other types of new technologies, but, you know, then it becomes new
technology --
DR. LUMPKIN: I would put it in both of the buckets, in No. 2 and No. 3, which is a new data stream because it is -- I mean, no one is really thinking about grafting it onto administrative data. Currently it is collected as a separate data stream.
MS. COLTIN: From a totally different source --
DR. LUMPKIN: That kind of data could be collected as an exit survey and appended to -- as a claims attachment. I am just trying to make a real wide consequence -- I don't think any of us would suggest that but currently with as a separate data stream in this new technology, I think surveys fit.
MR. HUNGATE: Okay. Do the report people know what to do?
MS. COLTIN: Well, I think the reframing of the recommendation is something I can certainly take on and do, you know, consistent with what -- the recommendations that move from recommendations into Section 5 or others, I think, would be appropriate for Gail and she and I need to coordinate on which those are.
PARTICIPANT: Which Gail?
MS. COLTIN: This would be the section on complementary initiatives. Some of those fit in that section. There are other things that are going on, either within other subcommittees or out in the environment, you know, the marketplace.
Then the other piece becomes going back and reworking some of the front end to bring in the pay for performance concept, the other things that we talked about up front, as opposed to -- you know, they have sort of been there a little bit, but they are not highlighted in some way. I would turn to Susan --
MS. KANAAN: I don't know if I can do that actually. I have been asking myself that question. I don't know that I know enough. I would be very happy to work with somebody on it or several somebodies.
MS. GREENBERG: Actually, I was thinking that Eduardo could -- we could work together on that.
MR. STEINWACHS: If you are building that section, can't you also talk about things like HEDIS and the National Health Quality Report, all those things that draw on --
MS. GREENBERG: Or were those in Section 5?
[Multiple discussions.]
MR. STEINWACHS: The framework that it talks to the accountability and mechanisms that may sharpen incentives.
MR. HUNGATE: I want to stick on the quality improvement model as being the central, long run core of where we got to get to. NHII is critical to us. But there are some things we can do between now and then that help that transition. I think we have got to get that in there because I think that is where we have as a committee the most room to do some things.
Now, John has got to give the approval for that.
DR. LUMPKIN: I only have one vote.
MR. HUNGATE: I can't operate, you know, as a sole operator.
MR. STEINWACHS: You are posing a framework. If there is a monitoring sort of framework, there are ways to enable organizations to create incentives by paying for quality or allowing consumers to choose quality. I think if we are going to introduce a quality improvement paradigm, then there has to be a write-up about what that is and how -- what data resources are needed to enable a quality improvement initiative.
I think if you took a crack at that -- I mean, we aren't talking about a long intro, but we are talking about an intro that tries to pull together what it is that we are trying to enable through this data structure.
MS. GREENBERG: Does Don Berwick -- he is a guru of quality improvement or --
MS. COLTIN: There has actually been a couple of papers written on that and --
[Multiple discussions.]
I wrote one also that was in one of the old TPR(?) publications. So, I think that background material is --
MR. STEINWACHS: -- is working very hard at this.
MS. COLTIN: There are materials that people could turn to -- what are the differences sometimes in the data that you need to do quality assessment, to do quality improvement? The data you need to do quality improvement are data within processes of care. You know, it is not simply did Mrs. Jones, who should have gotten a mammogram, get one, but it is why didn't Mrs. Jones get a mammogram. Did she get the referral, but she didn't go? Did she go but the results didn't get reported? Did the results get reported but it didn't make it into the record?
You know what I mean? What are all those process change kind of things that resulted in the fact that there is no mammogram? From the improvement perspective, you really need to look at all those little links in the chain to say where is the chain broken. Is it, you know, the referrals are getting made, the patients are getting told to go and they are not going? Or is it more complex than that? They tried to make an appointment and the waiting time is eight months and so they say I will call back and when they do it is nine months.
So, those kinds of data are the data you need for quality improvement and I don't honestly think that what we have recommended here gets us to that.
MR. HUNGATE: The point that I am trying to make is that you can't do a really good job of improving quality until you have an electronic health record, until you have the depth of information that you need to have when you need it. That is why Intermountain Health is able to do so much more than so many other people because they really do have the data where they need it.
Don Berwick in his early work did not argue for public accountability. As near as I can tell, he has changed his argument. He now does because he recognizes that the absence of public accountability at a population level for outcome means there is no incentive to improve. That is the quality agenda. That is the part that I think the claims information can facilitate that in an effective way. I think that argument has got to be in there.
That is more what I mean. Pay for performance is the quality improvement model for making the system improve.
[Multiple discussions.]
There are others but --
MS. COLTIN: Consumer decision-making and public disclosure with or without pay for performance.
MR. HUNGATE: Which is why I have argued for enhancing the ability to give individually a tailored list of death codes, which we talked about at an earlier brainstorming session.
MR. STEINWACHS: Okay. We have got a working team of now four, five.
MS. COLTIN: I think we can all provide information but I think we need to agree on some framework that we can use.
MS. KANAAN: And we have had a very decentralized decision-making and communication network. I think at some point, not necessarily today, we need to clarify where the buck stops or, you know, how these different pieces fit together.
MR. HUNGATE: Let's talk about this process a little bit because we would like to finish this in March, but I am not sure that we can. I think we ought to, we need to, because we need to get by it.
MS. KANAAN: I really need to have a deadline for mine. If we don't need it, we don't need it.
MR. HUNGATE: What is it going to take to do that?
MR. ORTIZ: Can we finish it by the end of the ICDM-10 implementation? Or we can try to go parallel.
MS. COLTIN: I think what we do is we aim for that March meeting and if that means we have to stop at the point of saying there are these three avenues that we can explore without fleshing those three avenues out to any degree. If that is all we can get done by then, then that is where we start.
MR. STEINWACHS: I was trying to see if there was a deadline where you could get like with --.
MR. HUNGATE: Let me try to articulate what I think I ought to lay out tomorrow morning. I think I ought to say we want to have a report that is ready for approval by the full committee at the March meeting. That is what we want to do. In order to do that, I think we need to make a new draft that organizes in the three categories and then this goes to the other subcommittees. I think we need to get the report out in a first draft in January probably.
DR. LUMPKIN: I would figure with our new structure we don't need the other committees because we have changed on this granularity recommendation and because we are not talking about specific changes in 837, we are just saying this is an option to do administrative -- that removes that problem.
MS. GREENBERG: And you are identifying the problems that you -- you are describing the problems that were identified and they can argue with your solution, but people can't really argue with the fact that this is what you heard. Now, the priorities and all that, but that is where it sort of is now.
DR. MAYS: I was going to say, but the one issue is the reason that we send it to the group is so that if there are some significant concerns or issues, et cetera, it resolves that and then it gets sent back. So, if you wait until it is very close to the meeting and if there are any problems, then you have a problem of not getting it through in March.
MS. GREENBERG: Getting it through the Exec Subcommittee.
DR. MAYS: No, no, no. That is what John said. We will do that, but what I am just saying is the timing of doing it. See, they were going to do it in January to get feedback, so if they wait and do it a couple weeks before in March, if there are any issues, they may not make the March meeting. If you think -- you know, because remember -- what is different about this is that the expertise has resided in a whole committee. So, the whole committee has heard everything. So, I think everybody has been --
[Multiple discussions.]
I was just saying you want to give yourself a little more time.
MS. COLTIN: And I think it needs to be sent out return receipt with a notice because if we don't hear back from you by --
[Multiple discussions.]
MR. HUNGATE: I would like to have a committee meeting of this committee in January and I would like to have a new report to discuss at that time.
MS. GREENBERG: We are talking about having potentially a full committee meeting now, adding one on at the end of January.
MR. HUNGATE: We could do this in conjunction with that.
DR. LUMPKIN: How long do you need for a committee meeting?
MR. HUNGATE: A half a day should be -- I think we could do it. I think this committee has to approve the report. You know, we have got to do the work now. I think we have internalized enough of the content from this process so far, that we should regard ourselves as the body who is going to say we think this is a good report and we have a 90 percent probability that the full committee is going to agree that we have done our homework well enough.
MR. ORTIZ: So, I am going to propose -- I am going to form a voting mechanism that we -- that Susan and I will go and work on the background, that Kathy will work on redoing the stuff that wants to redo, that she has got all these notes and all these recommendations. Gail, you have a section that you guys can work together on. Bob can work with us wherever --
[Multiple discussions.]
We will have this thing out ready for review for the January whatever meeting, the 29th meeting.
DR. LUMPKIN: Then we just add four hours up front of the full committee meeting, depending on when that starts.
MS. GREENBERG: You mean if the subcommittee and the -- they aren't having two full days?
DR. LUMPKIN: Well, but there has been really no significant over -- I mean, there is maybe a little bit, but there is not much of an overlap because they are meeting upstairs and we are all down here.
MS. GREENBERG: What you are saying is you could start a meeting -- I got it.
MS. KANAAN: So, you are saying it may be the 28th, not the 29th?
MS. GREENBERG: It could be.
MR. HUNGATE: It could be the afternoon of the 28th.
DR. LUMPKIN: You have to make choices in life.
MS. GREENBERG: I made a choice.
MR. STEINWACHS: Marjorie, do you need decision support?
MR. HUNGATE: Afternoon of the 29th also works.
DR. LUMPKIN: No. The afternoon of the 29th is a Friday.
MS. GREENBERG: No, it is Thursday.
DR. LUMPKIN: I just would not ask people to stay here. Yes, the afternoon of the 29th could work.
MS. JACKSON: I thought the 29th was the full --
MS. GREENBERG: It is, but it is only a half day, that morning --
DR. LUMPKIN: So, we could do the morning and then meet in the afternoon. I am just thinking of -- if it were a Friday, I wouldn't --
MS. COLTIN: So then you are saying after this committee goes through that at the end of January, then there will be another round of edits.
MR. HUNGATE: Yes, we will sure do some editing work.
MS. COLTIN: And at that point, we want to circulate it and ask for feedback. I don't think even if you do that, that you are necessarily going to avoid being
-- I mean, it is just the way it does --
MS. GREENBERG: But I think if the recommendations go up a level and they have these different options, a lot of the concern will go away.
MR. HUNGATE: So, I think that we will probably -- let's talk about -- I think that is the plan for getting the report. Now, we have got some time tomorrow morning. What do you want me to do with that time tomorrow morning?
MS. COLTIN: Can I just bring up one other thing? One of the big recommendations that came through in the comments and in the last meeting was we are making a lot of these recommendations, but we haven't gotten input on the recommendations. We have gotten input on the problems, what are the problems, but not on the recommendations, broad input from the field. So, there was a suggestion -- I think it was Simon's, somebody's -- that once we have this set of recommendations, we should really hear from those who have to supply this data or those who have to figure out -- [Multiple discussions.]
-- is to just leave them that way and not get that input? But we are still saying that these data elements should come on administrative. We are not saying how just somehow --
MR. ORTIZ: That is what you would like the ICD-10's letter this morning --
MS. GREENBERG: We have had testimony --
MR. ORTIZ: We have had testimony and --
MS. GREENBERG: I mean, that might be the first thing you would do with the report. Do the report. Get it through the committee and then maybe have a hearing to discuss the report.
MR. HUNGATE: What my inclination is to do --
MS. GREENBERG: But if you build that into doing the report, we won't see this report --
MR. HUNGATE: I know there is a discussion that needs to happen about the different -- the claims attachment versus change in claims forms and I think that needs to take place at the full committee. I have heard it from a couple people.
I think we ought to use some of the time tomorrow to --
MS. COLTIN: I am not sure, given where we are now -- relevant to saying you have to get into that discussion tomorrow. I think you are right if we kept it the way it is.
MR. HUNGATE: Can we in the next two hours come up with a moved up level of -- can we try something before the committee --
MS. COLTIN: I think what we are talking is incorporating race and ethnicity data into both enrollee reported and provider reported administrative trends without saying which of those provider reported transactions it would be. So, it could be a claim attachment. It could be a quality -- a new quality attachment. It could be the 837, but we are not making that recommendation. What we are saying is we need to know what the provider thinks is the race and ethnicity, as well as knowing what the individual thinks is their race and ethnicity because both factors are in play in terms of --
DR. LUMPKIN: I think that we don't want to have
-- if we have a change tomorrow, we don't want to have a discussion with the full committee that is going to lock you into something. So, I think that tomorrow morning a presentation of the conceptual model and a description of the process and if we end up not using the full hour and a half, which --
MR. HUNGATE: We shouldn't with that --
DR. LUMPKIN: And then we won't. It will be much shorter. Everybody will like the Quality Committee because it means I get done earlier.
MS. GREENBERG: We could move everything up except the National Child Health Longitudinal Study and then --
MS. COLTIN: I had put together a draft presentation when we thought we were going to go to the -- that I think actually worked pretty well because it doesn't deal with the recommendations. It deals with the problems.
I wish I had made copies for everybody, but it basically -- I have it on diskette and will work on it some more to get this new stuff at the end part of it.
But it begins with sort of what was the charge to the committee and what approach did we take and who did we hear from. Then it goes into the background that is in the report. Poor quality has heavy costs and it talks a little -- and then it says accountability is central to quality improvement and it talks about the role of data and you can't manage what you can't measure.
Then it is not the same old financing and reimbursement model and it talks about pay for performance and you can benefit. Then it talks about how fair IT adoption and what is happening out there that could facilitate --
MR. ORTIZ: I guess the only thing is that I don't know if this group needs to hear all that because I think they already know --
MS. COLTIN: I am not sure everybody read the report.
MR. ORTIZ: You are setting a lot of the stage for why this is important, preaching to a group that knows very well that this is important. So, I don't know if you have to get to that level of detail versus what John was talking about, what did we come up with in terms of our ideas today, our conceptual models, how we are going to go forward and letting them hear that because some of this stuff is -- yes, I know that is important but I don't think the committee needs to hear this.
MS. GREENBERG: Because that is going to go probably in the introduction to the report.
[Multiple discussions.]
MS. COLTIN: I thought you wanted to change that context.
DR. LUMPKIN: The problem with doing that tomorrow is the same problem that we have had with the work product of this committee, which is we spread it over such a long period of time. We have lost members. We are going to lose members again. We are going to have new members. And I think that it will not stick. Doing that walk through of the conceptual model as the presentation in March I think would be a useful time.
MS. GREENBERG: But I do think that it would be appropriate for you to report on this exercise because we ask people to do it. Not everybody who did it is here right now. I don't mean to just hand this out, but talk about how many responded --
[Multiple discussions.]
-- in some regards but there was a consensus in some others. You know, and if you looked at all this, but the facts really -- that you appreciate the people -- because like Simon, he spent a lot of time, a lot of comments and everything. So, I think the worst thing is to do that and then get no feedback.
So, I think that it is important -- it was a new thing. We had never quite done anything like it before. So, I think it requires -- I think you need to report back and you can decide whatever you want to say about, you know, this. That then led to this new kind of decision as to the way forward.
So, they won't feel like they just did it and that was worthless.
MR. HUNGATE: Any other suggestions, comments?
MS. JACKSON: If you need copies of the survey itself --
MS. GREENBERG: No, I don't think so.
I want to thank, of course, Bob and Kathy and Susan, but also Eduardo and Debbie for the work they did on, you know, shepherding the survey and compiling the results and everything. I know she was there several nights late and I know the two of you were e-mailing back and forth. So, I appreciate that because it was non-trivial. But I think it took us to the next step.
MR. HUNGATE: Okay. So, is everybody clear on the afternoon of the 29th as a scheduled time?
[Multiple discussions.]
MS. GREENBERG: Oh, you think the 28th is better for you? All right.
MR. HUNGATE: Debbie, why don't you circularize this and --
MS. GREENBERG: Yes, we will try to finalize that. So, you may not be able to come the 29th at all?
MR. STEINWACHS: I may not.
MS. GREENBERG: Or you might -- if we needed like a vote, you might be able to get on the phone?
[Multiple discussions.]
Now, if we are going to have this quality meeting, maybe you will want to come in, Vickie.
DR. MAYS: Well, we were just talking and I was -- you know, we have a report we have to write. So, I was trying to figure if he wants the full committee here, if John is saying the full committee is here, I would see whether or not we could also work around it
MS. GREENBERG: The more reasons to come in, the more reasons we would bring people in. If it were just strictly for this, you know, EHI thing, I think people who aren't on the Subcommittee on Standards, we could say, you know, if you are in the neighborhood, come by; otherwise, we will get you on the phone. But on the other hand, this could be a vehicle for -- it will break the budget, but it is okay.
MR. STEINWACHS: You don't have a budget yet, do you?
MS. GREENBERG: Although we are not supposed to agree to anything that we haven't done in the past and we have never had five meetings, but that is okay.
MR. STEINWACHS: Well, at least they are during the same period of time, right?
MS. GREENBERG: We still have four other meetings of the full committee.
MR. HUNGATE: All right. Thank you. I think we are adjourned.
[Whereupon, at 5:03 p.m., the meeting was concluded.]