[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
WORKGROUP ON QUALITY
September 2, 2004
Hubert H. Humphrey Building
200 Independence Avenue, S.W.
Washington, D.C. 20201
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
TABLE OF CONTENTS
- Call to Order, Review of Agenda – Mr. Hungate
- Discuss Role of Population Committee for the Next Upcoming Four Recommendations – Subcommittee
P R O C E E D I N G S [8:44 a.m.]
MR. HUNGATE: Let me call the meeting to order, we’re a little short on attendees at this point but we have two beneficial guests with us so let’s go forward. And we are being recorded, I’m Bob Hungate, Physician Patient Partnerships for Health, and chair of the Workgroup, and let’s go around and introduce ourselves.
MS. POKER: Anna Poker from AHRQ, staff lead to the Quality Subcommittee and staff on NHII.
DR. CARR: I’m Justine Carr, physician director and health care quality of Beth Israel Deaconess Medical Center in Boston.
MR. RODY(?): Dan Rody, American Health Information Management Association.
MR. DIAMOND: Lou Diamond, the Nextech(?) Group.
MR. HUNGATE: Okay, do we have the telephone hooked up live? Well, you’re just in time to introduce yourselves, we’ve called ourselves to order —
MS. GREENBERG: I’m Marjorie Greenberg, National Center for Health Statistics, CDC, and executive secretary to the committee and trying to be two places at once.
MS. JACKSON: Debbie Jackson, committee staff.
MR. HUNGATE: Okay, we started talking and I think we should pick up just where we were, the things I want to make sure we get through today are the planning for the next hearing which includes taking and picking up from the last one and seeing what things we need to do to make sure that we get to most on the next one, and we’ve gotten into a discussion here about an approach that I think we should continue and build on. Anna and I went to the Populations Subcommittee meeting yesterday to see basically how they wanted to approach another three of the candidate recommendations and we need to talk about that for a little bit.
And I don’t think there needs to be much time spent on the executive committee retreat, Marjorie do you think there’s anything that’s especially germane to this workgroup from the retreat? I couldn’t think of anything —
MS. GREENBERG: Well, the one issue though I don’t think any of the other groups have addressed it either, but is that each of the subcommittees and workgroups was asked to consider items for a research agenda and that certainly, we talked about the need to kind of pull together research recommendations, some of which have been already made by different subcommittees and workgroups and sort of scattered, and that often they’re the recommendations that don’t get picked up on, or don’t get followed up on, and yet they’re very important. And so we talked about trying to kind of bring those together so as a research agenda and include them in the next annual report, or biannual report, which will be on 2002/2003. So my thinking is John’s going to bring that up and that would probably be done, needs to be done by the groups between now and the end of the year, or the November meeting, at the November meeting, but I don’t think we need to discuss that now, just to mention that that’s something we’re going to do.
MS. POKER: Marjorie, that’s for the NHII recommendations on research, right?
MS. GREENBERG: No, no, no, that’s for all subcommittees and workgroups. Maybe NHII has specifically mentioned the need for a research agenda for NHII but there are populations issues, there are quality issues, and some of them dovetail with NHII but the standards, there’s a lot of areas.
MR. HUNGATE: Okay, good, Justine has pointed out that we need to spend some time and I agree with her thinking about the last hearing and using that information to populate our discussions in the next hearing in effect, and why don’t you pick up on that again Justine because it seems to me that part of what Lola is saying might be responsive to what you’re saying.
DR. CARR: I think our last hearing was on the first seven candidate recommendations, eight, sorry, and with an eye toward including, I wonder if we have a copy of them —
MR. HUNGATE: I do.
DR. CARR: I’ll refer to them out of sequence because I don’t have them memorized. One was including a data element on the claims form that would tell us something about quality and I think that stimulated a lot of discussion because when we heard from the groups the concept of what they wanted was many data elements that could be then building blocks toward any particular quality question. But Harry Reynolds from Blue Cross expressed serious concern about whether the feasibility of getting these data, of managing the data elements on the forms because they’re trying to streamline the processing now and trying to meet deadlines of turnaround and so if a data element had to be a part of that claim form then what, it has the potential for slowing that down.
A second concern was how does the information populate that form so that if blood pressure, which is one of the candidate recommendations, or a vital sign goes on there, where does that come from, is it the, is it chart review, is someone going to do chart review, is a nurse, clinical assistant supposed to then fill out a form that then goes to the billing office that then includes a blood pressure, and then when we heard from some of the groups they didn’t want just one blood pressure, they wanted all the blood pressures. So I think that we open Pandora’s Box, I think our initial thought was if you put a hemoglobin A1C on there you would have some data elements over time that maybe folks could say the patient is in good or not so good diabetes control, but to tell you the truth we’ve been looking at A1Cs in people with high sugars and they don’t even have diabetes and I don’t know what to make of their A1Cs, so I would even withdraw that —
MR. HUNGATE: That’s an interesting —
DR. CARR: Yeah, so the way we backed into this conversation was Lou was saying that their approach is to take, to have a disease state and to have a grid that is, I forget data element and utility of the element, kind of like what we did early on, easy to do, hard to do, informative, very informative, and think about that. But I would just like to step back and say what are we embarking on if we are pursuing this idea of having this field on a claim form because we’ve already heard from Harry pretty clearly that payers would not want it. We heard from Stan Hotchberg(?) that some challenges at the local level of getting this, and I don’t know that we have a clear idea, and we’ve heard from the people who might want it that they want way more then one data element. So what I said at the other meeting, I don’t mean to be dramatic, but I would withdraw that as a recommendation because I feel like it will create a huge amount of work and that there’s not a clear end point or value added and it just opens the door for meddlesome requirements that at the end of the day could contribute to chaos and not achieve the goal that we’re heading for.
MS. GREENBERG: Well, my understanding was we were going to get a summary of that hearing, which you did but I guess we —
MS. POKER: We didn’t, it’s not the agenda to discuss, I mean I didn’t think —
MS. GREENBERG: But then we were going to have the second hearing and explore these issues with these groups and then have some discussion about whether we want to withdraw all of them or take any of them forward as recommendations, from candidate recommendations to recommendations that the group would make to the full committee. So I think, I guess my sense is well, first of all, there are only two members here but also it’s a little premature to have, I’m not quite sure why we’re having this discussion right now.
DR. CARR: Well just so we that ask the right questions at the next hearing, to build on what we’ve learned and to fine tune as we ask for presentations for our next things, let’s get a little more traction specifically on these things, what would be the burden of doing these and what would be the value added —
MS. GREENBERG: I think that’s what we’re going to explore with the people later in the month right?
MS. POKER: Okay, I just wanted to ask you something Justine before you came so maybe synthesize what we heard at the next meeting, do we want to do that, because we’re changing the agenda now, do we want to —
DR. CARR: I’m not saying how the meeting should be conducted but for my thought process I want to be saying of all the things I heard in the first thing I now, these are high priority, these are low priority, and then when we hear from the next group there may be details that we didn’t learn enough about at the first meeting that we want to be sure to cover at the second meeting. So that may not be a public thing but it may be our internal agenda that we work from —
MS. POKER: Do you want to do that kind of review, what we did in the first hearing? Or you want to leave it all towards the end to review? Because I’m just looking, if we’re going to be switching the agenda where to put what and how to accommodate that.
MS. GREENBERG: Well, you did your summary —
MS. POKER: Synthesis, yeah.
MS. GREENBERG: — and people commented on it —
MS. POKER: Yeah, I got feedback.
MS. GREENBERG: Did you send that summary to the participants? I mean to the people who are going to be on the agenda?
MS. POKER: Yeah, they were sent, no, not the summary, they were sent, no, they did not get the summary —
MS. GREENBERG: You sent them questions —
MS. POKER: I just sent them the questions of the first panelists, they got the quality report, they got their questions too but they also in retrospect go the questions of the first panelists.
MS. GREENBERG: And I don’t know if we, they should have the minutes from that meeting, we do have the minutes?
MS. POKER: Yeah, we have the minutes from the last —
MS. GREENBERG: Have they gone out for review yet?
MR. HUNGATE: We have transcripts.
MS. GREENBERG: We have the transcript but do we have the minutes, so maybe you can check into that because that meeting was in June, the end of June, I think we should have minutes by now.
But in any event I think it would be very appropriate for Bob to give a brief summary at the beginning —
MR. HUNGATE: That’s what I was afraid of, I had trouble coming to a summary of what we did last time, I couldn’t —
MS. POKER: I have the synthesis —
MR. HUNGATE: Well, I haven’t seen that yet. I’ve been in Northern Idaho for the last two weeks.
MS. GREENBERG: But I thought you did that about a month and a half ago —
MS. POKER: I did, I did a long time ago —
MS. GREENBERG: It was right after the meeting.
MS. POKER: I sent it out pretty quick.
MR. HUNGATE: I lost about five days of email in my inbox overflow from the previous trip, I don’t use a Blackberry, and so I think I lost a piece of information, so I can work from that. But picking up on Justine’s concern, which I think it’s valid, and listening to what Lou said and I know Dan has expressed some worries about where we’re going on this and so I think that we can use their inputs to help frame this next discussion, and it may be that there’s something that we should hear from MEDSTAT with respect to what they’ve done in the disease centric measurement, that matrix sounds to me like it might be useful.
MS. GREENBERG: I’m sorry, I missed that —
MR. HUNGATE: So let’s go back to that discussion for a couple minutes, if I could call on our two consultants here.
MR. DIAMOND: Happen to show up and therefore we’re consultants.
MR. HUNGATE: Interest has an interesting way —
MR. DIAMOND: Well our approach to dealing with the issue at hand is to ask the question what is the purpose of the data that we want, and one of the things we landed on is we need to populate an evidence based quality measurement system to support public reporting, accountability, patient choice, and internal quality improvement. So the starting point at one level is a clinical performance measurement set, so you’ve got a set of measures on the vertical axis if you will which in this case could start with disease specific measurement sets, and I’m talking across all settings now, and then you across the vertical axis you display cells that attempt to define the various data sources in aggregate descriptions, chemistry, pharmacy, microbiology, radiology, etc., etc., vital signs, I mean we’ve got about 20 or 30 cells. And then you populate the cells and you end up with a matrix that at least answers a question to start with which is what data do you need to populate a quality performance measurement set.
You can then secondly ask yourself the question so now we know what the data source is, pick the data source blood pressure, vital signs, which is the measure —
MS. GREENBERG: By data source you’re not talking about where do you get it from.
MR. DIAMOND: We are talking, blood pressure actually backs into the medical record, cholesterol levels, hemoglobin monitoring, sorry, the hemoglobin monitoring or glucose monitoring, management monitoring, backs into chemistry so you end up with a chemistry docket if you will. And the measures are specifically defined, they can establish some criteria for accessing that data, asking yourself the question do you know where the data actually is.
Let’s assume that as an argument you’re looking at chemistry, in a hospital setting we know where the data is, it’s sitting in the laboratory in building ten. In an outpatient setting we actually don’t know where the data is and it may be in multiple locations so where is the data is a question. Second question is is the data coded, and can it be coded. Thirdly, can the data be accessed, can you actually access the data. You can then ask yourself the question of what is the cost of collection and attach some value equation to that data based on this matrix in terms of how many cells, how many measurement systems does it in fact support.
As an example when you actually construct the matrix you end up with for instance vital sign signs, blood pressure and other vital signs extracted out of the medical record, being of very high value. But you then look at some of the other criteria that we earlier discussed, it’s not easy to access the data, you look at chemistry and you see a large number of the cells are in fact populated and it’s relatively easy to collect the data because we often know where the data is, it can be coded, and that kind of stuff. So I mean that’s an approach that we would take.
I just want to make a quick comment about this UB92 connection, it seems to me there are two approaches to, maybe multiple, but at least two approaches to the question of attaching to the UB92 or the UB40 —
MS. GREENBERG: 04, the next version —
MR. DIAMOND: Whatever the next version is, one is to actually add the data to the form, which I think is what may or may not have been discussed at the last meeting. The other is to have it merely linked to the form, so you could have the UB92 or whatever the new version is and identify all series of data sources within the system, in this case within the hospital, and all you’re asking for is systems to be put in place to link the data, which gets away from some of the question of adding the data to the form, which as I understand it a lot of people objected to and probably appropriately so at the last meeting. Because linking gets you to the same answer, right?
MS. GREENBERG: Can I react kind of? I think this matrix you’re describing is absolutely, yeah, important and it would be great for I think the workgroup to partner with this work that you’re doing at MEDSTAT I gather. And I think linking in a sense was one of the options that was on the table, even with these recommendations, because you could use the attachment standard, which I’m told is going to be, there’s going to be a Notice of Proposed Rulemaking in my lifetime but I don’t know, actually it’s supposed to be in this year, you don’t even want it out there as an NPRM? Well, I mean there are —
MR. RODY: Not until they link the data to he electronic health record to build, the problem with UB, we used UB in Minnesota for years, we did not use it as part of the claims process because the insurers did not want it and we didn’t want to give them more data then they needed to process the claims. But we did run a second UB at the end of each quarter to report the last quarter decisions and that went to a state information bureau and they disseminated it to four citizens groups and the worker’s group and this process. So the form served because we had no other form. We’re not at a point where it would probably be almost as easy to develop a quality reporting form which if you’re doing electronic processing can be stapled, either stapled to the UB or sent separately, then to put it on the claims form, having sat on the UB committee for five years, and argue how you’re going to fit it and all that stuff and besides to look at development of electronic form then gets to Justine’s comment, if I have to collect blood pressure through the whole episode all I need is an identifier to tell me what the timing is on the blood pressure, I can give you all the blood pressures if you want it.
Then I take the example of making sure that when I build my electronic record system I’ve got the ability to not only collect blood pressure but I’m going to collect the time on the blood pressure because when I was reading your testimony some folks wanted time, some wanted dates, some wanted the beginning and ending dates of a procedure or times of a procedure, there’s a lot of data there, you’re not going to accomplish it on a claims form and that’s what you’re going to hear from these groups, you’re going to hear from us a concern about confidentiality and privacy on submitting it with the claim, and I think in the long run, again, if we’re trying to build a process and a system and indicators to improve quality or population health because I think it’s more then just quality, then let’s look at that. If we’re trying to get data in the meantime then let’s look at what would be the easier processes that we could agree to that aren’t going to just add to the work that we’ve already got.
MS. GREENBERG: Well, let me say the options that the workgroup discussed, and I think are discussed in the report, were one, are there any of these things that should actually be added to the basic UB04, UB92 04 837. And I think the one pretty much no brainer which I’m on the committee, the billing committee, is going to be added, is this one about the indicator for secondary diagnoses, whether or not they were present on admission.
A second option was some kind of an attachment, which didn’t have to go to the payer, you can use that attachment standard for other purposes, I mean once it’s agreed to as a standard it has the potential of being used for other purposes, there’s one on findings, or there’s one on laboratory findings and all of that so some of this stuff has already been codified in the claims attachment and it wouldn’t have to be sent to the payer.
A third one that John had put on the table was a separate transaction, a quality kind of transaction, that again does not have to go to the payer, but this could be for, this is the type of thing the employers are asking for these data, other oversight groups would be asking for it, so there are several different options. And I think what we agreed was even if you have the electronic health record nobody’s vision is, I hope, and certainly talking about privacy implications, just send the whole electronic record to these groups. So you’re going to have to have some kind of a transaction anyway, or some kind of a subset, so the real question is how do you populate it.
And I think that many of these things, to try to populate it from anything other then electronic health record could be so burdensome and could be at such high cost and so maybe questionable quality that you really can’t do it. There may be others that you could do it with, very selective, because of linking in to an electronic laboratory system or something else like that.
So it seems to me that all of that is on the table and so that probably only a small part of it might be on the table for actually adding to the claim. And I think that’s the kind of discussion we’re going to have with these folks on the 14th.
So if all of this recommendations one through eight being added to the claim, I think that was on the table as one option but I think it’s probably unlikely, but I think the question that we have been asking for years now, the committee has been asking for years, at least since it’s first review of the uniform hospital discharge dataset back in the late 80s and then into the early 90s is are there any quality items that could be added to the claim that would have high quality, I mean have good quality, high value, and be reasonable, relatively low burden.
And as I said I think this one about the indicator for secondary diagnoses has certainly come up, we brought up external cause of injury codes back in the early 90s, that got added too, so I think that’s been the question, is are there one or two things. And these are all candidates so I want to put this in perspective, I don’t think anyone, well, yes, some people were maybe, some of the consumer disclosure project people, they may all of this on the claim but frankly I don’t think they’ve figured out what they would do with it, but they are frustrated that they aren’t getting anything and I think that it behooves the committee, MEDSTAT, AHIMA, others, to try to respond to this need with this whole kind of possible cafeteria of approaches and to explore which ones might get us forward.
MR. HUNGATE: But I think the contribution could be to have a good model of approach which articulates these various pieces so part of what we’re contributing would be the decision model.
MS. GREENBERG: I don’t disagree.
DR. CARR: I think that something that has been very catalyzing is that any of these data elements in and of themselves are worthless, they’re only useful to the extent that they’re linked to an ICD-9 condition code, and I think —
MS. GREENBERG: Well, hopefully ICD-10-CM —
DR. CARR: Or 10, sorry, sorry, whatever, a condition, well you said you wanted it to today, so today we have ICD-9. But that then becomes the complexity because where is, actually to Anna’s point that someone comes in and has high blood pressure but is here for cellulitis and happens to have diabetes, so then how does this all play out, if there are four diagnoses for this episode of care then which data elements do you grab for today’s visit —
MS. POKER: I was just going to say also that ICD-9 codes if we’re going to use that, I don’t know yet the ICD-10, but for example they could have one thing for prostate cancer or prostate whatever, but it could be the acuity level of it is not captured. So in other words if we’re not creating potentially more problems. What I wanted to ask you Marjorie is you were talking about that providers don’t need some of the data —
MS. GREENBERG: They don’t want?
MS. POKER: That we may not have to transmit the payers some of the data that’s been collected, does that mean that you think, I mean are you envisioning something like a data repository for like quality data? And this being captured and like the hospital itself could collect data for internal review of what’s going on, is that what you were thinking of it, or not even thinking —
MS. GREENBERG: It’s sort of like what Dan described, almost all the states have these hospital discharge data systems, they mostly key off of the UB because that reduces provider burden and most of the items that they have wanted and that like were in the uniform hospital discharge dataset which is kind of what defines the content here were on the UB, or we got them onto the UB over the years, so that’s already a model that’s out there, and there’s some elements like race ethnicity that has always gone into that hospital discharge data system but doesn’t go the payer because they do one version for the payer and then they do this other version, and they had state assigned fields, so that’s already a model. I don’t think, when you start talking about repository I start thinking about that big database in the sky that nobody wants, or that some people want but that isn’t going to happen —
DR. CARR: But as you say it does exist, the hospital has their own dataset, in Massachusetts we send it to rate setting, and then there are people that can query it or there’s Medicare, you can query that and you can look for these data elements. So I understand more of what you’re saying so in terms, but that’s inpatient that dataset —
MS. GREENBERG: Yeah, at this point, I mean that’s inpatient. There are some states that are developing ambulatory —
MR. RODY: It’s outpatient for hospitals because hospitals report outpatient —
DR. CARR: Right, so you have the ICD-9 —
MS. GREENBERG: The real penetration is in the inpatient —
MR. RODY: — physician offices that are strictly 1500 based —
DR. CARR: I mean in our institution we have —
MS. POKER: [Off microphone.]
DR. CARR: So I guess I’m just sort of struggling with where are we going and then work back, so on an inpatient side to add to the administrative dataset a couple of data elements, then it’s all there and we use it the same way we use it now, show me everybody with heart failure and show me whether they had an ace inhibitor or beta block or something like that. And then on a different day you might say show me everybody with diabetes and admitted or something. But I think we need to be clear also in terms of are we, what is going to get better, this is inpatient care, if this is the dataset, and I had just a very kind of unclear sense on the last one both in terms of what we were thinking of and also what the people testifying at the hearing were saying.
MR. ROBY: When you collect data at your hospital having an electronic record you’re collecting more then just inpatient care.
DR. CARR: Yeah, we have multiple datasets and the key is to link them on an account number, so more and more we can do that but we’re still very not facile on outpatient, we’re just bringing up our outpatient order entry system —
MR. ROBY: And again, I think you have to look at this two pronged, if you get data today you may have to settle for an inpatient, because institutions can’t do that. But long term we have to face inevitably the fact how do we collect data across all sites of service because the bulk of patients are being seen in multiple units, and if any organization is going to build their little repository they’re going to need the data from all those places. I don’t think one large repository, no, but there are repositories, but if they’re going to do this right then they’ve got to deal with the outpatient field of service, they’ve got to deal with where else —
DR. CARR: That’s number six here, the episode of care and presumably that would be the, I’m assuming that’s what we mean by that, that someone had a breast biopsy —
MS. GREENBERG: This is actually not an inpatient exclusive, actually I mean we’re having Gene Narcissi is coming also from the National Uniform Billing Committee, I think, I mean some of these are more specific to like obviously the indicator is strictly for inpatient —
DR. CARR: Could be for outpatient too.
MS. GREENBERG: But like functional status, I think that certainly in some ways might make more sense in outpatient then inpatient to track, I mean in the sense that often people are released from the hospital so quickly that usually their functional status is worse then when they came in.
MR. RODY: You’re already selling me on eliminating a couple of your alternatives only because for example when I had cancer I could have been treated in a hospital based clinic, I ended up in a physician based clinic. I know for my own personal reasons that everything that was done in that doctor’s office was reported on the 1500. If I would have been in the Inova system everything would have been reported on the UB. If we’re going to collect data in a geographical area therefore I either have to have that element on both of those forms or I need to look at your alternative of some other method, whether it’s claims attachment that could fit on either, you can do anything you want with the 837 quite frankly, I can show you ways to do it, but some method that works for both groups because it shouldn’t be my choice of physician that determines whether you get the data or don’t get the data.
MS. GREENBERG: And frankly I don’t think we should spend a lot of time worrying about the paper form, I mean we told the consumer disclosure people that, again, I think you could handle this indicator on the paper form, there are a few things that you could handle on a paper form but I mean that’s not where the world is.
MR. HUNGATE: Well let’s decide overtly that we’re not going to worry about the paper form —
MS. GREENBERG: I think we should get testimony about the paper form —
MR. HUNGATE: Should we?
MS. GREENBERG: Well, they’re planning to, we’ve asked to even bring their forms. As I said —
MR. HUNGATE: They will each have their forms as part of their —
MS. GREENBERG: Yeah.
MR. HUNGATE: So will this cover the request that Justine had made to have that information presented?
MS. GREENBERG: Yeah, definitely ask them —
MS. POKER: I don’t know about them bringing the forms, we asked them to explain to us the process that they work and to bring if they have forms, like one of the things that Alexandra —
MS. GREENBERG: But she doesn’t have forms.
MS. POKER: Right, she explained she doesn’t have forms and she can’t possibly bring any —
MS. GREENBERG: I think the NUBC and the NUCC will, they should bring their forms.
MR. RODY: It seems to me, I almost want to say I don’t want to testify at that hearing to be honest with you, I almost think you need a hearing to talk to providers and maybe the insurance/payer community about whether things should be floating through their system a la the new Minnesota requirement or in other ways. But it would seem to me that X12 and HL7 both could also describe what other transactions or datasets could accommodate what you’re looking for as opposed to NUBC and NUCC which can only address their datasets —
MS. GREENBERG: That’s why they’re all on —
MR. RODY: Right, but I’m saying I think one discussion that’s not in the questions is what are the other alternatives that could be there to accommodate whatever way you see the data flowing from providers to payer —
MS. GREENBERG: You have the questions obviously? I apologize, these folders should have had the agenda in them, it should have had the questions —
MS. POKER: And this is the wrong agenda that you’ve been handed out, I’m sorry too, they photocopied it and this is not the one I gave them, it doesn’t have the list of the panelists that we’re having.
DR. CARR: I think it would be helpful to me is sort of to work as you say backward, that at the end of this exercise what will we be able to do, and then work back to sort of say we want these, let’s just take four major disease conditions and at the end of it these will be better. Now how will they get better? Well, we will know these vital signs or these lab tests or these condition codes and to know that we will need this dataset, the UB here or the 1500 here, and then to connect and see what are the data elements. And then if we see that there is no way to connect the 1500s and the UBs then that segment we’re going to put aside and say well is there something else we can get. Well, if we had the A1Cs we could go from here to here to here so we’ll put that leg of the table in place and it will stand on its own and we’ll be able to be better off.
What I’m worried about is that this talks about inpatient and outpatient electronic data elements, measured data elements, and perceived data elements, and we might get a little bit of everything and end up with all of nothing because we never got from the starting point to the finish point. So I would like to see a construct of to get here these are the building blocks that we need and there needs to be building blocks that connect to get from here to there because if you have the first three, the middle two, and the last one, but the missing elements for transmission and we’ll never get there.
MR. HUNGATE: Does the matrix that Lou presented seem to fit that same model?
DR. CARR: Well the concept does but not the, I mean what we’re saying here, just as we were saying, if we want the episode of care, to get that goal, an episode of care say for a patient with an oncologic issue that will have some surgery, some chemotherapy, some home therapy, and some radiation therapy, where will that information come from and have we, is there a way for that to go because otherwise we won’t get the episode of care.
MS. GREENBERG: I think that something much more modest was meant there and that was, it was particularly, I mean we’re really talking about episode I agree but it’s a much bigger issue, but what was, I think the main thing that drove this one, encourage payers to modify billing instructions to providers to align procedure start and end dates with services includes and selected global procedure codes, and you see we’re talking about selected global procedure codes. This was really driven by the delivery issue, that if you want to know when a person, what you really want to know with a delivery is, and if there’s a global code, when did the person’s prenatal visits start. You don’t want to know when did they come in and have the delivery, but right now, this was one that Kathy Coltin brought up and it’s described in the report.
So it’s really just for selected global procedure codes that by only having the date of the sort of final episode you are missing the information and payers have to respond as to when the first visit was, so when the prenatal visits started, they can’t get it if what’s on there is when they went in to deliver. So it’s not intended to —
DR. CARR: Well even so though, I mean then that would be a perfectly fine one to do, to say if a person gets their prenatal care in some local community and goes to the tertiary hospital for a high risk delivery, is there a way to link the private obstetrician office visit with the episode of care, how would that happen? I’m not sure we know how that could happen now because maybe the patient, maybe they’re expecting the patient at tertiary or maybe they’re not, they were transferred —
MS. GREENBERG: But I just wanted to make clear that I think this was intended for some specific areas that are problematic because there’s a global procedure —
MR. HUNGATE: I’m going to try to finish up our agenda for the next hearing.
MS. POKER: [Off microphone.]
MR. HUNGATE: Let’s finish up the agenda for the next meeting since it seems to me that what you want NUBC and NUCC should be bringing means that we don’t need a separate forms discussion on the agenda
MS. GREENBERG: I thought we took that off, we had an email discussion about that.
MR. HUNGATE: Could you have somebody be there on the 14th?
MS. GREENBERG: That would be great.
MR. HUNGATE: Okay, so I think we should add that in an appropriate way on the agenda.
MS. POKER: Well, since we have Wes Reschel(?) who may not show up, I just got an email last night —
MS. GREENBERG: Who?
MS. POKER: Wes Reschel —
MS. GREENBERG: He’s had some surgery, post surgery issues, and he doesn’t know if we can find a replacement at this late date, he’s going to try but he doesn’t know, I just read the email last night.
MR. HUNGATE: That’s a pretty big hole wouldn’t you say?
MS. GREENBERG: HL7 doesn’t come that’s a big hole.
MR. HUNGATE: So I think we have to fix that hole somehow.
MS. POKER: Well, I’m going to send him back an email.
MS. GREENBERG: He is actually somewhat disabled from the get go, although he’s very high functioning, but if he’s had surgery I don’t know, he may not —
MS. POKER: I don’t think he’s had recent surgery but as of the last surgery he had some complications he said —
MR. RODY: Somebody, Ed Hammond(?) certainly has history with that organization, they should be able to come up with someone.
MS. POKER: It’s kind of short notice, I mean I could try getting in touch with Ed Hammond —
MS. GREENBERG: There are so many people involved with HL7 —
MR. HUNGATE: Who’s our man from Utah on our committee who’s HL7? Stan Huff, I couldn’t come up with a name right now.
MS. POKER: Also Judith Warren is also doing HL7.
MS. GREENBERG: There are other people who are very active in HL7, somebody’s whose maybe been involved with the claims attachment standard, what about Maria —
MR. RODY: Gary Dickenson, that headed up one of the subcommittees for the electronic health record has been working with them for years and he’s was involved with the attachment —
MS. POKER: What’s the name again?
MR. RODY: Gary Dickenson, he used to be with Persey(?) and I think he’s on his own now. I could probably get you —
MS. GREENBERG: We can cover the persons travel —
MR. HUNGATE: Let’s fill that hole —
MS. POKER: Should I ask Stan then first because —
MS. GREENBERG: Sure, tell him that we were planning to have Wes and now we just heard from him and does he have any suggestions.
MR. RODY: We just need somebody who knows the gambit of what they’ve got to offer I think and not just —
MS. GREENBERG: Someone who’s probably been involved with the claims attachment and also maybe the EHR.
DR. MAYS: I mean if you get really stuck can’t he call in?
MS. GREENBERG: Well, that would be another possibility, maybe he could call in.
DR. MAYS: — have several other people and you only got to use him for a segment of time.
MR. RODY: [Inaudible.]
MS. GREENBERG: That’s true, he could call in and particularly if he had slides he could send them —
MR. HUNGATE: Okay, well let’s make sure the content is covered.
MR. HUNGATE: Now the other thing I want to talk for a few minutes about is Anna and I went to Vickie’s Population Committee meeting yesterday —
MS. GREENBERG: I was there, too.
MR. HUNGATE: And you were there, sorry. And would like to report on that, Vickie, where that ended up?
DR. MAYS: Well, I was going to say you can because I don’t know what part you want just for this group, but the discussion that we had is that you have I think five candidate recommendations, I don’t know what their numbers are —
MS. POKER: 11 through 13 were the ones you wanted to address.
DR. MAYS: I think nine and ten, and then hearing wise it might be 11 through 13 —
MR. HUNGATE: And 11 through 13 are germane to what we’re talking about.
DR. MAYS: Yeah, nine, ten, you all will have to read them —
MS. GREENBERG: There’s something with the enrollment form, there’s something —
MR. HUNGATE: Nine and ten were the survey kind of related —
DR. MAYS: Right, and we have enough, those are the whole issue of the inclusion of race and ethnicity I think within the survey concept and making sure that you do adequate sampling. We can discuss that because we’ve had umpteen hearings and we have the expertise on the committee to I think consult with you on that. 11 through 13 I think are a little bit —
MR. HUNGATE: Are race ethnicity, language —
DR. MAYS: Right, and those would be the question of in capturing race and ethnicity for example where do you want to capture it, it raises questions about are we talking about capturing it once somewhere, when do we capture it, do you deal with this every time the person comes in. The complexity that the subcommittee began to talk about is that, because I raised the question, and it’s what Justine is saying to some extent about your other things, where do we want to do, what is it that we really want to accomplish. We could capture the issue of race and will that give you for example better quality data, and so it got into the discussion that race is not all that one should be looking at, race to some extent may be not the thing to focus on only because some data will tell you that, some research has indicated that when you begin to control for variables like socioeconomic status, socioeconomic position, race may look quite different.
So what was raised was that in a perfect world if we could collect what we wanted it would be collecting race, it would be also collecting some socioeconomic position but we know that asking people their income is going to end up in a lot of missing data, so that instead a proxy that one might think about would be education, or educational attainment I should say —
MS. GREENBERG: Which the committee recommended about ten years ago.
DR. MAYS: I know, that’s what I was saying, when we go back and look at what we’ve done we’re redoing it so I’m at a point now where what it is that we can really get done and spend our time on that, so that that is a possibility of thinking about educational attainment.
And the issue of language, again if quality of care, and I think what we have to do is to make sure that we’re clear and this is what we began to discuss, are you interested in health care outcomes, actual treatment outcomes, are we talking about capturing information that has to do with health status, there are actually some differences in those perspective, what happens in the actual health care encounter, you may want something a little bit different then you would want if what your focus on is looking at health status because you’re interested in health disparities.
So Populations Committee thought that before it would be ready for a hearing on this matter that we really do need to have some time to discuss this, to say what is it that we hope to change, what is it we want to make better, what is the best way to do that, and then to be very clear about what the questions are that we would have in a hearing and who we would want to be asking those questions of. So that’s my sense. Marjorie, was there anything else, or Bob?
MS. GREENBERG: I think because the whole report is available some of the people are who are coming to the 14th may address race ethnicity, the feasibility, desirability, whatever, of getting that on the claim form. But it seemed to me that there needed to more, the bottom line from yesterday’s discussion was there needed to be more discussion before a hearing could be held.
MR. HUNGATE: But we have the people that would have been brought to a hearing if that were to be a hearing there, right?
MS. GREENBERG: They’re the same people —
MR. HUNGATE: They’re the same people —
MS. GREENBERG: That these particular recommendations are addressed to.
DR. MAYS: I was going to say if it is not taking it to far a field, if you’re sending questions before hand, if you just ask the question what is the feasibility of collecting data on race and ethnicity and where in the encounter do you see it as being easy or more difficult —
MS. GREENBERG: Or in the standard really.
DR. MAYS: Oh, I’m sorry, I’m saying it, so if you just ask that question it would get us started I think.
MR. HUNGATE: Well, that’s kind of my feeling, there might be some background that we could get from this process that would be illuminative —
MS. POKER: What is the second question?
DR. MAYS: I think Marjorie is probably better at that because I said feasibility and she said standards.
MS. GREENBERG: What’s the feasibility of collecting race ethnicity data in the standard that you maintain? Some of these, I mean they already have it, so it’s just kind of where. And one of them has to do with —
DR. MAYS: But let’s ask Justine’s question also and what is it linked to, it’s maybe somewhere but it’s like if it doesn’t follow the kind of data that’s going to be important to make a decision about quality it’s like okay, so what, because people collect a lot of things and then what you can do with it is so minimal because it isn’t really connected to the heart of what you would use to actually make a decision.
MS. POKER: But that would answer your question which was so important about where is the best place, where is it optimally important to collect the data, just because they collect it in that place doesn’t mean that that’s the optimal setting for it. That would be a great research question, one John wants to do, it’s like to do research on that, where in the continuum of care do you collect that kind of data, I think. And I think it would be a good one to do —
DR. MAYS: That’s not an AHRQ question —
MS. POKER: And I think that would be a great one to link together with the Quality Workgroup because that’s why, I mean that would be the research question —
DR. MAYS: I think what you want to do is capture that in your minutes so that we know to hold it over and to consider that that could turn out to be a recommendation to AHRQ, I mean for them to actually proceed with, or they may even have people already who are, RAND or somebody who’s doing work that they could ask them to look at this. So I would capture that because that may not necessarily be a hearing thing, that may be something that’s a discussion within AHRQ —
MR. RODY: There’s two questions there though, one is just collecting the data, one is collecting the data with the understanding of what the data could be used for.
DR. MAYS: Exactly, and see that last one is the one where I said Populations felt it needs to discuss that because the ultimate question is what is it we want to do, do better, I mean we started getting into discussions about health care treatment versus health disparities —
MS. GREENBERG: Obviously these forms are used for health care so that’s been —
DR. MAYS: But there’s also a way in which that data could be used in terms of illustrating disparities —
MR. RODY: But even there if I collect the information in an admissions department you will get different information then if I collect it as part of the history and physical.
DR. CARR: Absolutely, we see that everyday.
DR. MAYS: Wait, I know that but I want to hear it from you, why is it different?
MR. RODY: Because in an admissions office you are generally dealing with an entry level employee who is collecting data to fill in the blanks versus —
MS. GREENBERG: And may just eyeball the person.
DR. MAYS: Right, that’s what I wanted to make sure —
MR. RODY: Versus the history and physical where if you got a rationale why it helps health care you’ve got a health care professional then that will approach the question and make sure the answer is there.
DR. MAYS: Right, exactly, see it’s interesting because the presentation that we’re going to have today is only going to make this worse for you in the sense of another issue that has to be dealt with is the complexity of what you really want to know. I mean do you want me to just say to you well I’m black, and you really don’t understand that I could be Caribbean black, I could be whatever, so it’s kind of like, and if we only do the minimum OMB standard, which some people will say well let’s just do what OMB does in terms of a minimum recommendation, I don’t know that that’s going to get you what you want, I don’t know what you’re going to do when you have mixed race individuals, like okay now this person in the history has told me they’re part Navajo and they’re part something else, what do I do, and then the person has an expectation I’ve now shared this with you, what are you going to do differently in my treatment. And there’s no data to tell them what to do differently in the treatment. I mean there’s some issues here that Pop felt we should have some discussions about.
DR. EDINGER: I guess the same thing probably with language, if you don’t understand the language you might not be able to identify what it is, if I was coming into the admission room and I started talking to you and you had no idea what language, it might sound Germanic, or Slavic, or Spanish or something, Latin, but you had no idea which one it was you might capture it wrong.
MS. GREENBERG: I’m still kind of of two minds here because I don’t want to collect data, I mean we’re already collecting data in, almost all of the hospital discharge data systems do collect race ethnicity —
DR. MAYS: Not well, just understand that.
MS. GREENBERG: Well, I think they probably don’t do a bad job collecting some, they certainly under count others, but you’ve got these recommendations coming out from IOM, you’ve got Aetna, you’ve got other people now starting to recognize and reinforce the need to collect this because of health disparities. So it’s a question of are we going to let the perfect become the enemy of the good, are we going to say let’s stop collecting race ethnicity, ignorance is bliss, I don’t know. I mean of course in a survey and in a research project you’re better off getting socioeconomic position and getting various other things and finding out a lot of different types of detail, but we’re not talking about that environment. So they’re interesting questions.
DR. MAYS: But I think that one of the goals might also be, Marjorie, is to understand what goes on in the process and in those instances where data is already being collected maybe it’s an issue of training or guidance also, so I mean one doesn’t have to come in to upset the whole apple cart but it’s to know that in admissions there’s guidance given to the person who collects the data —
MS. GREENBERG: That we recommend best practices, that type of thing.
DR. MAYS: Yes, so it’s to some extent I don’t know that we’re like going to reengineer everybody’s system but we may say that if you’re trying to reach a best practice here’s some places that we can identify that having a set of, a chart or something to tell the person what to do, or to train them, might be helpful. So I think we also have to look at progressively what can we do while there’s a big fight about what is the place to collect it. We could in the interim say if you’re collecting it you can also do better.
MR. HUNGATE: There’s a political battle and there’s information reality battle, sometimes you do it politics ways and perfect to get you in the right direction so I think we have to be cognizant of that as we deal with this one.
MS. GREENBERG: I mean my sense is, actually once we have those people there it doesn’t hurt to ask them the question.
DR. MAYS: So I think if you could have that specific question, prepare them to answer that —
MR. HUNGATE: As much focused on race probably as on language, or is there a difference?
MS. GREENBERG: We could ask them about language too.
MS. POKER: The questions I wrote, do you want me to email it to you before I send it out?
DR. MAYS: Sure.
MS. POKER: I’ll email it to you, you and Marjorie.
MS. GREENBERG: I think on language was recommended that this be part of the enrollment transaction, and the NUBC and NUCC people don’t have anything to do with the enrollment transaction, that’s an X12 question.
MR. RODY: Again I think at some point then you need to ask these same questions and maybe given what you have here you’d be better off with a payer/provider panel at some other point because just taking that as an example, if I could it on an 834 that means I’m providing the data either via the employer or directly to the insurance company, I think you all have to ask the question if that question would be answered correctly on an enrollment form, would someone be willing to put their race and ethnicity data on a form that’s going to their insurance company or to their employer.
MS. GREENBERG: Right, well that’s why there was this other recommendation about it should just be for new patients. We’d actually talked about a separate transaction kind of for, but then you only get it for the people who actually come in, and it’s sort of then part more of a history, physical background, you’ve got to find out their language, etc., but the insurers, the plans, want to know the language of people who are enrolled but don’t come in because how can they do outreach to them if they’re not doing it in their language —
MR. HUNGATE: If you can’t talk to them how do you fill out the form.
DR. EDINGER: Marjorie, would another quality issue be the business case, if you have people available to speak the language — [inaudible] —
MS. GREENBERG: That’s also true, because that’s why you need, but maybe it shouldn’t be when they actually apply for insurance because that could be discriminatory, so it’s complicated.
DR. MAYS: In this instance there are I think people we can talk to, like we can see what Aetna is doing, we can see what LA Health Care has a whole system its put into place —
MS. GREENBERG: I would actually not deal with language at this hearing on the 14th —
DR. MAYS: I was just going to say let’s just do race because I don’t want them to get defensive, I don’t want, it’s like let’s just take, the business case has been made really clear for race, I think there is —
MS. GREENBERG: Would you say race and ethnicity?
DR. MAYS: I’m sorry, race and ethnicity, and I think it has been made somewhat for language, but language is, because there are regulations around language it makes people more nervous around whether they’re going to get sued —
MR. HUNGATE: That was my reading.
DR. MAYS: So I think we should wait on that one until we —
MR. RODY: We were going to testify at this hearing but I’m not sure that’s appropriate given what you’re trying to do.
MS. GREENBERG: Why?
MR. RODY: Because our approach is going to be what does it take to fill out this information, collect this information, to put it on one of these vehicles as opposed to will or could this be the vehicle, and it seems to me that if you take Justine’s comment about what is it we want to collect, you add to that the comment of what the potential vehicles could be, then the third group you might want to go to after that point would be the providers of the information and maybe also the insurers of the information, that they may be the vehicle for the collection. I’m only saying that because you only have one day to do that so you need to look at the panel maybe a little differently, I mean we can, no problem on getting you the testimony, I just don’t know if that fits with what I hear you tracking here.
MR. HUNGATE: I think it still fits.
MS. GREENBERG: We’ve got this report, we’ve got these recommendations, you know what it is —
MR. HUNGATE: Is it you or is it someone else?
MR. RODY: It’s myself and the director of medical records for Hackensack University Hospital.
DR. CARR: Just as kind of trying to synthesize what we’ve talked about, I went through the first eight recommendations and identified whether they affect inpatient or outpatient or both, and then I identified four obstacles, data collection, data linkage, data interpretation, or other, and kind of scored them. And then also put four or five major conditions that we would be trying to make better, and it’s just a start but I think the only thing that doesn’t have any obstacles is number three, the modifier as you say. These other ones, this may be a beginning of a template but to begin to say what are the obstacles and what’s the condition that’s going to be better at the end of it. I think it will help us focus to find one thing that will work and certainly the modifiers I think definitely will help. But these other things, I think we then have to pick our battles, are we going to try to figure out a data collection or data interpretation, I think data interpretation is very important because knowing that a blood pressure is 140 over 70 is only going to be important if the person has a history of hypertension and is now better, or used to be 90 over 60 —
MR. RODY: And is it at the beginning of the admission or the end of the admission.
DR. CARR: Exactly. On a medication off a medication, so I’ll put this on my computer and email it to you —
DR. MAYS: Can I just ask a question? Would you want at all to have anything about the quality of the data? Some things that we collected are, maybe it’s accurate, maybe it’s not, the whole issue about the quality of the data.
MS. POKER: Justine, is this something you want to send out to the panelists?
DR. CARR: No, we have to get our thinking together —
MR. HUNGATE: This goes to the workgroup, staff members —
DR. CARR: So for example a lab test it is what it is, a vital sign might be flawed.
MS. POKER: And Justine are you going to do this? Are you going to send it out to us?
DR. CARR: I have my computer, I’ll do it —
MR. HUNGATE: What is our best distribution method to get information reliably to all of our group?
MS. POKER: Email, except you —
MR. HUNGATE: I’m on email but I’m just not always there.
DR. MAYS: I’m off and on this mailing list, so whatever, I don’t, sometimes I get stuff sometimes I don’t.
DR. CARR: I’m going to have this done in about 20 minutes, I have my computer, and if you give me a desk or something —
MR. HUNGATE: But the methodology is —
MS. GREENBERG: Well, things should be probably —
MR. HUNGATE: Is it Debbie, is it Anna, who maintains the list of who the workgroup is?
MS. GREENBERG: Gracie is the one —
MR. HUNGATE: So Gracie is the distribution point —
MS. GREENBERG: If it doesn’t need review by someone else it can just be sent to Anna and Gracie, or Ed to Anna, then Anna can look at it, that’s what I would do, I’d email it to Anna, have her look at it, and then Anna sends it to Gracie and say please send it —
So send this to Gracie and say please send it out to the workgroup members and staff, tell them that Dr. Carr developed this —
DR. CARR: We ought to probably have those at every workgroup meeting even though I’ve committed it to memory, every time you look at them there’s something new that I didn’t see before.
MS. GREENBERG: And I guess in the future too, and we failed to probably communicate properly with you but you should be letting Gracie know what you want in the folder for these meetings, whether it be the breakout sessions or the hearings or whatever.
MR. HUNGATE: Now we have not gotten to the risk adjustment agenda, and we have to —
MS. POKER: But have we finalized this? Because I need to, is this a final, were going to have two panels, and we’re going to have discussion, what is the discussion —
MR. HUNGATE: Have you added Lou Diamond to that?
MS. POKER: I don’t know which one to add him, where would you put —
MS. GREENBERG: Well, I would think he’d be separate.
MR. HUNGATE: I think separately after lunch.
MS. POKER: So right after lunch. What time would you like —
DR. CARR: This sounds like an administrative discussion, could we just get to the risk adjustment because we have eight minutes.
MR. HUNGATE: We don’t have time to deal with risk adjustment now, we’ll have to do that —
MS. GREENBERG: Do we want to have a conference call?
MR. HUNGATE: I think we probably better have a conference call.
DR. CARR: And the question that we’re going to address is?
MS. GREENBERG: Well, I guess whether we want to have hearing, I don’t know, go back to, several meetings we talked about going up to Boston and meeting with some of the experts and talking with them —
DR. CARR: I think we should stat with the goal that we’re trying to accomplish is what, and then work back from that, in order to get to that what are the building blocks that need to be understood and what are the obstacles, same sort of format as this. Because I’m not sure the goal we want to accomplish is, maybe Bob can articulate —
MR. HUNGATE: I will articulate as well as I can where I think risk adjustment fits in the quality agenda and how it can make the system work better in a quality improvement sense because I think that’s the case that needs to be clear and it’s part of what is in this but it’s not visible.
DR. MAYS: Can I make a suggestion? And I may be totally off base here but I just don’t know, is that Quality was earlier doing a lot of its work in the full group so would this risk adjustment benefit from having a full committee discussion?
MR. HUNGATE: Yes.
DR. MAYS: So maybe if you could like bring it up maybe when we each do our reports that that’s something, and then maybe a preparation to discuss it in November, at the November meeting?
MS. GREENBERG: Well, but I think it should be a structured discussion with maybe a few experts —
DR. CARR: I like your idea to sort of say, because even Bob and I have very different ideas about it and before we get a workgroup to work on it I think we should have a clearer top down sort of where this fits in the big picture and what aspects of it because —
DR. MAYS: I think even before you bring in a panel is that I think what’s going to happen in the full committee —
MS. GREENBERG: We had quite a bit of discussion at the executive subcommittee retreat when you weren’t able to go to —
DR. MAYS: But it’s translated to what? That’s what I want to know, I wasn’t there but I did I get it? How do I benefit from it?
MS. GREENBERG: I think it was somewhat documented in the minutes.
MR. HUNGATE: I can find the minutes and transcript and send you that —
DR. CARR: But even so, just synthesizing the goal of risk adjustment and risk adjusting for what, toward what end.
DR. MAYS: That’s what I was just going to say, and I would like to see, I think the committee, honestly, I think the committee is going to have lots of different directions it wants to go in so I’m just saying if you’re responsible for reporting to the full, if that’s still it, then you need also direction from the full I would think.
MR. HUNGATE: Yes you do, I agree. That’s fine.
MS. GREENBERG: And if you want to have something, address this on the November agenda, let us know.
DR. CARR: So today at the big meeting the report out on Quality Workgroup was about what?
MR. HUNGATE: The hearings that we’ve held, the one that’s coming, and to raise the issue of risk adjustment as an agenda for the full committee and say how we need to deal with it.
DR. CARR: Okay, so we should probably have a statement for the meeting today to say risk adjustment of this for this purpose is something we want to address, does everybody agree, but I mean I think it has to be specific.
MR. HUNGATE: I don’t think we have that kind of specificity from this discussion.
DR. MAYS: What about from the executive committee? Did you all decide something at the executive committee about where you want to go on risk adjustment?
MR. HUNGATE: Risk adjusted outcomes as indicators of quality are argued by many to be the best way of guiding quality improvement, accountability for risk adjusted outcomes is felt by many to be a good driver of quality improvement. I think that’s what Simon Cohn and I at least agree, beyond that I’m not sure where the whole committee is, it hasn’t had a broad enough discussion —
DR. CARR: And I think that risk adjustment is alive and well in cardiovascular domain but beyond that I don’t know that there are accepted risk adjustment models for outcomes so I think that’s the issue and I think the STS and the American College of Cardiology have all of this well in hand and there are state reporting requirements now growing across the country for people to report their data and its risk adjusted so again —
MS. GREENBERG: For cardiology.
DR. CARR: For cardiology, so you have, for some outcomes, but because you need a certain volume threshold and so on, so if that’s in place and then there’s a huge chasm between either the evidence or the volume or the decided upon risk adjusters then I think we would be premature to be advocating that this be imposed when there isn’t the really evidence base —
MR. HUNGATE: I’m not proposing advocating imposition, I’m proposing that we understand the limitations, the benefits, the uses. American College of Surgeons collects a lot of outcomes data and how do they deal with, they have a quality improvement person, Scott Jones, who does that. I think the American College of Surgeons could talk about how they see risk adjustment in surgical outcomes as it sits in their profession, I think there’s a content there that is background for discussion.
MS. GREENBERG: Actually part of the discussion that we had at the November executive subcommittee retreat was I remember Ed Sondik said he really felt that this was a major research area, major area for additional research. What we’re talking, the subcommittees, workgroups, full committee, coming forward with identifying high priority research areas so I think it fits in very well with that. And as I understand what the Quality Workgroup was doing was exploratory discussions to try to even help frame what some of the research questions are.
DR. CARR: I mean I did follow Elise Siazoni(?) and we’ve had correspondence but the study that she did a couple of years ago showed the greatest predictor of inpatient mortality was the question on admission can you bathe yourself. They looked at labs —
MS. GREENBERG: See that’s what I said, we need functional status.
DR. CARR: I’m not saying it’s too related, I’m only saying that was the only thing that came up.
MR. HUNGATE: Okay, thank you all, the other meeting is beginning.
[Whereupon at 10:00 a.m. the meeting was adjourned.]