Hubert H. Humphrey Building
200 Independence Avenue, S.W., Room 325A
Washington, D. C.
CHAIRMAN IEZZONI: Why don't we get started so maybe we can get done early. We are a group who believes in wishful thinking. I think we are supposed to go around and introduce ourselves first because there are some new faces at the table. This is, again, the Subcommittee on Population for the National Committee on Vital and Health Statistics. I am Lisa I. Iezzoni from Harvard Medical School, and I will be chairing this session. Then I will be turning it over to Kathy who will be chairing the Quality Work Group Session. So, Kathy, do you want to introduce yourself?
KATHRYN COLTIN: I am Kathy Coltin. I am from Harvard Pilgrim Health Care.
ELIZABETH WARD: Elizabeth Ward from Health Information Institute.
VINCENT MOR: Vincent Mor, Brown University.
DEBORAH MAIESE: I'm Debby Maiese with the Office of Disease Prevention and Health Promotion staff subcommittee.
SUSAN QUEEN: Susan Queen from the Health Resources and Services Administration.
CHAIRMAN IEZZONI: And you're replacing Mike?
SUSAN QUEEN: Yes.
CHAIRMAN IEZZONI: Great.
PAUL NEWACHECK: Paul Newacheck from the University of California at San Francisco.
DALE HITCHCOCK: Dale Hitchcock from here in the Department ASPE, staff for the subcommittee.
CAROLYN RIMES: Carolyn Rimes, HCFA, staff to the subcommittee.
LEON GOE: Leon Goe, HCFA, staff.
KATHERINE JONES: Katherine Jones, staff.
CHAIRMAN IEZZONI: And Patrice Upchurch. And?
JANE BEACH: Jane Beach, Court Reporter.
CHAIRMAN IEZZONI: Okay, nice to see you, Jane. Okay, Clyde Tucker from the Bureau of Labor Statistics will be joining us at 2:30 to talk to us about an update on where the OMB Guidelines on Recording Race and Ethnicity are, or the Implications of Implementing The Guidelines, to be more precise. And the rest of the business that we have for the afternoon frankly is the reports. We have got to get them done, folks, so we can get them off our plate, and then Kathy will be taking over at 4:15 for the Work Group. So why don't we start with the Islands and Territories Reports that Dale told us this morning at the Full Committee meeting will need a new title, because "Insular Areas" is a pass term, and so we have to come up with a new name.
DALE HITCHCOCK: You know, I am not sure which is pass and which is politically correct anymore, but there is some disagreement over what we should call it.
PAUL NEWACHECK: Maybe we should use both.
DALE HITCHCOCK: Well, that's been done.
CAROLYN RIMES: There's also, what, the Cohn-Lynn approach?
CHAIRMAN IEZZONI: Which is?
CAROLYN RIMES: One title, colon.
PAUL NEWACHECK: Or, also known as.
DALE HITCHCOCK: Yes.
CHAIRMAN IEZZONI: Also known as.
DALE HITCHCOCK: I don't know how many of you got to see, how many of the comments everyone got to see, so I made copies. There aren't many. I'll hand them out and we can look at them. Depending on what you want to do, did you want to let people look at them for a few minutes? Or I could walk you through them.
CHAIRMAN IEZZONI: Well why don't we first see whether committee members have any comments, and while people are talking they can be glancing at the comments that we did get from the field. Do any committee members have any specific comments about the, I'm going to call it "The Pacific Insular Areas and the Puerto Rico and Virgin Islands Report"?
DALE HITCHCOCK: We pretty much covered it in the last meeting.
CHAIRMAN IEZZONI: Okay.
PAUL NEWACHECK: We're all very happy with that.
DALE HITCHCOCK: Or something of the sort.
CHAIRMAN IEZZONI: I was gone. Okay, can I just ask you a couple of questions, Dale?
DALE HITCHCOCK: Sure.
CHAIRMAN IEZZONI: This version didn't have any boxes in it with quotes. Did you decide to get rid of those?
DALE HITCHCOCK: It must have been a problem with your printer.
CHAIRMAN IEZZONI: Well it was mailed to me, and every apostrophe was replaced by an equal sign in my copy.
DALE HITCHCOCK: Did you download it first into Word Perfect?
CHAIRMAN IEZZONI: I didn't-- Look, I got this mailed to me. It was double-sided mailed to me. It was sent through the mail.
PAUL NEWACHECK: I think some of the copies were made that way. I suspect it's a word processing problem.
DALE HITCHCOCK: When I send them out, they always have boxes and maps.
ELIZABETH WARD: What was approved had boxes and maps.
CHAIRMAN IEZZONI: The one that was approved had boxes and maps?
DALE HITCHCOCK: Yes.
CHAIRMAN IEZZONI: Okay. Good. Well so nobody has any extra comments on it other than, Dale, I have a few, you know, like little typo types of things.
DALE HITCHCOCK: We'll be glad to incorporate them.
CHAIRMAN IEZZONI: Well, you'll be glad to disincorporate the typos.
DALE HITCHCOCK: The comments we're going to incorporate, yes.
CHAIRMAN IEZZONI: Right.
DALE HITCHCOCK: The typos may be something that happened in the transmission, too.
(Laughter.)
CHAIRMAN IEZZONI: And I think that, other than that, I mean I thought that you've done a good job of talking about the diversity and bringing in the recommendations. So--
DALE HITCHCOCK: Do you have the most recent copy, Lisa, that has the--
CHAIRMAN IEZZONI: This was mailed to me within the last couple weeks.
DALE HITCHCOCK: It has 15 recommendations? Could you double-check and see? That's one way we can tell.
CHAIRMAN IEZZONI: Okay, it goes to beyond HCFA. This was what got mailed to me and I just stuck it in my folder for today's meeting. No, I have 11 recommendations.
PAUL NEWACHECK: So you have an earlier draft.
CHAIRMAN IEZZONI: Well, here, Dale. You've probably corrected all the typos by now.
DALE HITCHCOCK: Yes, this is a May 24th version, and the one you should have is a June 25th version.
CHAIRMAN IEZZONI: Okay, does anybody have any-- I have a new secretary, by the way, after the temp from hell-- sorry. That could be one of the explanations for part of this. So the comments?
DALE HITCHCOCK: Do you want me to go through them sort of briefly? There's a couple of major overriding sorts of concerns, and then there are just some specific things.
CHAIRMAN IEZZONI: Okay.
DALE HITCHCOCK: Maybe I would just mention what I see as the overriding concerns and then briefly go through them. One concern is this thing that Joe Eiser raised about the title, and I'll talk some more about that.
CHAIRMAN IEZZONI: Right.
DALE HITCHCOCK: And how to refer to the Islands. The second thing is sort of the issues of fairness and coverage. I sort of limited the report to what we learned at those meetings on the 14th and 15th, although I did put in some other background sorts of material.
CHAIRMAN IEZZONI: Yes, because HRSA sent a very long letter.
DALE HITCHCOCK: A very long letter. Lots of things have changed since we originally met. On the other hand, we had a senior person there from HRSA who has had ample opportunity to review this and then to present it to us on the 14th and 15th. I'm ambivalent about what to do about the HRSA comments. I can see if we allowed SAMSA and CDC to update us on what happened since July 14th and 15th, in other words what might have happened since we sort of put a spotlight on the whole set of activities, we could open this up again.
CHAIRMAN IEZZONI: Yes. My recommendation, Dale, and other committee members should chime in, is if there are factual errors in our report that were brought to light by the HRSA comments, that you should correct them. So any absolute factual errors should be corrected.
DALE HITCHCOCK: Sure.
CHAIRMAN IEZZONI: And there should be a blanket statement at the beginning that these are our findings as of--
DALE HITCHCOCK: Yes.
CHAIRMAN IEZZONI: --and state the date, and say that there have been--that we are aware that there have been numerous ongoing activities since that time; that we have not been monitoring carefully, and so therefore our recommendations should be viewed within the context of current, ongoing activities within the appropriate departments. Does that sound reasonable to committee members?
VINCENT MOR: Yes. I think so, otherwise you have a major problem of coverage, and time frame, and stuff.
CAROLYN RIMES: We would never end. We would just continue to update this forever.
CHAIRMAN IEZZONI: Yes. And so the date that you should specify should be the July 14th-15th, was it, 1998. We have to have the year in there. That was the date of the comments. Paul, did you want to say something?
PAUL NEWACHECK: I was just curious. What are the recommendations that they're making within this context?
CHAIRMAN IEZZONI: They just talk about all these new initiatives that they are undertaking.
DALE HITCHCOCK: Yes. It's in that letter there.
CHAIRMAN IEZZONI: Yes.
DALE HITCHCOCK: That we've got.
CHAIRMAN IEZZONI: And they're doing a lot. I mean Linette Arocki is now at HRSA, and so Linette is about to go to Hawaii and the Pacific Islands, if she hasn't already gone. So there's obviously things happening right now that we didn't know about in July of '98 and I think we have to--
ELIZABETH WARD: Yes, we could use words, "so we're pleased to understand that in fact since this report"--
CHAIRMAN IEZZONI: Right. Have been some activities.
PAUL NEWACHECK: Some of these pre-exist the report that still were not included in our report.
DALE HITCHCOCK: Yes, that's true.
PAUL NEWACHECK: Do we want to include those, or not?
DALE HITCHCOCK: That is sort of part of the same question, yes.
PAUL NEWACHECK: Like this State System Development Initiative has been around for a few years that's referred to here. So is that something that they-- they didn't bring it up before.
DALE HITCHCOCK: They didn't bring it up before.
CHAIRMAN IEZZONI: Well that is why I think we need to correct factual errors. So if there is a part of the report that touches on that, or would have direct bearing on that, you can mention it in a sentence or two.
DALE HITCHCOCK: That's what I was thinking.
CHAIRMAN IEZZONI: Because we don't want to give an appearance of erroneous information. But we also--it would look very disproportionate if we added all the HRSA material and didn't add all the material from the Department of the Interior, from the CDC, from the other places. So, Dale, is that enough for you to work with?
DALE HITCHCOCK: Sure. Sure.
CHAIRMAN IEZZONI: Okay. So the process will be, did you give the Islands and Territories' folks a date by which they had to have commented to you?
DALE HITCHCOCK: Yes, we did. I think it was the 13th of September. We only got one. We only got one from Palau.
CHAIRMAN IEZZONI: From Palau.
DALE HITCHCOCK: And Puerto Rico may give us a comment.
CHAIRMAN IEZZONI: Yes, she corrected her name. I think what you should do is, I don't think our committee can finalize what the title of this report should be because we just don't know all the issues.
CAROLYN RIMES: Did anybody give you any suggestions other than the one gentleman?
DALE HITCHCOCK: Joe Eiser is the only one who noted this.
CHAIRMAN IEZZONI: Why don't you pass it by Roy Lynn and Pass it by Linette, because if you can get Roy Lynn and Linette to sign off on the titles I think that you will be in okay shape.
DALE HITCHCOCK: I mean even Linette's letter, if you look at that, she referred to it in a couple of different ways. On the front page she's talking about the Pacific Insular Areas, and then later on it's the Pacific Basin Jurisdiction. So...
CHAIRMAN IEZZONI: Well just give her a call. When does she get back?
SUSAN QUEEN: I don't know. She's not back yet. Paul got back today.
CHAIRMAN IEZZONI: Okay, so she'll be back probably within a week or two.
DALE HITCHCOCK: So do you want me to sort of develop a consensus from our experts?
CHAIRMAN IEZZONI: We're not in a position to give you any input on this.
ELIZABETH WARD: If they have an official way of referring to these places--
DALE HITCHCOCK: There is no official way. I mean, I'm not interested in your expertise in the background; I'm interested in the process I need to go to to arrive at what we can call it.
CHAIRMAN IEZZONI: Why don't you ask each of--you know from the one person what his preferred title is. Ask Linette what her preferred title would be. And ask Roy Lynn what her preferred title would be. If those titles are hugely disparate, then you can float it by the committee on an e-mail and we can maybe help you. But if the titles have some concordance to them, I'm happy to change the title.
DALE HITCHCOCK: Sure. I am, too. Okay.
CHAIRMAN IEZZONI: Okay? So is there a process by which somebody will do just kind of a final proofreading of this, Dale? How does that work? Just to make sure it's all perfect?
DALE HITCHCOCK: Actually I'll have to rely on other subcommittee staff to help out.
CHAIRMAN IEZZONI: Could somebody just read it over? Because it's always good for a fresh pair of eyes just to--
DALE HITCHCOCK: Oh, absolutely.
CHAIRMAN IEZZONI: Make sure that there aren't any blatant typos.
CAROLYN RIMES: We're getting better at that.
CHAIRMAN IEZZONI: Yes. You can look at the version which I have which had some things that have probably been corrected by now.
CAROLYN RIMES: You may need to give me one more copy.
DALE HITCHCOCK: Sure.
CAROLYN RIMES: Actually, HCFA's doing some stuff, too, which is good.
CHAIRMAN IEZZONI: Okay. So are we set on that?
DALE HITCHCOCK: I should probably just note also that I didn't get any comments back from Census, and I think they're a little displeased with the report, that we've apparently painted them in a sort of negative light, but I've gone back and reviewed the transcripts and nothing is in there that's been made up.
CHAIRMAN IEZZONI: There are problems.
DALE HITCHCOCK: Yes, there are problems.
CHAIRMAN IEZZONI: There are problems, and that is what we identified.
CAROLYN RIMES: As well as other agencies.
CHAIRMAN IEZZONI: Okay, now what is the process? Is this report to be sent to Shalala? Or should we send it to the Data Council?
DALE HITCHCOCK: It usually would go through the Data Council to Shalala.
CAROLYN RIMES: But isn't it from the full committee?
DALE HITCHCOCK: Yes.
CHAIRMAN IEZZONI: It has to be from the full committee. So it would have to have a cover letter--
VINCENT MOR: Was it not voted on last time?
DALE HITCHCOCK: Yes it was.
CHAIRMAN IEZZONI: But it would have to have a cover letter signed by John Lumpkin. And so the process for drafting that cover letter to be signed by John is?
CAROLYN RIMES: Well we need two things. We need an executive summary and a cover letter, which Dale and I can work on.
CHAIRMAN IEZZONI: So Dale and Carolyn will draft that. And so as soon as you've finalized the report by adding in any residual comments and making the few changes relating to the HRSA letter and the title, you will get a draft letter for John Lumpkin to review and sign, and then we can submit it.
CAROLYN RIMES: Through you.
CHAIRMAN IEZZONI: Through me? Okay. All right, great. So hopefully that will be done within the month?
DALE HITCHCOCK: Yes, I would think so.
CHAIRMAN IEZZONI: Okay. It would be nice to get this one done.
DALE HITCHCOCK: It would be very nice.
CAROLYN RIMES: At least what Palau wrote.
DALE HITCHCOCK: Yes.
CAROLYN RIMES: I was concerned that not any of those groups--
ELIZABETH WARD: I thought it was a nice--
CHAIRMAN IEZZONI: It was a very nice letter.
DALE HITCHCOCK: Joe Eiser, who knows those folks well, said that he would be surprised if any of them sent back comments. I don't really know what his reasoning was, but I was surprised and pleased that Palau sent back the note.
ELIZABETH WARD: Remember their descriptions of their work load, I can imagine that this would not be easy.
CHAIRMAN IEZZONI: Okay, great. Any more comments on that report?
(No response.)
CHAIRMAN IEZZONI: So we will not talk about it again.
CAROLYN RIMES: Other than when we get the letter to Shalala, you guys might want to know that.
CHAIRMAN IEZZONI: Please tell us when it's gone. That would be nice. Okay. Dale, thank you for all of your hard work. And Carolyn, thank you for working with Dale on that and getting this done. Thank you. Now to the slightly weightier issue of the Medicare Managed Care Report. Again, I wasn't here in June, and so I wasn't aware of the full nuances of everything that went on at the June meeting. So if I make statements that show a lack of appreciation for the discussion in June, it is because I don't appreciate the discussion in June.
VINCENT MOR: I hope you had a good time in, Italy?
CHAIRMAN IEZZONI: I went to Italy for a couple days, but we were in northern-- Switzerland, so it was very northern Italy. Yes, it was really nice. It was a long time ago, or it feels that way.
PAUL NEWACHECK: The afterglow is gone.
CHAIRMAN IEZZONI: Yes, it's gone. Everybody has had ample opportunity to comment on this report. Let me just make that statement right up front. This report has-- you know, this dog has trotted around the race course a number of times.
CAROLYN RIMES: And it still has legs.
CHAIRMAN IEZZONI: It still has legs.
(Laughter.)
CHAIRMAN IEZZONI: And so I think everybody needs to understand that you all have had an opportunity to comment on it. And so what we are doing now is commenting on the version that Carolyn as yet does not have an executive summary of, but why don't we agree that the executive summary will be basically plucking a section from the report and moving it en masse without changing any words into the executive summary. So what we approve today--
CAROLYN RIMES: And I also got lazy, because I wanted one final review on the recommendations before we did--
CHAIRMAN IEZZONI: Okay, well that's what we're going to look at right now.
CAROLYN RIMES: Anything.
CHAIRMAN IEZZONI: We're going to look at the recommendations which start on page 42.
PAUL NEWACHECK: Before you get to that?
CHAIRMAN IEZZONI: Yes.
PAUL NEWACHECK: The comments we got this morning in privacy and confidentiality--
CHAIRMAN IEZZONI: Yes.
PAUL NEWACHECK: --and also the comments that we haven't heard yet from Simon, do we know anything about what his concerns are?
CHAIRMAN IEZZONI: I don't. I mean, did Simon raise concerns at the last meeting?
PAUL NEWACHECK: I don't think so.
(Several simultaneous voices saying 'yes'.)
PAUL NEWACHECK: Did he?
ELIZABETH WARD: Oh, yes. Significant ones.
PAUL NEWACHECK: Which ones?
CHAIRMAN IEZZONI: I think Simon was concerned about health plans' responsibilities, at least that's what I remember hearing. Let me just state up front what I propose, and the subcommittee needs to help me decide whether this should be our position. I want to get this report out because I think that a lot of people have worked on it a long time, and it contains useful information. However, it is also important to give all points of view. And so if Bob Gelman, or Simon Cohen, or other people want to write dissenting opinions, or a statement that they would like appended to the report, that sounds fine to me. I would feel comfortable with that.
DALE HITCHCOCK: Since I wasn't here this morning--
CHAIRMAN IEZZONI: Shall I tell you what Bob did this morning?
VINCENT MOR: Yes. That would be helpful.
CHAIRMAN IEZZONI: Okay. After I introduced the report, in about three minutes Bob raised his hand and said he had a comment. His basic statement was that at no point during this process had privacy been looked at explicitly. And he also stated that when we heard from consumer representatives in the field that we had made no effort to hear from consumer representatives who are explicitly interested in privacy matters. He therefore felt that the report should not be released. Now he is right in the following sense: That what we put together are panels. We did not explicitly say who were privacy experts around Medicaid Managed Care, because at the time there was a very active ongoing Privacy Subcommittee meeting, if people remember back two years ago, and we didn't explicitly state that we wanted to have Medicaid Managed Care privacy people speak to us. Nonetheless, we heard a lot about privacy from people that we heard from. The woman in Phoenix from Families USA talked a lot about what it was like for her as a parent to have people repeatedly come and ask her to again define the disabilities that her child has, and again justify why her child should get Medicaid Managed Care. And so we did not make an explicit effort to hear about Medicaid Managed Care privacy related concerns, but we did hear about privacy. With that stated, I think it is very fair for Bob if he wants to write a dissenting opinion, or an expectative opinion, to explain his point of view that he should be able to do so. Paul?
PAUL NEWACHECK: I agree that he should be able to do so if he wants to. I think it would be preferable if he didn't, and if Simon didn't as well. I think it weakens the report and increases the likelihood that it will just gather dust if there are dissenting views presented in it. So if we can address those issues at all I think we ought to try to. If we can't, then that's I think the best route. I was looking through the Recommendations Section at our lunch break and there are at least five instances in the Recommendations Section where we refer to the privacy/confidentiality issue, and we did that explicitly I think to accommodate his interests and his concerns that he raised last time, too, as well. So I feel like we have tried to be responsible in terms of addressing this issue. I'm not sure about Simon's issues, but his, to the extent that we can, given that he raised his concerns after all the hearings were over and after the testimony was taken.
CHAIRMAN IEZZONI: That would be his exact point, though.
PAUL NEWACHECK: Yes, I understand. But I'm just saying I think we tried to do a responsible job to the point where it almost looks odd in the recommendation section because it kind of comes up at the end of each one. And remember, we have to be careful about privacy and confidentiality. So it is almost like we are going overboard in a way, and I think we are, actually.
CHAIRMAN IEZZONI: Well I would like to come back to you for some recommendations about exactly what to say to Bob in the public hearing, in the public meeting tomorrow. Vince? VINCENT MOR: I actual don't believe that Bob will be satisfied with modifications.
(Many voices in agreement.)
VINCENT MOR: I went through--I was here last year--in June, heard him. We had a long discussion about these issues. If he doesn't think the report should be released, that is an issue not just of process but it's also one of politics, I believe. I was also very pleased that the actual statutory basis for the right that states and the Federal Government have for collecting data are very, very explicitly outlined in every one of those tables up front. So really the argument is with existing law and statute and not with the report. And that is a different issue, which is perfectly appropriate for him to say separately, because I think really it is that this is a major danger in the absence of privacy legislation. Bingo!
CHAIRMAN IEZZONI: Bob is very much focused on the process that we went through in this exercise. So he has not explicitly articulated, at least not within my hearing--he may have done so in June--that his concern is with the legislation. Because, Frankly, I think Kathy called him on it on the May 6th meeting when Kathy said: But states already have the legal right to collect these data. And he hadn't realized that. I think his concern is a lot more with how our committee has gone about this entire enterprise. I cannot argue with him on that point.
VINCENT MOR: But it wasn't the charge of the committee.
CHAIRMAN IEZZONI: Right.
BARBARA STARFIELD: What do you mean you can't argue with him on this?
VINCENT MOR: He had every right. We had it open as part of the charge of what this task was for this subcommittee, it was open and approved by the entire committee, and there was no language in that charge explicitly organized around privacy. He had the option. We probably would have been very happy to have incorporated it at the time.
BARBARA STARFIELD: He said this morning that we made "no effort".
CHAIRMAN IEZZONI: That's right. That's right.
BARBARA STARFIELD: And, I mean, you know, other hearings don't go out and find the opposition, either. I don't think what we did was wrong.
VINCENT MOR: No. Our issues were a very different focus.
CHAIRMAN IEZZONI: Okay. Subcommittee, I need some advice on how to deal tomorrow in the public meeting in the morning with Bob because I think that he will continue to raise his concern, and we have to figure out what we want to say in response. My feeling is, Paul, that it wouldn't actually weaken the report to have Bob write a page about it, because, frankly, we all agree that privacy is important.
PAUL NEWACHECK: Well I'm actually not concerned that he would say just that, but that he might say the report is biased and incorrect. I mean, he might put a lot of negative verbiage in there as well as just pointing out that privacy is important, and that we need to pay attention to it.
CHAIRMAN IEZZONI: Other suggestions, then?
VINCENT MOR: I don't think we have any other choice but to say that, you know, you want to write a dissenting piece, then the committee has to vote on your dissenting piece just like the committee has to vote on the report. And if he says that there was no effort made to incorporate issues of privacy, then we just have to say I'm sorry that's not correct. Everyone in the room, everyone on the committee had the opportunity to specify, to make modifications to the charge on this particular task and that was not taken up at that time.
PAUL NEWACHECK: I think we have a standing rule, though, that says that people can write dissenting opinions without getting the rest of the committee's approval.
CHAIRMAN IEZZONI: I don't know. Do we?
PAUL NEWACHECK: I thought we agreed to that about four or five meetings ago.
CHAIRMAN IEZZONI: I'm not sure.
CAROLYN RIMES: It came back up with John in a conversation, but I don't remember closure.
CHAIRMAN IEZZONI: Actually, it would be
good to get a read on this. Dale, do you know where Marjorie is right now?
DALE HITCHCOCK: I can find out.
CHAIRMAN IEZZONI: Because it would be good to get a read from this from here, because I think the issue is how hard do we work tomorrow morning to try to keep Bob from writing a dissenting opinion. Or, do we just say, look, he has the right to do it and let's just let him do it. And frankly, I think the weight of the report is such, with all the statutory language in it, and all the clarity around that, that that would overwhelm the objections.
BARBARA STARFIELD: I think Paul's right. I think that he has a right, but he doesn't have a right to make false statements. Now he did have the opportunity to make suggestions, which he did for the privacy--
CHAIRMAN IEZZONI: Could I ask tomorrow morning that other committee members weigh in with me on this? Because I am perfectly happy to go toe-to- toe with Bob, I've done it before, but I think that it is important to have other committee members back up our position on this.
ELIZABETH WARD: I certainly think you need to meet with John and decide whether we--
CHAIRMAN IEZZONI: Yes, and--
ELIZABETH WARD: --to me, it is less important to get out. I don't know what one month or three months difference could possibly make at this point in terms of who out there may use this. What I think is important is how the process is managed at the time we take a vote in there, and if there is going to be a dissension that that is managed and managed well, and that John decides how he wants to do that as Chair of the National Committee.
CHAIRMAN IEZZONI: Well, we're--
ELIZABETH WARD: If it is a matter of postponing it tomorrow morning one more time, I would rather do that. I think there are some really critical means for stating how we came about doing it, what the goals of the report were. I think we started off with goals that were going to be biased in a way that isn't antithetical to how the confidentiality committee is going. I think that in hindsight we could have chosen a very different way to go there. But I think we need to be up front and put those kinds of things into the preamble as we present this for a final vote. We were going based on the bias that logged in collecting these billions and billions of dollars worth of data exist. Rather than going in and starting off with the question: Should HCFA still be allowed to do that? And as everything gets automated, what is the risk to that? We said, we are just assuming that that is there, and what we are looking at is how to improve that process, not question that process. And that is where we diverted. I think we took off on entirely different routes. We were bound to have this conflict.
KATHRYN COLTIN: I think that we need to reiterate to Bob what the state of data collection for the Medicaid population was before Managed Care.
CAROLYN RIMES: That is good point.
KATHRYN COLTIN: What happens with Managed Care to change it. And that basically we are just trying to bring the pendulum back to say, okay, can we preserve what state medicaid agencies had always had available to them, but which was threatened with the increased enrollment of Medicaid eligibles in Managed Care, the loss of data that they could potentially experience, and that this was intended as redress, essentially, for that problem. So I think that is one piece. And I think you're right. He wasn't aware of that, and he may since have forgotten it as well. The other thing is the Notifiable Disease issue, is the other data that we're talking about that is personal data besides the encounter data. And again, we are not talking about data that they didn't have before and that they don't still have the statutory right to collect, but rather the fact that Managed Care intervened in that process through some of its contractual relationships with laboratories that made it difficult for states to continue to have the data stream that they have previously had. So it is not like-- He seems to think we are trying to collect a lot more, and what we are really trying to do is to fix a problem to bring things back to where they had been for at least a decade. So I think that, first of all, is a perspective that needs to be made clear to him. I think the only recommendation that we have in this report that relates to the collection of new data is the recommendation around collecting data on patient experiences. There is not a confidentiality issue in that particular recommendation because the letters that go out to the beneficiaries make it quite clear that it is voluntary and they don't need to provide the data.
CHAIRMAN IEZZONI: Marjorie, we are trying to figure out how to--I would actually like Elizabeth, if we could, and we're not going to push this if we can't, but if we could I would like to have a final vote tomorrow so we could get the Medicaid Managed Care Report off our plate, because we have other things that we would like to move on to do. And yet we have this issue about Bob, and possibly Simon--we're not sure what Simon's issues are--but we know explicitly that Bob recommended this morning that the Report not even be released.
MAJORIE GREENBERG: Right.
CHAIRMAN IEZZONI: And so I made a proposal a few minutes ago that we allow Bob to write a dissenting opinion, and there was agreement, or a consensus actually among other subcommittee members that that wasn't appropriate because Bob in fact had had opportunities to weigh in on the scope of work through when we presented our subcommittee's scope of work which laid out explicitly what we were going to be doing here. Also, Bob doesn't understand the full context and so was making opinions, as Kathy has ultimately stated, that really do not understand the context of the opinions. So what we wanted to get from you was a sense of, number one, is there now a proviso whereby anybody can write a dissenting opinion and have it appended to a report that comes out of NCPHS? And Elizabeth had, I think appropriately, suggested that I speak to John Lumpkin beforehand about how to handle this tomorrow, but since you are here maybe we could have you weigh in on both issues.
MAJORIE GREENBERG: Yes. I think we would want to, whatever discussion or kind of tentative agreement we reach here, would want to discuss it with John before the meeting. Because one thing that--well, to the extent a few people didn't receive their agenda books, I would take responsibility and so they didn't have the report--I think that made it hard to press for other comments. Because I think the majority of people had seen the report, well had seen it a few times, but had seen this version, but we didn't ever find out whether people had other problems with it. It was just kind of tabled. I would say, based on the experience with the annual reports, the HPRA report and the 1996-98 report, that at least at this time there is a precedent that any member could write a dissenting view, and that this will be made part of the report. This is not something that we have had in the past, because it wasn't I guess desired by any member to do so. Also, when we agreed to that, when the committee agreed to that in regard to the 1996-98 report, I think that John Lumpkin said he didn't think that was necessarily precedent-setting. But on the other hand, I would say that I think at this stage it is. That's my understanding. And, that probably it would be difficult and possibly futile to attempt to revise the report sufficiently to satisfy Bob's concerns. I think I don't-- I mean I totally respect your desire to get this substantial piece of work out and behind you, but also out. I think the subcommittee has probably put as much into it--
CHAIRMAN IEZZONI: Right, as we can.
MAJORIE GREENBERG: --and staff, and contractors, as you really feel you are able to. So I think that if there are any--and I don't know what Simon's concerns are going to be. Most of the concerns that were expressed at the June meeting did have to do with the draft contract language, which is basically alluded to but certainly not made a part of the document. So unless there are particular areas that you feel you would like to revise today--
CHAIRMAN IEZZONI: Well there has already been the point that privacy concerns are raised about five different times in our set of recommendations, but obviously that was not sufficient to assuage Bob's concern. And so--
MAJORIE GREENBERG: Well one concern I had about it, because I was very attuned to that as I was reading it, is that it frequently refers to "current privacy confidentiality protections," or whatever, which I think we all agree are inadequate. So in some ways that doesn't acknowledge the concern, but--
CHAIRMAN IEZZONI: Well when we--
MAJORIE GREENBERG: --acknowledges the concern, but it doesn't make people feel that much better when you talk about the current situation, which is not that good.
CHAIRMAN IEZZONI: Well what I was going to propose is that actually after this initial discussion, we need to hear from Clyde, but then we go through each point, point by point, as a subcommittee, even though it would be time-consuming to do that, that we do that, and that at that point maybe we can highlight these areas and, Carolyn, we can change the language slightly to say that we need to build in future protections that Congress or the Secretary may develop for privacy and confidentiality and data security.
MAJORIE GREENBERG: You might also want to say up front that it really is not the intent of the committee to expand--
CHAIRMAN IEZZONI: Right.
MAJORIE GREENBERG: --the data collection, or data requirements that are currently extant; but to try to rationalize them--
CHAIRMAN IEZZONI: Yes.
MAJORIE GREENBERG: --to try to standardize them in a context that could be beneficial to the program.
ELIZABETH WARD: It could decrease the amount of stuff that in fact is collected that we found, over and over again, is not being used by HCFA.
CHAIRMAN IEZZONI: Right.
ELIZABETH WARD: That was my particular interest in looking at this, because you've heard from the states and my experience is that because they just sort of say we want everything, and then eventually we may be able to use it, and that's how you get inappropriate bunches of data being collected on people that are identifiable that you put at risk because they don't have a rationale from the beginning that says this is why we, HCFA, have to have the following data items. We have standardized them. We have told you what we're going to use them for. And then you walk through and you can lay the confidentiality over that.
MAJORIE GREENBERG: And that could be emphasized in the executive summary, also. But I would think in terms of going through it as you wanted it, then the minority or dissenting opinion.
CHAIRMAN IEZZONI: Yes.
MAJORIE GREENBERG: And it could be more than one. I don't know.
CHAIRMAN IEZZONI: So, Marjorie, when you go back up and see John, just tell him that he should call me tonight and we should talk.
MAJORIE GREENBERG: Okay, I will touch base with him. He may come down here later.
CHAIRMAN IEZZONI: Right. For the Quality Work Group. Okay, so thank you, Marjorie, very much.
MAJORIE GREENBERG: I can come back again. I'm just upstairs.
CHAIRMAN IEZZONI: I think that we do not have anything to gain by fighting at length with Bob. I think that--let me just ask the subcommittee, do you want me to try to argue with Bob at all? Or shall we just say--
PAUL NEWACHECK: I think it is worth saying that we attempted to address these issues as carefully as we could and, given our--
CAROLYN RIMES: I liked also your perspective.
ELIZABETH WARD: Well I think Katherine's, too, on whether each of us can individually state that. But I think that is part of what has been so confusing to the public, why we're sitting in that room trying to figure out why two pieces of the committee are fighting with each other. I could not understand it, and I think that is Kathryn's comment, too, that we started off in an entirely different place and went down a road. And if we had wanted to--and that is why I agree there is no point in trying to sort of fix that something that--because we started it in two entirely different places.
CHAIRMAN IEZZONI: Well I think Kathy's point needs to be made in the very opening paragraphs of this report, Carolyn, because you made it at the May 6th meeting and it was a beautifully made point at the May 6th meeting and it actually caused Bob to pause.
KATHRYN COLTIN: For a minute or two.
CHAIRMAN IEZZONI: Yes, that was helpful. And so I think that actually, Carolyn, we really do need to make that point.
ELIZABETH WARD: The historical perspective is not there.
CHAIRMAN IEZZONI: The historical perspective that--
ELIZABETH WARD: Why did we decide to do this report.
CHAIRMAN IEZZONI: Right.
BARBARA STARFIELD: I guess I'm sort of not sure. One of the reasons it's so difficult to deal with Bob is you don't really, or at least I don't really know what his agenda is. You know, it may sound like privacy, but maybe he doesn't want us to collect data. Maybe he doesn't want anybody to collect information. I mean, it sort of sounds to me a lot of the time that that's the case. So I think the historical argument is not going to weigh important to him. We say we made a mistake in the past--
ELIZABETH WARD: No, I don't think we made a mistake--
CHAIRMAN IEZZONI: But it gives context to why--
BARBARA STARFIELD: It gives the reason why we did it, but it is not going to make him happy.
CHAIRMAN IEZZONI: Right.
ELIZABETH WARD: I am talking about the public who is trying to understand the conflict, and that we started off assuming that HCFA had this enormous authority to collect enormous amounts of data, and that the states individually also have that obligation. We didn't start--he would have preferred us weighing in on that subject saying they ought not be able to collect that amount of identifiable data. We didn't go there. We said we're starting over assuming that that legal process is in existence and isn't going to change. How can we make that a more rational process? If we could at least explain to our public what the sort of difference of view is, in sum, if it is possible.
CHAIRMAN IEZZONI: All right. Why don't we now go through-- Now, Clyde, you were supposed to be on at 2:30, but you're sitting here.
CLYDE TUCKER: It's okay with me. I can start whenever you want me to.
CHAIRMAN IEZZONI: Why don't we have Clyde talk to us now, because this is a clean break. Then after Clyde finishes, let's make a plan that what we will do is maybe take a brief break and then go through each recommendation. Because there are some words and phrases in them that I think it would be good to just tidy up a bit, and then hopefully we will be able to move to Kathy's Work Group at 4:15. Okay? So--
DALE HITCHCOCK: And Clyde needs the overhead, I guess, right?
CLYDE TUCKER: Well I can do it with or without. It might be easier with, but I can do it without.
CHAIRMAN IEZZONI: Oh, so we are waiting for the overhead?
DALE HITCHCOCK: We have it, but I have not set up in this room before. I think we have a screen, or a wall that can be used. We have a wall for a screen.
BARBARA STARFIELD: In our Child Health Status we asked if you could see the blackboard, and the kids didn't know what we were talking about.
(Laughter.)
CHAIRMAN IEZZONI: Because there are no blackboards, that's right. That's very cute.
(Laughter.)
ELIZABETH WARD: Take out your slide rule.
PAUL NEWACHECK: I once said, I think I have a mimeograph of that somewhere, and they said "What's that? Mimeograph?"
BARBARA STARFIELD: Language that doesn't exist anymore.
PAUL NEWACHECK: Right.
CAROLYN RIMES: And it's a shame, too, because just think what they're missing with that wonderful smell.
ELIZABETH WARD: And the work you got to do in grade school to go to the office.
CAROLYN RIMES: That's true. That was an honor, or in my case it was so I'd be quiet for 10 minutes.
(Laughter.)
(Pause for setup.)
CHAIRMAN IEZZONI: Okay, Clyde Tucker is from the Bureau of Labor Statistics. Clyde, could you tell us a little bit more about yourself, like 30 seconds' worth, and then talk to us about the OMB status.
CLYDE TUCKER: I am a Senior Survey Methodologist at BLS. I have been working on the new standards for race and ethnicity since 1994. In October of 1997, the new standards were put out, and since then I have been chairing some groups that are working on developing guidelines for implementing the standards.
CHAIRMAN IEZZONI: Thank you.
CLYDE TUCKER: Olivia told me to talk about everything, so I will try to talk about everything in 30 minutes.
(Vu-graphs.)
OLIVIA CARTER-POKRAS: I mentioned the fact that some members had not been to his earlier presentation.
CLYDE TUCKER: Okay, I am going to very quickly go through the main changes in the standards. Is that--
CHAIRMAN IEZZONI: That's perfect.
CLYDE TUCKER: So the main changes are that we have added an additional category, Native Hawaiian or Pacific Islander. That is breaking apart the Asian and Pacific Islander category. We now with our instructions allow respondents to mark one or more races, and we ask the Hispanic ethnicity questions separately and before the race question. And finally, that there have been some terminological changes. "Hispanic" is now "Hispanic or Latino", "Black" is now "Black or African American." Those are the main name changes. These, what I am talking about mostly will be about self-identification of race.
(Vu-graph.)
Okay, when these guidelines were, these standards were put out, there was a need expressed that we develop some help for people to implement the new standards. And so in the October 30th Federal Register Notice there was some discussion there about users who are in areas of enforcement or have legislative or constitutional mandates to carry out, particularly redistricting and reapportionment, those kinds of things, enforcement of equal opportunity laws, and voting rights. And we gave a number at that time, a number of different options that they might use.
What is currently expected is that the Public Law File that is used for reapportionment and redistricting will contain 63 combinations, counts for 63 combinations, at probably the block level. There basically are 63 combinations if you go all the way up to the possibility of being of five races. I can tell you that in a Census dress rehearsal I don't believe they ended up with all 63, but they ended up with somewhere around 50 being in just those two main sites. In addition to that, we proposed in the Federal Register Notice that a set of groups be put together to look at implementation guidelines for monitoring economic, social, and health trends, including complying with antidiscrimination programs, sample design, and survey controls for weighting. The denominators that needed to be developed were vital statistics for property rates, and also for being able to do analytical work involving trends bridging to the past to deal with distribution.
(Vu-graph.)
Now there were three groups set up. One was the Policy Group, which basically reviewed the work of the other groups and provided information about how the new standards would affect their agencies. The second group, the Procedureal Design Group, has been working on a questionnaire format, questionnaire design, instructions for interviewers, respondents, and training of interviewers, and also the methods for reporting aggregate data. The Tabulation Group has been working with developing bridges to, bridges-to-the-past distributions and also how to report future tabulations under the new standards.
(Vu-graph.)
Now we know that in terms of looking at the past, we know that there will be possible changes in a data series, demographic characteristics for Hispanics, for instance could change. Counts of demographic and substantive characteristics of races could change because of the pulling out of these combination categories from single races. An example for Hispanics is that on the current population survey we know that instituting the new standards is probably going to increase our coverage of Hispanics from 83 percent to over 90 percent. The people we are going to pick up are going to be actually higher SES than the ones we've been picking up. So there is going to be some slight change there.
(Vu-graph.)
Okay, as I said, the counts that are going to likely change are going to be for smaller races, particularly the American Indian populations. The demographic characteristics could change for smaller racial groups, and I'll talk to you a little bit more about that, and distributions of substantive characteristics by race could be affected.
(Vu-graph.)
Now this is the table of contents of the guidelines which is currently available on the OMB Web Site. It is quite extensive, and the results on procedural design and on bridging are basically summarized in Chapters 2 and 5, but the full technical reports are available in Appendices B and D. I am not going to talk too much about the others. I am going to spend most of my time talking about tabulation and comparing data of all the new standards and a little bit about procedural design, but let me say a couple of other things. Under the tabulation, because the decision for how the PL file would be laid out, what has just been made, we're not yet sure about how that is going to affect all the tabulation. We know it is going to change the way they were originally planning on tabulating the race and ethnicity in the Census, so that part is sort of being revised. The same is true for the tabulation information and some of the examples given in the report for other surveys and administrative records. So that whole chapter is still in flux. Also that chapter also includes, actually Chapter 4 is another part that I won't be able to talk about much. In Chapter 3 there are examples in the report, but as I say some of them are out of date.
In Chapter 4, there is some information on redistricting. That is changing. Equal Employment Opportunity we still haven't settled on the option that is going to be chosen by the EEOC. Vital Records and Intercensal Estimates, there are supposed to be discussions going on between Census and NCHS on updating Intercensal Estimates using Vital Records, and I'm not sure where that stands at this point. And this last part has just been--we've written part of D but we haven't finished it. That was not in the old version. So let me talk about some of the other parts, particularly procedural design and bridging.
(Vu-graph.)
So the first thing I want to talk about is cognitive testing that's been done for implementing the new classification standards. That has been done in order to develop questionnaire formats, in order to develop instructions, and training procedures.
(Vu-graph.)
So it has been used for coming up with guidelines for data collection, particularly for self-identification questions across various modes. That is, whether you are collecting it in person, by mail, by telephone. And also to work on developing aggregate reporting forms for institutions such as hospitals, businesses, prisons, schools. And there is also--this has not happened yet--but we are in our final phase of work that will lead to development of guidelines for interview of training and collection instructions.
(Vu-graph.)
VINCENT MOR: Is that just for Census, or for others?
CLYDE TUCKER: No. It covers all Federal Service. I mean, it is meant to be used by all Federal Service.
(Vu-graph.)
We are working with the Field Division at Census in developing this. You have to understand that most of the Census is collected without interviews. So most of it is mail questionnaire. Okay, some problems that we were worried about: Misinterpreting categories; Misunderstanding or ignoring instructions to mark or select more than one group, if appropriate; And the confusing concepts of what ethnicity is, what race is.
(Vu-graph.)
Now the standards do not give definitions for "ethnicity" and "race." What they do is they give definitions for categories of race, and for Hispanic Ethnicity.
(Vu-graph.)
So here are some examples of some self- administered questions that we believe will work in the field. This is a short version, just a yes/no Hispanic or Latino question. It could be followed up with a question asking more specific information about what group they're from. Then the short version of the race question. These would be probably most appropriate on the telephone, and here are examples of what were used in the Census dress rehearsal.
(Vu-graph.)
This is the short form. They are boxes. You can't see it. They don't Xerox well. They are boxes for writing in responses. You can see in this case they use a longer version of the Hispanic question, and obviously a more extensive question for race in terms of the categories.
(Vu-graph.)
This is the race question that was used on the long form. I think it was the same Hispanic question.
(Vu-graph.)
Yes?
SUSAN QUEEN: Does the Census do the special exceptions?
CLYDE TUCKER: Yes. They've always been given that, and they will have it on there this time, too.
SUSAN QUEEN: Because most surveys don't.
CLYDE TUCKER: Most surveys don't, and they would have to ask for an exemption. What most surveys do is they have an "other" category, which they do not read or show to the respondent. And if the respondent says "I'm 'other'", then they will be asked to specify what that is.
(Vu-graph.)
Okay. So we have already completed Phase I of our research and testing and self- identification questions. We did it all in the D.C. Metro. We currently have a contract with Westat to work on collection in other geographic locations. I think we are going to Los Angeles, Honolulu, Tulsa, Oklahoma, and I think New York City. Those are our cool places, if someone is anxious to go to Honolulu.
ELIZABETH WARD: Compared to some of the other places--
(Laughter.)
CLYDE TUCKER: Pardon?
ELIZABETH WARD: Compared to the other sites.
CLYDE TUCKER: Well, I like New York City. I'm not sure I would want to go to Tulsa, but it's hard to find American Indians. We have recruited our subjects and brought them into the laboratory. We conducted either a face-to-face kind of interview in the lab and asked them questions and followed it up with more indepth probing. We also conducted telephone interviews in the lab where we brought people in. Occasionally we called them at home, but we like to bring them into the lab, put them in another room and conduct if over the telephone, and then get back together with them.
(Vu-graph.)
This is a breakdown. We conducted these interviews with 44 people. We had trouble getting enough American Indians, as you might expect. We are talking about pure American Indians as opposed to those of many races, and the same thing with Asians. That is why we are now going out to the rest of the country. We will not be doing Blacks or Whites because we think we've addressed the issues for Blacks and Whites. So to keep our costs down, we are focusing on these other groups.
(Vu-graph.)
Here are some examples that we are using in our testing. I am not going to spend too much time going through these.
(Vu-graph.)
I might say that in a minute I will tell you some of the problems we have run into in administrating this. We are testing different versions, as you can tell.
(Vu-graph.)
Here is an example of some telephone questions.
(Vu-graph.)
And you can go as far as you want to. I mean, you don't have to even ask Asians about particular countries, but it is up to you. We tested out long and short versions.
(Vu-graph.)
Our conclusions are that we will use an instruction that allows but does not require multiple responses to the race question. So, usually rather than saying you can mark more than one, we usually say you can mark one or more, or select one or more, something that doesn't appear to be a command. We considered using instructions to answer both race and ethnicity. That's what is going on in the Census. They start out--this doesn't show up here--but at the beginning of the section Race and Ethnicity, the Census will have something that says please answer both questions, because we considered providing definitions. Currently we are not providing definitions of the racial categories. The problem on the Census is there's just not enough space. But some surveys could, if they wanted to, provide definitions, especially if they have enough space. Typically the problem is people have less space than we, not more space. We contain the wording and format according to the mode. I'll give you an example of that in a minute. Use follow-up questions or write in options where we need more detailed information, and of course follow the standards.
(Vu-graph.)
An example of where we have problems would be-- in person this is not a problem, but on the telephone we have a problem when we read the list of races and we go: White, Black or African American, American Indian, and go on down the list, because when they see Black or African American, on the telephone it sounds to them like two separate categories, unless we are very careful as to how we're reading that. So they can't decide. If they're both Black and African American, which one are they supposed to mark? Well it's the same category. We have the same problem to some extent on Hispanic or Latino, but that really has not been as much of a problem as having Black or African American. The problems we have do differ by mode. We also can't present a very long list on the telephone. We have to break the question into three or four questions in order to get it on the telephone, whereas otherwise we could just do it like the Census in one question.
(Vu-graph.)
Okay, testing aggregate forms. This has been the hardest part of our work in our procedural design. We have convened an expert panel that is expert in the sense that they are experts in establishment surveys to review and discuss our reporting forms. We have developed three different kinds of reporting forms. We have recruited subjects. We have conducted interviews in the laboratory, and we've also conducted interviews on site at establishments.
(Vu-graph.)
The problems we have, particularly in conducting these at establishments, is that the data that they have that we ask them to bring to the table are not collected like under the new standards so it is very difficult for them to get any practice or understanding of what they're supposed to do with the new data since they don't have it yet. We can make up data in the laboratory, and that is what we do. Even then, we have found there0 have been a lot of problems primarily because people in establishments for so long have considered Hispanic a race, even though the old guidelines in large part were separating that. Still, now we are really pushing them for self-identification and we use the two-question format, and they are very confused by that because they are not quite clear what we're doing with ethnicity. Similar agencies and organization of cells make it hard to develop. Everybody is collecting data in some different kind of format, at least at the individual level. When they report it to the Federal agencies in many cases we have uniformity. The task is hard for them to actually aggregate the data. Testing requires a lot of pretesting. One of the problems that we have is that in using these forms, we have used these forms, aggregate reporting forms, a paper and pencil task, whereas for many, many establishments now they can just computerize this task and it will be much easier for them. Of course they have to be willing to expend the money to change their data formats, and also change their programs for tabulation.
(Vu-graph.)
Okay, here are our Guidelines for Aggregate Reporting so far. We're still working on this. We need to make sure we have them understand what ethnicity is, this is what race is, and understand what a single versus a monthly report is and how to deal with it on these reporting forms. Allow for reporting of every combination, if possible. Or, in collapsing multiple responses we need to work out some instructions for how exactly we go about doing that. You need some way of dealing with people who don't report Hispanic origin, or don't report race. There has to be a category for them to go in. Basically we need to spend more time developing these forms.
(Vu-graph.)
So overall in terms of procedural design, we are testing self-identification questions in other geographic locations, as I mentioned. We are going to be reconsidering what we're doing with aggregate forms. And we are going to be, probably by November, developing and testing interviewer training and interviewer responding and instructions.
(Vu-graph.)
Tabulations. There are two kinds of tabulations. There are future tabulations using the new standards ongoing into the future, and then there's bridging to the past, which involves taking the new data we collect and somehow trying to move it into the old distribution so that we can look at trend analysis.
(Vu-graph.)
So let me talk a little bit about bridging first.
(Vu-graph.)
Here is the guiding principle we use in the research we did on trying to develop effective bridges. If we can find data sets in which they simultaneously, or almost simultaneous, ask the same people their race under the old categories, under the old system and their race under the new systems, we can use those two to try to develop a distribution, a way of allocating the new data back to the old distribution. The problem we have is that what we want to do is get it to match that old distribution right now. They are collected simultaneously, therefore there is no change over time. We don't have a variance in the sense that we don't have it from two different groups. We don't have that problem. So we have a repeated measurement. If we can get a bridge that really matches to the old distribution for these people, then we will be able to, from this point forward, use that bridge and measure true change as it goes on under the old distribution. So that is what we have been working on.
(Vu-graph.)
Now let me say that there are a number of different bridging alternatives that we have looked at, and they are really classified according to whether or not a person's multiple responses are assigned to a single category or not, and whether they are assigned by a fixed rule or by a probability method.
(Vu-graph.)
Okay. Let me go through these.
(Vu-graph.)
The first group is called Deterministic Whole Assignment. That means that there is a fixed rule and they are assigned to only one category. One method of doing that is to assign a person to the smallest group they mentioned. So if they mention White/Black, they'd give them Black. A lot depends upon however at what geographic level you're talking about making this assignment in. On a particular tract, or something like that, you could find there are more Blacks than Whites so you might go White if you did it that way, but at least at the national and state level chances are it is going to be the smallest groups we can think of. Another possibility is assigning it to the largest group other than White. The largest group that is, period, which means that most people would be assigned to White if they named White. If they named a combination with Black but not White, Black would be the most likely assignment. Then the plurality assignment. That is, looking at some data that we have on how people who respond to this when they are asked to choose which is their primary race, which one they choose, and make the assignment that way. That is, if you are Black and White, we know that the probability is 60 or 70 percent that you're going to chose Black. So use that as the decision rule.
(Vu-graph.)
Now the deterministic fractional assignments. In this case we're talking about assigning to more than one category. So one possibility is if you say you're Black and White, assign a half a response to White and a half a response to Black. Or based upon data we have from the HIS to make a fractional assignment, the fractions will depend upon the probabilities of choosing Black or choosing White for people who are Black and White.
(Vu-graph.)
In looking at bridging, we had a number of criterion we wanted to look at. The main criteria was we wanted to be able to measure true change over time. Others were minimizing disruptions to the single race distribution list. That is, could we develop a bridging method which would bridge back to the old distribution without disturbing too much what the single-race counts were on the new data. This will allow us to actually during a transition phase be able to move more easily from the old distribution to the new distribution over time. How widely could this dispersion method be applied? Whether or not it would meet confidentiality and reliability standards? Whether it is statistically defensible.
(Vu-graph.)
How easy was it to use? Going along with that is what skill was required to be able to do it? And how easy were these bridging methods to understand for both the data users and the general public? And also we weren't particularly concerned so much with bridging, but in our future tabulations if we want to make sure whatever tabulation we did, and also for bridging to the extent we could, that we were able to use the maximum amount of information from the respondent's answer.
(Vu-graph.)
In carrying out this work, we had three data sources in which the same people were asked almost simultaneously about their race under, usually, the old question and the new question, the National Health Interview Survey from 1993 to 1995, the May 1995 CPS Supplement on Race and Ethnicity, and a survey done by Washington State.
(Vu-graph.)
The analyses we did was to look at the produced distribution using allocation under one of these bridging methods and compare it to the referenced distribution under the other-- that is, the way they answered under the old standards. The second thing we looked at is how much misclassification was there of race when reallocating? That is, we allocate them to a race under the old standards that they didn't choose. Then finally we wanted to look at some outcomes. You know, how did this allocation procedure and this new distribution affect substantive characteristics?
(Vu-graph.)
Okay, our results. Measure of change over time. We compared the reference distribution to the new and we found that the closest matches were for the largest group method, the plurality method, and the fractional assignments methods. The other methods, the ones used in the Largest Group Other Than White, those did not work very well in terms of matching back to the distribution. We couldn't get back to it as well. Overall, misclassification was was greatest for the Smallest Group and the Largest Group Other Than White methods, but they also had the smallest misclassification when you looked just at American Indians. The American Indian/Alaska Native Category is the most problematic in all of this largely because it is the smallest. It also has the largest rate of intermarriage.
BARBARA STARFIELD: I'm sorry? Largest rate of what?
CLYDE TUCKER: Largest rate of intermarriage.
CHAIRMAN IEZZONI: Intermarriage.
(Vu-graph.)
CLYDE TUCKER: Outcomes. Well, we looked at health outcomes. We looked at the percent having no insurance, percent having poor or fair health, and percent of children with single mothers. They were largely unaffected by bridging except that the American Indian category for no health insurance which was affected. Possible effects on the American Indians for unemployment and labor force participation, we looked at that. But the main thing we concluded was that the affects will vary depending upon the characteristics we're looking at, and of course we don't know how they're going to be affected.
(Vu-graph.)
The fractional methods, plurality method, and largest group method were closest to their single race proportions under the new standards and the Other Whole Assignment methods showed substantial differences for the American Indian category. In general what happens under the Smallest Group or the Largest Group Other Than White is that we blow off the American Indian category because there are so many people who on the old method would have said White who are now saying White and American Indian, or Black and American Indian.
(Vu-graph.)
We found that the Largest Group and the Equal Fractions Methods were least sensitive to the context--that is, geographic level. The Smallest Group and the Largest Other Than White varied quite a bit by geography. The Plurality and the HIS Fraction methods were more difficult because we only have a national distribution currently based on the HIS, and so we can't carry it down below the national level. Everything has to be allocated according to how those in the national survey responded when they were asked their primary race. Confidentiality and Reliability. This is not really a problem because we don't really have--we don't really have a confidentiality problem because basically we're not doing anything differently than we did in the past. Now to the extent there was a confidentiality problem in the past, we are going to have that. But in terms of reliability, we were a little concerned about the reliability for the Smallest Race Groups and any bridge distribution to come up with. All the methods other than Equal Fractions were based on observed distribution. The Smallest Group and the Largest Other Than White methods were based on historical record, and the others were based on minimization of error. The Fractional Assignment requires duplication of records, or multiple variables. It's very complicated--some would say, very complicated to use. Equal Fractions is relatively simple. The other methods are pretty easy, except that they could differ at different geographic levels. So you can imagine somebody who is in California being classified as one thing, and if they move they'll be classified as something else.
(Vu-graph.)
Deterministic Whole Assignment requires relatively little skill compared to the Fractional Method. So if we are going to do something where we just do Whole Assignment based upon something like the Plurality method, that would be easier than having to deal with fractions. But everything is relative. It depends on how skilled you are as to how difficult this is, and how much money you have. Understandability and communicability. We were concerned that the Fractional Assignment may be difficult for people to accept; we're splitting people into more than one race, so having parts of the response. Although at the aggregate level it sums up to a nice distribution.
(Vu-graph.)
Future work we're doing, we're now comparing the previous combined distribution. That is, we have been able to manufacture a distribution that is similar to what people have done before when they were just asking one question, including Hispanic as a race and we're going to also do bridges back to that and compare and see what happens there. We hope to be able to do the same analysis we have already done on the dress rehearsal data from Census. Then we hope that in the end we will be able to devise suggested strategies for how to do this bridging to different groups.
(Vu-graph.)
Okay, I think I am done.
CHAIRMAN IEZZONI: Clyde, that was actually remarkably clear and really an excellent presentation of very complicated issues. So thank you for that. I am reassured that you are on the case, but how long do you have to figure this out?
CLYDE TUCKER: It is unclear. It can't be more than 12 years because I'm going to retire.
(Laughter.)
CLYDE TUCKER: After that, they will have to hire me at an exorbitant consulting rate.
CHAIRMAN IEZZONI: We won't ask if our Government does that.
(Laughter.)
CLYDE TUCKER: I guess I would say that we hope to put out another set of provisional guidelines, an updated one with a lot of decisions and more of the research, the bridging research we talked about and the other procedural design research by the end of the year.
CHAIRMAN IEZZONI: We should get that. So will that be guidelines that will be for open comment?
CLYDE TUCKER: They will be for comment. I'm not sure how the comments are going to be requested. I don't know that. But I do know that after that I doubt we will put out anything else until we see Census data.
CHAIRMAN IEZZONI: Okay.
CLYDE TUCKER: So that will probably be at the end of 2000 or sometime in 2001.
CHAIRMAN IEZZONI: Okay.
OLIVIA CARTER-POKRAS: Cathy Wohman said we could certainly call for the committee to send in comments at any point in time.
CHAIRMAN IEZZONI: Yes.
OLIVIA CARTER-POKRAS: So if there is something from this presentation that stimulates any discussion, we can certainly do that. They received very few comments. Cathy Wohman thought somewhere between half a dozen and a dozen comments in all. It was amazing considering the hundreds of comments that went in for the initial decision.
CHAIRMAN IEZZONI: Yes. Clyde, is it possible to get a copy of your slides?
CLYDE TUCKER: It is possible, but probably not right this minute. I mean I could probably quickly organize them.
CHAIRMAN IEZZONI: Well maybe Patrice could help you organize them, because I think it would be helpful for us. We should get an extra copy for David Takeuchi because that would remind us what you said when the material does come out in the Federal Register before the end of the year. Now, Olivia, will you be monitoring this and able to tell our committee when we should be looking at the Federal Register?
OLIVIA CARTER-POKRAS: We can certainly do that. Part of the reason why this didn't get as many comments, this program, one is the very technical nature of--
CHAIRMAN IEZZONI: Yes.
OLIVIA CARTER-POKRAS: --the provisional guidelines that came out. And also because it was more of an informal comment period rather than the formal comment period that they had before. They accompanied those with public hearings, and all sorts of things like that.
CHAIRMAN IEZZONI: Okay. Well we might to just want to say something like you're doing a great job, keep it up, but I think that we should at least have the option. Paul, do you have a comment?
PAUL NEWACHECK: I just have a question. In your cognitive testing, was there any confusion over the concept of Native Hawaiian in the sense that say a Black or a White person who was born in Hawaii might view themselves as a Native Hawaiian like a Native Californian?
CLYDE TUCKER: We think there is going to be some confusion when we go to Hawaii among people who were born in Hawaii, no matter what they are. But we are going to have to evaluate that and see how it is going to work.
PAUL NEWACHECK: It seems like it would be very confusing to a lot of Hawaiian folks.
BARBARA STARFIELD: I mean, aren't there standard--I mean I guess you can't afford to define all these things when you send out a survey, but do you in your own minds have the definition of what Native Hawaiian means?
CLYDE TUCKER: Native Hawaiian would be someone who can trace their ancestry back to--
PAUL NEWACHECK: But that's not--like I'm a Native Californian. It's like Native Hawaiian is what?
CLYDE TUCKER: Well let me tell you what makes this a little bit easier. If you're in Hawaii, people know what a Native Hawaiian is.
PAUL NEWACHECK: I think that is probably true.
CLYDE TUCKER: The complication is, there are very, very few, probably less than 1 percent of people in Hawaii who are 100 percent Native Hawaiian.
PAUL NEWACHECK: Yes. There are lots of Japanese there, too, who have been there many generations but-- well, I don't know. It just seems like--
CLYDE TUCKER: The more complicated problem in some cases is actually whether, if you're from Japan or the Philippines, but the Philippines particularly, or Indonesia, are you a Pacific Islander or are you an Asian?
PAUL NEWACHECK: Right; yes.
CHAIRMAN IEZZONI: Yes.
BARBARA STARFIELD: A Pacific Islander or what?
CHAIRMAN IEZZONI: Asian.
CLYDE TUCKER: Asian.
BARBARA STARFIELD: Asian? All right.
VINCENT MOR: In the OMB Web Site on that Appendix D, does it actually have the results of some of these data?
CLYDE TUCKER: All of the data analyses are there, all the tables.
VINCENT MOR: And that's up on the Web in a pdf file?
CLYDE TUCKER: It is in a pdf file, and it is up on I guess--I didn't bring the log, but it is basically in the OMB section under miscellaneous reports or something like that.
CHAIRMAN IEZZONI: What is the answer to people who are on the Philippines? Are they Pacific Islander, or are they Asian?
CLYDE TUCKER: We consider them Asian.
CHAIRMAN IEZZONI: You consider them Asian.
SUSAN QUEEN: OMB considers them to be Asian.
VINCENT MOR: There must be a parallel that is crossed at some point.
CHAIRMAN IEZZONI: Hmmm. That's interesting. And how did you come down on the Middle East, people from the Middle East?
CLYDE TUCKER: Well this is actually discussed in the standard, so you can get this from the Federal Register Notice on October 30th.
CHAIRMAN IEZZONI: Right.
CLYDE TUCKER: They decided that it was very difficult to deal with the people from the Middle East because how do you classify them? By language? By area? Or by Religion? So basically what the new standard says is that there will be more research conducted on this because we don't know enough to do anything yet.
VINCENT MOR: That was my recollection. I remember we had that discussion--
CHAIRMAN IEZZONI: I remember we talked about it, and I just couldn't remember what the final decision had been, which doesn't surprise me now that I get the numbers since there wasn't actually a final decision.
CLYDE TUCKER: Right.
CHAIRMAN IEZZONI: This was very helpful. Are there questions around the table?
SUSAN QUEEN: I have one.
CHAIRMAN IEZZONI: Yes.
SUSAN QUEEN: The Procedural Design Group, did they put out a report on aggregate of the data?
CLYDE TUCKER: There is a discussion of it in the Appendix B. They have their own separate report that covers both self-identification and aggregate.
SUSAN QUEEN: And that is in the implementing guidelines?
CLYDE TUCKER: Yes, the Draft Provisional Guidelines.
VINCENT MOR: One other question. At one point last Census, or maybe even the one before, probably the last Census, the state agencies and even some of the counties got sort of little software fixes when they were working with their Census data that they'd actually do some work with if they could calculate numbers, and so on and so forth, basically little macros or something like that. Does your office plan to do anything like that? Or instance, alternate versions of calculating what the racial distribution in any particular area is likely to be? Sort of some kind of user-friendly thing to go with the Census CDs?
CLYDE TUCKER: The current plan is probably to let agencies get vendors. I know that we went to a meeting at the Department of Education where they actually have been attending the meetings so they could see the kinds of things we were doing. There is no plan. Actually, there is no one working on this on the technical side, other than they come to the meetings and everything, but there's no one working on it from OMB now. They don't have much of a staff. So they're basically borrowing staff around, and we have yet to agree that we will do consulting or anything like that. We obviously have programs with them because we've been able to do the allocations and stuff, but they are written in SASS and they're written particularly for the data system we've been working with.
VINCENT MOR: So it would be some kind of C++ macro or something to run on something?
CLYDE TUCKER: Well if there's enough demand, the market will require it, right.
VINCENT MOR: Right. In all of its glory.
(Laughter.)
CLYDE TUCKER: The question is--
VINCENT MOR: In correct versions.
CLYDE TUCKER: Well the question is: Can we in fact have enough, exercise enough control so that not only are they correct but that they provide, you know--
VINCENT MOR: The Health Care Financing Administration might be very helpful in that interaction in trying to preempt and put up specs for what the market is supposed to do.
CLYDE TUCKER: Well that's good if you have people who can understand how to write detailed specifications. There are a lot of agencies who don't have people like that. And they can't task that out very well, either.
(Laughter.)
CHAIRMAN IEZZONI: We may want to go off record at this point.
(Laughter.)
CHAIRMAN IEZZONI: Oh, okay.
DEBORAH MAIESE: I had one other question about the extent to which in the test people didn't move beyond the Hispanic Ethnicity question to answer the Race? I mean, to what extent are there problems, and thereby this issue of defining ethnicity versus race? How often did that come up? And how is that being addressed?
CLYDE TUCKER: Well we know that in the dress rehearsal there's a problem. I can't give you the percentage, but certainly I would say it is probably about a third of Hispanics either don't answer the Race question or answer Other and say they're Hispanic, or they say they're Guatemalan, or whatever. Overall, the total number of people we have to deal with who don't answer the Race question or the Ethnicity question is down because before when we had Race question first, so many Whites and Blacks skipped the Ethnicity question. On other surveys, it is unclear because you don't see that Other category. And remember, even on the Census, you know, we are getting two-thirds of people to give answers. And the other things is that--well there are two other things. One is of course there's the instruction that says: Please answer both questions. So of course an answer would be: Other, Hispanic. But the problem is how people answer, Hispanics answer the Ethnicity-- Race question depends upon where they're from. Cubans are quite likely to say they're White. People from South America or Central America, below Mexico, are likely to say that they are either Hispanic or something else, or their country of origin. Mexicans are likely to say Other, and put in Mexican. So it does differ depending on where you're from. And Puerto Ricans are more complicated because--and Cubans, too, because there are a number of Blacks in there, too. And Dominican Republicans, also. The Caribbean is probably quite different from the Mainland south of Central America.
CHAIRMAN IEZZONI: Good. Okay. Thank you, Clyde, that was very helpful. And thank you, Patrice, for helping us get copies of the overheads. Why don't we take a ten-minute break, come back at 3:15. We'll spend an hour finalizing the Medicaid Managed Care Report. We're going to do it by sheer act of will. Then Kathy can take over at 9 4:15. How does that sound?
(General agreement.)
(Recess.)
CHAIRMAN IEZZONI: Why don't we get started so we can wrap this up. Folks? Okay. What we are going to do is we're going to go through the recommendations starting on page 42. This is your last opportunity to give Carolyn comments on the recommendations. So are there any concerns on page 42?
(No response.)
CHAIRMAN IEZZONI: Going once, going twice. Okay. Page 42 is just kind of background. So page 43 is where we get into the actual recommendations. So are there any comments on recommendation number one?
BARBARA STARFIELD: Could I just ask a question? I didn't bring a copy, the July copy. I think I made a few suggestions.
CAROLYN RIMES: I have them.
BARBARA STARFIELD: If I can assume they were put in there, then I don't need it.
CAROLYN RIMES: The July one, or the one you sent me in September?
BARBARA STARFIELD: The latest one, whatever.
CAROLYN RIMES: You sent me some in September, which I haven't done anything on until today.
BARBARA STARFIELD: And I don't know anything about them because I didn't bring them.
CAROLYN RIMES: I do.
BARBARA STARFIELD: So I can't say anything unless I see them.
CHAIRMAN IEZZONI: Okay, Kathy has a comment on recommendation one.
KATHRYN COLTIN: Actually I had a comment that is in the very last paragraph--
CHAIRMAN IEZZONI: Page?
KATHRYN COLTIN: --on page 41, but it repeats in the recommendations in a couple of places--
CHAIRMAN IEZZONI: Okay.
KATHRYN COLTIN: --so I thought I'd--
CHAIRMAN IEZZONI: All right, tell us about it.
KATHRYN COLTIN: It was just the word "satisfaction", where it says "including patient satisfaction."
CHAIRMAN IEZZONI: Yes, um-hmmm.
KATHRYN COLTIN: I would just like to modify that to say "and/or reports of care"--
CHAIRMAN IEZZONI: Of care. Exactly.
KATHRYN COLTIN: Ratings, rather. And/or ratings of care.
CHAIRMAN IEZZONI: Okay.
KATHRYN COLTIN: Because you've covered the experiences, which is the reports. But some of the questionnaires use "ratings" rather than satisfaction scales.
CHAIRMAN IEZZONI: Exactly right. Perfect. That's exactly right.
KATHRYN COLTIN: And then that repeats. I'll point out when it comes up the next time.
CHAIRMAN IEZZONI: Okay. Those are the exact kind of comments we need. Okay, recommendation one on page 43. Vince?
VINCENT MOR: I'm not sure. Page 43, top paragraph, or the paragraph before, I'm not sure. I circled this as I went through, and I'm not sure I understand 100 percent, nor is it understandable to readers who aren't really versed in all this, of what "optional menu of contracting language" might mean. I think I recall some of the background of that, but I didn't find it terribly--it's not sufficiently explicit.
BARBARA STARFIELD: I had the same problem. In fact, I said that, it refers to an optional menu, but later in this it's referred to as "required." Somewhere around here it's referred to as required. So I was very confused by that.
CHAIRMAN IEZZONI: I believe that--
VINCENT MOR: I don't remember all the politics, but this doesn't clarify it.
CHAIRMAN IEZZONI: I believe that this sentence is in here to indicate that the subcommittee may have some interest in the draft contract language. So this sentence is intended to--
CAROLYN RIMES: It's a placeholder, if anything.
CHAIRMAN IEZZONI: It's a placeholder for anything. But we could consider just eliminating that sentence entirely.
CAROLYN RIMES: Eliminate it, or I can footnote what that might mean. There's a couple of options.
BARBARA STARFIELD: Are we talking about the sentence that says "An optional menu"--
CAROLYN RIMES: Yes.
BARBARA STARFIELD: --and then it says it makes-- "requirements"?
CAROLYN RIMES: Yes.
CHAIRMAN IEZZONI: Right. Should we eliminate that sentence?
VINCENT MOR: My recollection was that we actually had draft contractual language that had been proposed, and then everyone said this is too much, deep six it.
CHAIRMAN IEZZONI: Yes.
VINCENT MOR: If we are not going to incorporate it into the report--
CHAIRMAN IEZZONI: We are not going to incorporate it into the report.
VINCENT MOR: Then I'm not--
CHAIRMAN IEZZONI: But we're not getting rid of it.
VINCENT MOR: Getting rid of what?
CHAIRMAN IEZZONI: Entirely.
VINCENT MOR: Getting rid of what?
CHAIRMAN IEZZONI: The draft contract language. It's still, Sarah Rosenbaum and her team are still vetting it with stakeholders. They'll bring it back to us and we'll have to decide what we want to do with it.
PAUL NEWACHECK: So what will happen to it?
VINCENT MOR: Will it be part of this report?
CHAIRMAN IEZZONI: No.
VINCENT MOR: All right, then I would scratch it from this report.
CHAIRMAN IEZZONI: Okay. Carolyn, please--
CAROLYN RIMES: Delete the sentence?
CHAIRMAN IEZZONI: --please delete the sentence that begins "The subcommittee is supportive of existing efforts to develop an optional menu."
PAUL NEWACHECK: So the draft contract language will either be a standalone report--
CHAIRMAN IEZZONI: Yes.
CAROLYN RIMES: It will be.
PAUL NEWACHECK: --from the committee, or?
CHAIRMAN IEZZONI: Or it will just go into the ETHR.
PAUL NEWACHECK: In which case they can still use it, then.
CHAIRMAN IEZZONI: Yes.
PAUL NEWACHECK: You can still get it out into the public domains.
CHAIRMAN IEZZONI: It is something that--
BARBARA STARFIELD: I don't know.
VINCENT MOR: ASCI will probably have a substantial interest in that.
CHAIRMAN IEZZONI: --was paid for with public money.
CAROLYN RIMES: Well there are a number of entities that have an interest in it, so it is not going to disappear.
CHAIRMAN IEZZONI: Yes, it's CDC, and other people.
CAROLYN RIMES: SAMSA, a variety.
PAUL NEWACHECK: Can we sell it to somebody?
CAROLYN RIMES: Well I'm sure that that's been considered, but, yes.
CHAIRMAN IEZZONI: Again, you know I think that we heard from the field that they wanted standardized contract language.
PAUL NEWACHECK: Right.
VINCENT MOR: Yes.
CHAIRMAN IEZZONI: And so when Sarah Rosenbaum brings it back to us, I think we have to think afresh about how we want to position it vis-a- vis us, recognizing that there are other players who are more intimately involved than we were.
BARBARA STARFIELD: I need to go back to page 41, since I got this.
CHAIRMAN IEZZONI: Okay, page 41.
BARBARA STARFIELD: In the second paragraph, it says "Focus Study," but I guess I don't know what the Focus Study is.
CHAIRMAN IEZZONI: Yes. That's Anne's language. She uses that phrase "Focus Study" a couple of places. It's Anne's lack of facility with English, I think. I think she's talking about her survey, for Focus Study.
CAROLYN RIMES: It's the little survey that they did to the states.
BARBARA STARFIELD: Whatever it is, if we don't understand nobody else will.
CHAIRMAN IEZZONI: The phrase "Focus Study" needs to be changed.
CAROLYN RIMES: And I saw your note. I mean that's why -- I agree with you.
CHAIRMAN IEZZONI: I circled it, too. It's Anne's lack of facility with English.
BARBARA STARFIELD: Then in the fourth paragraph, I--
CHAIRMAN IEZZONI: The fourth paragraph on page?
BARBARA STARFIELD: On 41-- 42.
CHAIRMAN IEZZONI: 42.
BARBARA STARFIELD: Whether we need that. It's so motherhood and apple pie, and I think it basically would anger somebody like Bob, unnecessarily, you know. We have legitimate reasons to anger, but this is-- you know, HPA related decisions on... I don't know that we need this paragraph at all. It doesn't really add anything.
VINCENT MOR: My recollection is that some language like this was inserted precisely to mollify, as it were.
CHAIRMAN IEZZONI: Let's leave it in.
PAUL NEWACHECK: It didn't work.
VINCENT MOR: It obviously didn't work, so...
CHAIRMAN IEZZONI: But it doesn't say anything that anybody disagrees with.
BARBARA STARFIELD: Except he might, because he says--
CHAIRMAN IEZZONI: Well he--
CAROLYN RIMES: That's not going to matter.
CHAIRMAN IEZZONI: It's not going to matter.
CAROLYN RIMES: He's going to disagree with--
CHAIRMAN IEZZONI: Bob will write what Bob will write. All right, let's move on to 43. Recommendation one. Any comments on that?
BARBARA STARFIELD: Let's see. Third and fourth paragraph. Yes, that's the third and fourth paragraph. They use the word "smaller" but we don't know smaller than what, at least I didn't.
CHAIRMAN IEZZONI: Where is that, Barbara?
BARBARA STARFIELD: "The Subcommittee supports the adoption of a smaller data set".
ELIZABETH WARD: Smaller than the existing? Is that what we meant?
CHAIRMAN IEZZONI: Yes. I'm not sure what that means.
ELIZABETH WARD: I think it was related to the issue that we kept hearing from-- we heard from the states we went to that they did not understand, and they could find no rationale for all of the data that Medicaid requires, and that they said if we could get very good at standardizing a smaller number of items, we think the data would be more meaningful.
CHAIRMAN IEZZONI: But Barbara is correct. You know, the word "smaller" means "smaller than"--
ELIZABETH WARD: Than what is currently being requested.
VINCENT MOR: Right. That's all.
CHAIRMAN IEZZONI: Smaller data set, or targeted data set.
BARBARA STARFIELD: Either one of those words would be fine. Targeted may be the better word.
CAROLYN RIMES: I'll piddle with something that won't be offensive to me.
VINCENT MOR: Smaller than, maybe.
CHAIRMAN IEZZONI: Okay, Barbara, you had a comment on the next paragraph, as well?
BARBARA STARFIELD: I think not. I think it was the same. The next paragraph also has "smaller data set."
CHAIRMAN IEZZONI: Okay, so we have to do a forward search for the word "smaller" and give some consideration to it. All right, recommendation 1(a), Common Definition of An Encounter. This is something we also heard from the field repeatedly, that they wanted a common definition of "an encounter." Are there any comments on that recommendation?
(No response.)
CHAIRMAN IEZZONI: Recommendation 1(b), Enrollment Data. Kathy?
KATHRYN COLTIN: In the second paragraph, third--well, second and third lines--"Reason for Enrollment," isn't that reason for eligibility, as opposed to enrollment?
VINCENT MOR: Yes.
CHAIRMAN IEZZONI: Yes. Yes. Okay, so instead of "enrollment" use "eligibility."
CAROLYN RIMES: Eligibility?
CHAIRMAN IEZZONI: Yes. Perfect. Okay, other comments on 1(b)?
(No response.)
CHAIRMAN IEZZONI: Recommendation 2.
VINCENT MOR: You don't want to add the new OMB-approved mode of collecting race data?
CHAIRMAN IEZZONI: It says "be consistent with the Office of Management and Budget Standards for Maintaining". I think it is explicit there.
CAROLYN RIMES: We're trying, guys. Come on.
CHAIRMAN IEZZONI: Yes. I think it's there. Recommendation 2?
KATHRYN COLTIN: Yes. Top of page 45--
CAROLYN RIMES: Are we done with 44?
CHAIRMAN IEZZONI: Well, this is--
KATHRYN COLTIN: Well it's part of Recommendation 2. Again, including "satisfaction" with "and/or ratings of access to".
CHAIRMAN IEZZONI: And the Consumer Assessment of Health Plan Survey.
CAROLYN RIMES: No, it is study.
ELIZABETH WARD: It's study.
CHAIRMAN IEZZONI: I thought John Eisenberg said survey this morning.
CAROLYN RIMES: No, no, no, it's study.
BARBARA STARFIELD: That's why it's called a CAPS Survey.
PAUL NEWACHECK: --only pays for it.
CHAIRMAN IEZZONI: Oh. He only pays for it?
CAROLYN RIMES: We verified it four times, truly.
PAUL NEWACHECK: But this is not really very clear because it's not really CAPS that is the standard tool. So I think we should say: For example, those developed by the Consumer Assessment of Health Plan Study, so it's clear.
CHAIRMAN IEZZONI: That's a good point.
CAROLYN RIMES: Say that again, please?
PAUL NEWACHECK: E.g., those developed by.
CHAIRMAN IEZZONI: That's right. Any other comments on Recommendation 2?
(No response.)
CHAIRMAN IEZZONI: Okay, Recommendation 3, Services Covered Under The State Medicaid Plan But Not Under The MCO Contract.
BARBARA STARFIELD: Yes. I think this is a red flag, especially the second paragraph in there because this clearly implies linkages, right? We should share relevant data without any--
CAROLYN RIMES: A red flag to who?
BARBARA STARFIELD: Well I think anyone interested--
KATHRYN COLTIN: To the data sharer, the data matching--
CHAIRMAN IEZZONI: But you agree with the idea.
BARBARA STARFIELD: Well of course I agree with the idea, but maybe we need to say something like "while recognizing the problem of developing unique identifier".
CHAIRMAN IEZZONI: Or this needs to be done within the context of privacy--
VINCENT MOR: But every state has unique identifiers for Medicaid. These people are chattled once they're eligible. That's it. They have no more freedom. We can mandate that they go to an MCO. Forget it.
CHAIRMAN IEZZONI: That's true. For this population, that's true. I think we should only make changes that we feel are conceptually appropriate. We shouldn't make defensive changes unless we feel they are justified. So is there another phrase you'd like to add, a sentence? Anybody?
VINCENT MOR: Encouraging MCOs-- states to share data with MCOs, some of which might be confidential. That is, because they sort of carve out mental health service or otherwise, has a potential positive benefit association with quality improvement, but a potential negative consequence of basically labeling. And whether we want to have a clause to that effect, I don't know. But that's really what it is. I mean, it's one thing to say that the state has all that for quality assurance and so on and so forth in its purview, but rather than say the state gets it for multiple contractors and divvies it out, so that each MCO can then do its own QI or quality assurance processes, that's a different issue. And they are very explicitly saying what we would like the world to do, but if you're telling somebody they're crazy, you know, this one over here is crazy and I don't know they're crazy over on this side, then that might be a--
BARBARA STARFIELD: Well maybe we could just say something, "using the standard Medicaid" you know, whatever it is that is used. It's there. We're not recommending that anybody develop a unique identifier.
CHAIRMAN IEZZONI: No, we're not. No way.
PAUL NEWACHECK: --anything else--
BARBARA STARFIELD: Well, yes, because it could read, to someone who doesn't know this, you could read you need to develop--
PAUL NEWACHECK: Well--
BARBARA STARFIELD: --do the sharing.
VINCENT MOR: Or just drop that sentence. The problem is, the sentence embodies our positive desires.
CHAIRMAN IEZZONI: It's important, yes.
VINCENT MOR: And in fact we heard it fromthe field.
CHAIRMAN IEZZONI: Right.
BARBARA STARFIELD: So could we say that? Using existing--
VINCENT MOR: I don't think it's an issue so much of the technical, the technical issue of ideas. The conceptual problem of--that's the downside of data sharing on this particular context is labeling.
CHAIRMAN IEZZONI: Is there a phrase--
VINCENT MOR: Maybe nothing.
CHAIRMAN IEZZONI: It doesn't sound like there is.
KATHRYN COLTIN: Well I think you could,for instance, say to share relevant data needed by MCOs in order to comply with QSMIC requirements, because that--
VINCENT MOR: That's much better. Yes. That's really explicit.
CHAIRMAN IEZZONI: Okay--
(Many simultaneous voices.)
VINCENT MOR: In lieu of, pursue their quality improvement strategy.
KATHRYN COLTIN: Instead of "their quality improvement strategies,"--
CHAIRMAN IEZZONI: Oh, yes.
KATHRYN COLTIN: To say "in order to comply"--
CHAIRMAN IEZZONI: In order to comply.
KATHRYN COLTIN: --"with QSMIC requirements."
CHAIRMAN IEZZONI: So it's the fourth line from the bottom.
VINCENT MOR: Much better.
CHAIRMAN IEZZONI: Do you see it?
CAROLYN RIMES: Yes.
KATHRYN COLTIN: Because that was my concern, that they tell you you have to do something and then don't give you the information you need to do it.
CHAIRMAN IEZZONI: Have we defined the acronym QSMIC earlier in this report?
SIMULTANEOUS VOICES: No.
CHAIRMAN IEZZONI: Okay, we need to define it.
VINCENT MOR: It's earlier some place.
CAROLYN RIMES: It's way early.
VINCENT MOR: Yes, way early, yes.
CHAIRMAN IEZZONI: We need to define it here.
CAROLYN RIMES: I'll cross it.
CHAIRMAN IEZZONI: Okay. Any more comments on Recommendation 3?
(No response.)
CHAIRMAN IEZZONI: Recommendation 4. In the last sentence on the first paragraph, the "more benefit the collectivity will reap" needs to be changed a little bit.
VINCENT MOR: You could probably drop that last sentence altogether.
PAUL NEWACHECK: It sounds communistic.
(Laughter.)
ELIZABETH WARD: Where is that? I missed it.
CHAIRMAN IEZZONI: It's the last sentence on the bottom.
CAROLYN RIMES: And I'm so used to reading it, I can't see it.I
VINCENT MOR: I think you could readily drop that last sentence.
CAROLYN RIMES: You want me to edit it out, guys?
CHAIRMAN IEZZONI: Yes.
ELIZABETH WARD: That should be quoted somewhere.
(Laughter.)
CHAIRMAN IEZZONI: I know.
VINCENT MOR: "Collectivity." I haven't heard that phrase in a long time.
CHAIRMAN IEZZONI: Yes. And the "reaping" I thought was particularly metaphoric here.
CAROLYN RIMES: Well, we've reached symbolic behavior.
(Laughter.)
CHAIRMAN IEZZONI: That's right. Top of page 45, last paragraph of that recommendation.
(No response.)
CHAIRMAN IEZZONI: Okay, moving on to Recommendation 5, Training To Increase Staff Analytic Capability.
KATHRYN COLTIN: I had a question.
CHAIRMAN IEZZONI: Yes.
KATHRYN COLTIN: What are IPA Programs?
CHAIRMAN IEZZONI: I know.
KATHRYN COLTIN: I've circled that with a question mark.
CHAIRMAN IEZZONI: I did, too.
VINCENT MOR: Interagency Personnel Agreements.
CHAIRMAN IEZZONI: Oh, please. Okay.
KATHRYN COLTIN: I think we need to spell that out.
CHAIRMAN IEZZONI: What is that?
VINCENT MOR: IPA.
CHAIRMAN IEZZONI: It's Interperson--
VINCENT MOR: Interagency Personnel Agreements.
CHAIRMAN IEZZONI: You need to spell out all acronyms, even AHCPR, and even PHS, because readers of this report are not going to be within the Beltway, we hope. So I think we need to spell all those out.
CAROLYN RIMES: Okay.
CHAIRMAN IEZZONI: And what is an IPA-like program? I mean--
CAROLYN RIMES: It's personnel in lieu of cash.
VINCENT MOR: Yes, basically--
CHAIRMAN IEZZONI: Does that mean anything to people who are not bureaucrats?
CAROLYN RIMES: Well--
CHAIRMAN IEZZONI: Sorry.
CAROLYN RIMES: It used--
VINCENT MOR: The CDC does it a lot. HCFA does it a lot. People go from--
CHAIRMAN IEZZONI: But can you explain in English what it is?
VINCENT MOR: Yes. You can go from one agency or one institution to another, spend one or two years there on a personnel-exchange basis.
CHAIRMAN IEZZONI: Okay. Write in parentheses "personnel exchanges for training purposes" or something like that, so people would understand as an EG what an IPA-like--
PAUL NEWACHECK: It will just be written in English, then.
CHAIRMAN IEZZONI: Yes.
PAUL NEWACHECK: What the IPA was.
CAROLYN RIMES: I'll straighten that out.
CHAIRMAN IEZZONI: Some people understand what that means.
CAROLYN RIMES: I kind of liked it, but, you know.
CHAIRMAN IEZZONI: But you knew what it meant, the rest of us were clueless.
(Laughter.)
CHAIRMAN IEZZONI: You're a bureaucrat, Carolyn, just admit it.
(Laughter.)
CHAIRMAN IEZZONI: Any more comments on 5?
VINCENT MOR: You know, as more than just a CPR provides fellowships, so you could have PHS Fellowships or Scholarships, because that's under the whole thing, Public Health Service Fellowships and Scholarships.
CAROLYN RIMES: Done.
CHAIRMAN IEZZONI: But does HCRP even do fellowships?
OLIVIA CARTER-POKRAS: Actually PHS is just certain agencies. It's not every agency in the Department. So I don't know if we want to say HHS Fellowships and Scholarships.
CAROLYN RIMES: Maybe HHS. That's a good point.
CHAIRMAN IEZZONI: All right. Other comments on Item 5, because believe it or not we are on our last recommendation.
VINCENT MOR: The only question I have about 6 is, it has some degree of overlap with 4 to the last clause of 4.
CAROLYN RIMES: This is one we wrestled on a lot.
CHAIRMAN IEZZONI: Yes. It has its own specific context, though, I think.
VINCENT MOR: I understand. So it probably needs to stand alone.
CHAIRMAN IEZZONI: Before we get into the details of it, one, two, three, four, five, sixth line down, I think it is out-of-state laboratories to conduct tests.
CAROLYN RIMES: Conduct test, yes.
CHAIRMAN IEZZONI: Okay. I think that-- was this the recommendation that got Simon going in June?
BARBARA STARFIELD: Well we discussed this. Kathy, you had some questions about this.
KATHRYN COLTIN: This was--there were a number of people--
ELIZABETH WARD: --were addressed.
(Simultaneous voices.)
CHAIRMAN IEZZONI: Okay, so people feel--
CAROLYN RIMES: Is it okay? Or do you have any other recommendations? Because we really wrestled on this one.
CHAIRMAN IEZZONI: Yes.
ELIZABETH WARD: I think that what was good was that when you talked about having MCOs report this kind of--these kinds of results, having found a state that actually has done that.
CAROLYN RIMES: We thought that would help.
ELIZABETH WARD: Is very good. Because 1 then you can see that it's not debating the theory of should it or should it not work. We certainly, Kathy and I all have different opinions, all think--
KATHRYN COLTIN: I mean it was a major discussion.
CAROLYN RIMES: But at least you can say that at least one state is doing it.
CHAIRMAN IEZZONI: Yes.
BARBARA STARFIELD: The penultimate sentence is "We think the MCOs may be in a better position to encourage the recording of these data." It is not obvious to me why that should be the case, and I wonder whether "encourage" is not the right word. Because "encourage" means you sort of like encourage--
CHAIRMAN IEZZONI: Well they don't pay the providers if they don't.
BARBARA STARFIELD: Well I understand that MCOs may be in a better position to do something, but I'm not sure "encourage" is right. Maybe something like "augment" "enhance" "further the reporting" but encourage means they encourage other people.
KATHRYN COLTIN: You also wanted a stronger word.
CHAIRMAN IEZZONI: Why don't we just say "to improve the reporting".
BARBARA STARFIELD: Okay. That's better.
CHAIRMAN IEZZONI: To improve the reporting of these data for public health surveillance and disease tracking.
CAROLYN RIMES: Okay. I've got that.
CHAIRMAN IEZZONI: Okay, so I guess all the heavy lifting work that you did on Recommendation 6 in June really paid off, because it seems like we've got something here that people feel okay about. Did we do it?
CAROLYN RIMES: Well, and you guys know-- does anybody have--can I ask, are there any other parts of things that people are uncomfortable with?
CHAIRMAN IEZZONI: I think, again, this is a report, Carolyn, where you need another staff colleague to read through it.
CAROLYN RIMES: Yes, I can't--
CHAIRMAN IEZZONI: You can't see it. You're too close.
CAROLYN RIMES: --I did a final cold read, and somebody over there did, but we're done.
KATHRYN COLTIN: I think it just had one awkward sentence.
CHAIRMAN IEZZONI: Yes.
KATHRYN COLTIN: The rest of it--
CHAIRMAN IEZZONI: The focus study appears repeatedly.
CAROLYN RIMES: Where is the awkward sentence, Kathy?
KATHRYN COLTIN: It is on page 33 in 8, when you talk about "utilization and encounter data"--
CAROLYN RIMES: Where?
KATHRYN COLTIN: Where it says "In 1997, eight states, or three states" I would say "Eight states collected encounter data and three states collected aggregate utilization data".
CAROLYN RIMES: That's easy.
KATHRYN COLTIN: "While the majority collected both."
CAROLYN RIMES: Perfect. That we can fix.
PAUL NEWACHECK: So just in terms of tomorrow, we're not going to tangle with Bob, we're just going to simply say we did our best job and we feel comfortable with it?
CHAIRMAN IEZZONI: Yes. And I think, though, I think that we should make the statement that we are not proposing additional new data collection. Because in fact--
ELIZABETH WARD: We're recommending the opposite.
CHAIRMAN IEZZONI: --we're recommending the opposite, that there be more focus and more rationality around the data that we're collecting. And I might turn to Kathy to explain how Harvard Pilgrim had to report data in the past for caring for Medicaid patients.
KATHRYN COLTIN: We're not a good example because Massachusetts was one of the few states that did not require encounter reporting.
CHAIRMAN IEZZONI: But you're much better at articulating this than I am. So I will try to articulate it, but if I get tongue-tied I might ask Kathy to rescue me. Because I think that that is an important point for the whole Committee to understand, because Bob's position may become catechism to other people, but it is obviously an erroneous position. So I think once we make that position clear, then we say, Bob, if you feel you'd like to write a dissenting opinion, you should do so.
BARBARA STARFIELD: John needs to say that, not you.
VINCENT MOR: Right. I agree.
CHAIRMAN IEZZONI: I need to talk to John, yes.
VINCENT MOR: I think one of the things I'd be happy to point out, if it's necessary, is at least with regard to California and New York which traditionally have had the most complete reporting and Medicaid reporting systems, some incredibly detailed research linking files on everything imaginable has been done on HIV comparatively between New York and California, only because the data were complete on every single possible encounter.
CHAIRMAN IEZZONI: And was it research that added to the knowledge about quality that improved care for people? Or did it just go to the CVs of the investigators? Because that would be Bob's type of response.
(Simultaneous comments.)
VINCENT MOR: The point was actually that not--
CHAIRMAN IEZZONI: Right. Because -- quality--then that's --
VINCENT MOR: --not only that it was valuable, but that the data systems were in place long before those two states went to Medicaid Managed Care. And they were in place in a very, very detailed manner through essentially a claims process mechanism.
CHAIRMAN IEZZONI: Okay. I think that members of our subcommittee should feel that they should weigh in with Bob as they have points that they'd like to make. Kathy?
KATHRYN COLTIN: I just have one question and then perhaps a comment. It has been so long that I can't recall, but I know that when we've held hearings we've usually put together a set of questions that we ask.
CHAIRMAN IEZZONI: Yes.
CAROLYN RIMES: We did have that.
KATHRYN COLTIN: Did we have anything on that list relating to privacy and confidentiality? Because if we did--
(Simultaneous speaking.)
KATHRYN COLTIN: --and other people chose not to address it--
CAROLYN RIMES: I think we did, but I'll be damned--
CHAIRMAN IEZZONI: Well actually that's a really important point.
CAROLYN RIMES: I'm even trying to think if I would have kept a list.
CHAIRMAN IEZZONI: We should find that out.
CAROLYN RIMES: It's been so long, I know it's off my PC. Because I can remember writing them.
KATHRYN COLTIN: Arizona was the first one?
CHAIRMAN IEZZONI: Arizona was the first one, but--
CAROLYN RIMES: Well, no, but we did--we did a number of them here before Arizona. And I know I had written something out for those because it was important to do that.
CHAIRMAN IEZZONI: Well, if we could find that out--
KATHRYN COLTIN: But I just think that--solicited--and if we didn't then I think we deserve adissenting opinion.
CHAIRMAN IEZZONI: That's a really good point.
CAROLYN RIMES: It's a really good point. I just--
CHAIRMAN IEZZONI: You're absolutely right, that if we ask people-- Carolyn, is there any way?
CAROLYN RIMES: Well I know it's not on my PC. The only thing I can think of to do quickly in the morning is get ahold of Barbara who kept everything and was very organized with what went out, and I will--
CHAIRMAN IEZZONI: Barbara Hetzler?
CAROLYN RIMES: Um-hmmm.
BARBARA STARFIELD: They're probably in our agenda books from way back.
CAROLYN RIMES: Well they would have been in the mailing material that went out, and if I kept a file, absolutely. Well, the Washington meetings and the Boston meetings because all that went out.
CHAIRMAN IEZZONI: Okay. If you can let us know that before 10:00 is when the meeting resumes tomorrow morning, isn't that correct?
KATHRYN COLTIN: Right.
CAROLYN RIMES: She's usually in when I am, very early, so I will call her.
CHAIRMAN IEZZONI: So if you can let us know that. And, Patrice, there is a telephone number in the conference room up in 705 that Carolyn could call us on. Can you give Carolyn the number?
PATRICE UPCHURCH: Sure.
CHAIRMAN IEZZONI: All right.
KATHRYN COLTIN: And I also had a comment.
CHAIRMAN IEZZONI: Yes.
KATHRYN COLTIN: That is, that if it turns out that Bob does elect to write a dissenting opinion, I would like us to have an opportunity to check it for factual errors--
ELIZABETH WARD: Good point.
CHAIRMAN IEZZONI: Yes.
KATHRYN COLTIN: --and then to be able to write a response.
VINCENT MOR: I don't(?) agree.
CHAIRMAN IEZZONI: That's reasonable. I think I need to talk to John.
ELIZABETH WARD: Oh, absolutely. He has to have a very clear strategy for how he is going to deal with this in the morning. Who is going to get called on. Who is going to say what. How is he going to manage it? Or we should just not talk about it?
CHAIRMAN IEZZONI: Okay. Well maybe, Patrice, what you could do is, could you just make sure, I asked Marjorie to tell John--could you just find out whether John is coming back down here? I just want to make sure that I have a chance to talk to John tonight. Just find out whether he's planning to come down here.
CAROLYN RIMES: For the Quality session.
CHAIRMAN IEZZONI: For the Quality session.
PATRICE UPCHURCH: Okay, sure.
CHAIRMAN IEZZONI: Because the Quality session is about to start.
PATRICE UPCHURCH: Okay.
ELIZABETH WARD: But he doesn't know that.
CHAIRMAN IEZZONI: But he doesn't know that, and the basic ploy is that I need to speak to him at some point this evening so he can be prepared. Okay, does anybody have any final words on the Medicaid Managed Care Report?
VINCENT MOR: Now would you communicate, at least from me, I like it. I thought it was very informative. I want to give it to my students. It is very interesting.
CHAIRMAN IEZZONI: Yep. I think that the way it is photocopied you're going to have have to look at, because some of the boxes that are blacked in don't photocopy well.
CAROLYN RIMES: Now I have everything.
CHAIRMAN IEZZONI: Are we allowed to have Sarah Rosenbaum as the author of this, even though she didn't--
CAROLYN RIMES: Well she edited it, and it is under her tutelage. You know, of all the things I would fight about, boy, that wouldn't be one of them.
CHAIRMAN IEZZONI: Okay. I just didn't know, because usually our reports don't go out with other people's names on them.
BARBARA STARFIELD: Well this says "prepared by the subcommittee."
CHAIRMAN IEZZONI: Okay.
CAROLYN RIMES: It's very carefully--wechanged that, too, at one point.
CHAIRMAN IEZZONI: Okay. All right.
PAUL NEWACHECK: One issue, though, is that the way it is implied here that they wrote the recommendations, too, which they may have actually drafted but we gave them the--
CHAIRMAN IEZZONI: They drafted for us, right.
PAUL NEWACHECK: And I kind of wonder about that. Does it seem like to the causal reader that this is a contractor's recommendation?
CHAIRMAN IEZZONI: Yes. I am uncomfortable with this face page.
ELIZABETH WARD: It has to be ours.
CHAIRMAN IEZZONI: Yes, it does have to be ours. I think that there should be an acknowledgement section right up front.
PAUL NEWACHECK: Could it be something like Final Report of--
CHAIRMAN IEZZONI: They were the contractor, yes.
PAUL NEWACHECK: --of the Committee on National Public Health Statistics with the assistance of, or something like that?
CHAIRMAN IEZZONI: Yes.
PAUL NEWACHECK: Or collaboration of, or something.
BARBARA STARFIELD: Or based on background work by.
CHAIRMAN IEZZONI: Right.
CAROLYN RIMES: All right.
CHAIRMAN IEZZONI: Because these are our recommendations, and we were the ones that did the hearings, and so the whole first part is really what we found.
CAROLYN RIMES: Done. That one I can fix.
PAUL NEWACHECK: We are also the Subcommittee on Population.
CAROLYN RIMES: Oh, no, no. I know.
CHAIRMAN IEZZONI: Okay, so we will have to--hopefully this will get approved tomorrow by the Full Committee, and we will have to have a letter that John would have as the transmittal letter to the Data Council or to the Secretary.
CAROLYN RIMES: And I need to do a final thing and then pull the attachments together.
PAUL NEWACHECK: Do you think it might be worth trying to just touch bases with Simon to see if, before tomorrow's meeting, just so there are no surprises?
CHAIRMAN IEZZONI: Well, yes.
BARBARA STARFIELD: It's too late. Anything's a surprise now.
PAUL NEWACHECK: But if he has relatively minor reservations, or they've been addressed, that makes it very different than if he's going to drop the bombshell tomorrow.
CHAIRMAN IEZZONI: Yes. It's just I don't know how long his--do you know what hotel he's staying at? Does anybody have any idea?
CAROLYN RIMES: He usually stays at the Marriott.
PAUL NEWACHECK: Marriott something or other.
CHAIRMAN IEZZONI: Marriott, because--
ELIZABETH WARD: He's usually at the Marriott Metro Center.
CHAIRMAN IEZZONI: Because I can't wait around here after our meeting finishes to wait for him to finish, because I'm just--
PAUL NEWACHECK: So we'll just play it by ear?
CHAIRMAN IEZZONI: Well, no. Maybe Patrice can go and find his phone number.
ELIZABETH WARD: Or you can write a little note to him right now.
PAUL NEWACHECK: That would be nice.
ELIZABETH WARD: Why don't we have her take a note to him and say the committee is done. We need to know whether you're going to have a complaint tomorrow.
CHAIRMAN IEZZONI: Okay. All right. So when Patrice comes back from her mission to John, we'll send her on a mission to Simon. Would you mind writing that note for me, Carolyn? I don't write very well.
CAROLYN RIMES: I'll actually be glad to go find him.
CHAIRMAN IEZZONI: Oh, actually, why don't you because Stan Edinger is the lead staff on Kathy's Work Group, so why don't you go.
CAROLYN RIMES: Unless I'm missing something, I can probably leave for a few minutes.
CHAIRMAN IEZZONI: Okay, Kathy, do you want to sit here, or do you want to sit there-- Stay where you are?
KATHRYN COLTIN: Yes.
CAROLYN RIMES: Could somebody take decent notes tomorrow if there is something I need to hear, because I'm not going to be able to come in tomorrow.
CHAIRMAN IEZZONI: Well, tomorrow, let me just remind people--
PAUL NEWACHECK: We'll transcribe it for you.
CAROLYN RIMES: That's true, too. You will, won't you.
CHAIRMAN IEZZONI: Tomorrow we are going to talk about our effort to assess functional status, which I think is going to be a lot of fun. Carolyn will be assisting with that, but Jerry Hendershot and Paul Plaseck have agreed to be the lead staff on that, which is great because Paul has been working with the Interagency Subcommittee on Disability Statistics for like 15 years or something like that, and Jerry Hendershot obviously was extremely involved with the National Health Interview Disabilities Supplement. So what we are going to be talking about tomorrow for an hour is basically where we want to go with that. Okay, so--
PAUL NEWACHECK: Is there a reason we're doing that from 8:30 to 9:30 instead of like from 9:00 to 10:00?
CHAIRMAN IEZZONI: Because I wanted to have a little time afterwards for people to make phone calls and stuff before the Full Committee starts. It's better than 8:00.
PAUL NEWACHECK: Yes, it's better than 8:00.
CHAIRMAN IEZZONI: We could always start at quarter to 9:00 or something.
BARBARA STARFIELD: The late start is because of Dan?
CHAIRMAN IEZZONI: Well staff needs to come in. Paul Plachek drives in from the Eastern Shore of Maryland. I think he's the one that's going to be staffing this tomorrow. I think he is the only one of the few that could come tomorrow. Okay, with that, Kathy.
KATHRYN COLTIN: Okay, our folders don't have a copy of our charge and work plan, so I will start with just sort of reminding people what it was that we had said that we wanted to do as a work group. There were two main initiatives that we had talked about. One had to do with identifying some of the specific quality related data issues that 17 came out of the Medicaid Managed Care Report. As it turned out, somewhere along the line the baton was dropped. We had talked about having a subcontractor go through the testimony that we had taken and identify all of the topics and discussions that related to quality measures and what some of thelimitations of data were for creating quality measures for Medicaid Managed Care, and using that as a springboard the idea being that the problems that existed for Medicaid probably also existed in other payor data to a large degree, although not all. That didn't happen, and there was not aseparate subcontract let. At this point, with all of the controversy that has gone on around this report, I am hesitant to reopen it. On the other hand, I think we do, as we proceed forward, want to think about what we learned in this project in the context of other initiatives that we take on, specifically.
We had also decided that we were going to try to look at the recommendations that had come out of the President's Advisory Commission on Consumer Protection and Quality In Health Care, and to try to identify areas where some of those recommendations rested particularly heavily on data systems that might or might not exist, or might or might not be up to the task of supporting that particular recommendation. There is among the recommendations one that deals with disadvantaged populations. I think that may be the area where we can revisit some of what we learned through the Medicaid Managed Care Initiative in thinking about the issues around race and ethnicity data that we looked at, issues around SCS that we looked at, in the context of this. So that is sort of my compromise position given that we don't have a subcontractor to go back and reorganize what we learned from the Medicare Managed Care Initiative. How do people feel about that? Is that all right with folks? Does that fit with your sense?
BARBARA STARFIELD: Yes, it does fit with my sense. There is the other thing that I raised this morning. By the way, I thought she answered the questions absolutely magnificently, and that's the whole business about what we know about quality from outcomes, from bad outcomes, but we don't know a lot about what led to them. It was not really all of it error. You know, how can we build in data systems to do this surveillance when we--
KATHRYN COLTIN: We don't know what the predictors are.
BARBARA STARFIELD: --don't know what the predictors are, that's right.
KATHRYN COLTIN: That's true. Well, all right. Then since people seem to be in agreement with that, one of the objectives that I have this morning for bringing John Eisenberg in to talk with us is that both the interagency initiative and the fora are working off of, to a large extent, the Report of the President's Advisory Commission and their recommendations. So what at least the quick committee, or task force has done, it seems is to have identified certain priorities based on that. The ones that John mentioned this morning that looked to me as if they may be promising in terms of examining some of the data issues using them as sort of probes to say well what are the data issues fell into a few areas that he had identified.
One is the initiative on key opportunities for clinical quality improvement. He had identified two clinical areas where work was going on, one being diabetes through the Diabetes Quality Improvement Projects, and a second being depression. It seemed to me those might be interesting probes to look at in terms of data issues and underlying data necessary to support quality measurement and quality improvement in those areas, one being a mental health area and getting us into some of the issues around mental health data. The other, diabetes, includes measures that rest on information that is not typically available in administrative data like laboratory results and vital signs' information for blood pressure control, for instance. So those might be examples of areas that we might want to look at and look at options for how that information might or might not be incorporated into some ongoing data collection processes, as opposed to having to rest on medical record review, or survey information. And then the third that he mentioned under that same initiative was medical errors and systems for being able to identify, track, and prevent medical errors.
BARBARA STARFIELD: Do you mean medical "errors" or "adverse effects"?
KATHRYN COLTIN: I think it is a broad category. I think you could include adverse effects, as well, preventable events, negative events, whatever. There were some other areas that he also listed that fell under "other initiatives" that lend themselves to looking at inadequacies of data as well, and I tried to mark them as he was going through his presentation. One was under the initiative for improving quality measurement where he listed risk adjustment. Lisa obviously is an expert on that, but clearly there are limitations to existing data for doing the kinds of risk adjustment--
CHAIRMAN IEZZONI: Barbara is kind of an expert on this, too.
KATHRYN COLTIN: And Barbara is, too, right. So that may be another area that we would want to explore further and say, you know, what are the key categories of data or information that would be important to collect to be able to do a reasonable job of adjusting for risk, and what do we not currently get. I know we talk about this constantly in our own organization around targeting disease management programs to high-risk patients, and not all diabetics need to have a diabetes disease management program. But the data that you need to identify the subset who do are often not available through administrative data. They may rest upon current laboratory results. They may rest upon, you know, other medical history factors that you may not have access to through claims. So there are a number of potentially interesting avenues that could take us if we wanted to pursue that. I am laying these out as options because I don't think we can do all this, and I would like to have us talk about what it is people would like to do. So I am kind of putting out a menu.
CHAIRMAN IEZZONI: Yes, and Kathy remember that one of our thoughts was to have the functional status initiatives be carefully melded with the Quality Work Group, too, because frankly I think functional status information is going to be more useful for quality assessment than almost anything else.
PAUL NEWACHECK: And for risk adjustment.
CHAIRMAN IEZZONI: And for risk adjustment. And so I want to put that on the table as an option for this, too.
PAUL NEWACHECK: I was thinking, Kathy, as you were talking, I was thinking about our morning discussion and all the different activities that are going on with the FORA, and the QUIC, the IOM effort, what NZQA Factor is doing. It sort of seems to me it is such a field--I mean the field is moving so quickly and there are so many different initiatives going on, that figuring out what our little piece is that is unique and adds value is going to be very complicated.
KATHRYN COLTIN: Well I think the piece that--well, working on functional status is unique, but it is something the subcommittee is doing. So the question is, does the Work Group basically sit back and say we're not going to do anything that's here; we're just going to do what the subcommittee is doing? Or is the Work Group going to have an agenda and some activities that it pursues? And, you know, I am open for that discussion. I mean, I am not wedded to saying, okay, we have to have--but we did approve a charge, we did approve the work time, and to the extent that we can meld the two I think that is fine. But I don't think they need to necessarily overlap 100 percent. So I think this is something we can talk about.
BARBARA STARFIELD: You know, John said something really interesting. I thought he challenged us. He says he looks to the committee to look to the future.
CHAIRMAN IEZZONI: Right, to be on the cutting edge.
BARBARA STARFIELD: We have to be on the cutting edge. That was a sort of novel way of looking at this committee.
CHAIRMAN IEZZONI: Well, and also the people that I--
BARBARA STARFIELD: This is in the form of a question, not a suggestion. To what extent does-- we're going to have ICD-10 at some point for sure, and it's got a lot of other things besides diagnoses now, I think. I haven't studied it. To what extent is this relevant to considerations of data for quality? I mean, do we know anything about that? Is it worth just having a look at that? That's the future.
VINCENT MOR: Let me just offer an alternative future one that is actually in place right now, which is HCFA's approach on this Quality Indicator work in the world of nursing homes, which actually does blend the ICD diagnostic information with a fairly interesting perspective on for whom the 1 information--who is the consumer of the Quality information? Is it internal? That is, for quality improvement, quality assurance, and disease management kinds of stuff? Which is quite a bit different from the NCQA distribution of some form of ranking system, whether it's risk-adjusted or not risk-adjusted. I think that you might want to spend some time taking a look at how those different audiences-- the implications of the use of these kinds of data for different audiences. I am struggling--
KATHRYN COLTIN: Well we're getting way off of what we have agreed to as a charge, though, with that.
VINCENT MOR: Okay.
KATHRYN COLTIN: I mean, I agree with Paul, there are lots of other groups out there thatare working now on issues of data dissemination and how to package information for consumers, and how to get it to them. FACT has an initiative in that area. NCQA has done work in that area. We heard today the IOM has an initiative in that area. I am not looking for us to repeat what others are doing.
VINCENT MOR: Okay.
KATHRYN COLTIN: Also, there are lots of groups out there that are developing performance measures. I'm not looking to us to do that. What I am looking for us to do is to deal with the infrastructure on which those activities rest, which is the existing data systems in this country and their ability to support those kinds of initiatives. So whether those groups out there that are developing quality measures are tearing their hair out because they can't get what they need from the administrative transaction data and are forced to go to medical records to get information that's critical for monitoring quality of care and for improving it because the data are not always the same. When you're talking about improving, you need different kinds of data, more intricate information, detailed information, about the process. So I think that's where I saw our contribution being somewhat unique because none of the other groups are doing that. Okay? There are lots of groups out there that are talking about coding systems, about transaction standards, but not with a view toward evaluating them against a set of needs for performance measurement and quality improvement.
JOHN LUMPKIN: Do you think the Forum will do that?
KATHRYN COLTIN: They may, but we'd be guessing about that. And I think if we laid that out as an area where we have a particular expertise and can make a contribution, I think that is something that we could either collaborate on or could share recommendations with them. But it is not as if there is an initiative already underway and we're trying to tread in the steps where they've already gone. I mean, I really feel like we don't need to spend our time doing that, we've got so many other priorities to address. So if we're going to make a contribution, I think it would be helpful for it to be in an unique area, and in an area where we have a particular unique expertise. And data systems and data policy are our area. So that's what I thought we had agreed to in terms of our charge. However, that is still a very broad expanse. And I think within that we need to think about a focus for the next year's work. And we had talked about using the recommendations from the President's Advisory Committee in the areas where we knew they were headed in terms of developing measurement systems, or looking at improvement systems and saying, well, what do we know about the data that exist to support those kinds of initiatives? And what would need to be done to improve those data to be able to help?
VINCENT MOR: So that's the data infrastructure needed to support the sets of recommendations regarding quality measurement, and for instance the President's Commission Report. That's still a very, very broad--
KATHRYN COLTIN: It's still very broad.
VINCENT MOR: Because the nature of the data infrastructure is quite different as a function of the sector of the health care system.
KATHRYN COLTIN: Oh, absolutely. Absolutely.
PAUL NEWACHECK: And depending on the quality measure.
CHAIRMAN IEZZONI: And, Kathy, NCQA wrote a really nice report that came out two years ago that actually is going to be excerpted and published in JAMA pretty soon about the information needs to support quality measurement.
BARBARA STARFIELD: That was Roadmap, wasn't it?
CHAIRMAN IEZZONI: Yes. Yes, it was actually quite good. So how would you see what we're doing--
KATHRYN COLTIN: Well what they laid out was at a very high level, and it didn't talk about specific steps that the health care system would need to take to be able to get there. It talked about health plans because it was done within the context of health plans in saying, well, health plans should try to provide incentives for doctors to invest in electronic medical records. It was that kind of a recommendation. But it didn't talk about things like, you know, looking at the existing coding systems and saying whether or not those systems could capture information that would be necessary to support quality improvement. I mean, Simon and I, for instance, sit on the AMA CPT-5 Work Group on Managed Care, and we put forth a set of recommendations jointly with the Research Committee that CPT establish a set of tracking codes that were not tied to reimbursement but would enable plans to collect information about specific processes that occurred within the context of delivery of a service that was billable. So there might be a broader CPT Code, but then there might be a tracking code that was more specific around, say, the type of counseling that was 1 provided to a patient. Or one of the things we even looked at were tracking codes that related to test results, that might say this person's LDL cholesterol was between 100 and 130. There might be a code. And that code could be used not just to track outcomes but to identify high-risk patients. So that is an approach that is being looked at in that group already. But it's those kinds of potential opportunities for improving the data infrastructure to do this kind of measurement and improvement work that I had in mind. Vince?
VINCENT MOR: A couple of thoughts. I liked the concept of thinking of the Work Group being in sync with the Full Committee on Functional Status, but prior to that there is also something that you may want to think about. That is, the NHII process and trying to pull that together. I think there may be some significant contribution if we can think about how measurement of quality fits within the conception of what it is we want to propose as a national health information infrastructure. Just the example that you gave of the process codes are the kinds of things that we would want to suggest would be enabling technologies to allow us to actually have a national health information infrastructure.
ELIZABETH WARD: I think that's it. I was hunting for the language, and I think that's where I find, having spent 10 more years in meetings with others looking at how are we going to improve quality, all the recommendations that are made never talk about enabling the data systems to make it happen. They all stop at the point they make recommendations about what we need to do. And nobody is talking about enabling somebody's data system to look different--
KATHRYN COLTIN: And that's really what I'm suggesting.
ELIZABETH WARD: --because if we keep trying to go there with what we've got over here, we can't get there.
KATHRYN COLTIN: And that to me is where the difference lies between what NCQA did and the kind of specific action that I was just giving as an example. I mean, they sort of say here's where youneed to go. It's like saying we want to go to Los Angeles. There it is. That's where we want to go. And what we're talking about is how could we get there.
ELIZABETH WARD: Are you going to fly or take a--
VINCENT MOR: But I think, in reference to NHII, the first thing that you could do is to describe what it is that's missing and what needs to be there, and that can be then incorporated into the NHII document, and then subsequently to begin to get more into the details. A lot of measurement development that I've 19 seen go on is focusing, because you can't get access to the clinical data, then you're going to use more surveys. We probably ought to comment about that as part of the process. So I think there's a fair bit of stuff, some stuff that could be done and ought to be done fairly quickly since we're looking at trying to have maybe a first draft of the new NHII document by March, by our March meeting, that there may be some good input from the Work Group and more detailed work after.
KATHRYN COLTIN: To me, that--
BARBARA STARFIELD: And the other--
KATHRYN COLTIN: Go ahead, Barbara.
BARBARA STARFIELD: The other overlap is with the Vision work, the Vision papers from the other countries. They don't address quality specifically, but the way they are approaching data collection has certain implications for information for quality, so maybe we ought to get those papers spread around. Kathy is going to be at the Workshop for at least part of those presentations, but I don't know whether the rest of the people are. Maybe you could be the filter, Kathy.
ELIZABETH WARD: And it might be the melding would start with functional status--
CHAIRMAN IEZZONI: Well, in terms--
ELIZABETH WARD: --in terms of what your--the larger committee wants to do about how to recommend that that be done better. I would hope that we won't just do another paper on 'this is what we need,' but we have no idea how you are actually going to get there. And then the Work Group could sort of try out that on that particular goal.
KATHRYN COLTIN: Okay, any other thoughts on John's suggestion?
CHAIRMAN IEZZONI: A practical thought. There's a lot of overlap. I think John is the only person who doesn't overlap with the full Subcommittee on Populations and the work group. Kathy, our energy level, I mean you guys are great to come to all these meetings. We've got these four meetings scheduled to do functional status stuff through next July, but it has been awfully hard for Patrice and our contractor to hear from you all, when you're available, you know, what dates you can make, trying to figure out dates for us to get together and meet has Patrice and I literally, I mean it's amazing we're not bald, both of us--
(Laughter.)
CHAIRMAN IEZZONI: --right now, because we've been pulling our hair out the entire summer and trying to just find dates. So I guess we need to have a discussion about the practical realities of having two conceptually full and interesting parallel efforts that have some overlap but not complete overlap, that are proceeding simultaneously, given that we are talking about the same people except for John, and how this is actually going to work in practical terms. Because, I'll tell you, I feel totally deflated. It has been very, very, very difficult to even find these dates, and I am hoping that tomorrow--I already know Paul can't make the October dates, or maybe he can, but I'm hoping that tomorrow you can bless our October, our January, our April, and our July dates. And if they can't, Patrice, have we figured out a punishment for them?
(Laughter.)
PAUL NEWACHECK: Withhold their federal stipend
(Laughter.)
CHAIRMAN IEZZONI: I don't think that's a punishment.
KATHRYN COLTIN: You get an assignment.
CAROLYN RIMES: You get to be your own work group.
(Laughter.)
CHAIRMAN IEZZONI: I just don't know how this is all going to work itself out.
PAUL NEWACHECK: I agree with you, though, Lisa. I'm not talking just about the administrative part, but there are so many meetings. It's a lot, particularly for people I guess on the West Coast. I mean, it's an extra day every time there's a meeting here.
CHAIRMAN IEZZONI: Right. And I would love to come to that meeting the evening of November 3rd, but I just can't. I've got three trips to D.C. between last Wednesday and this Friday. So even though I'm not from the West Coast, it's just like I am a ping pong ball. So I am just wondering how this is all going to happen. Not that I don't want it to happen.
BARBARA STARFIELD: What is the specific problem? Is it that most of us can't make those meetings, or can't make the third meeting? I mean, what is the specific problem? Or is it that there are pieces of us?
CHAIRMAN IEZZONI: We haven't even tried to schedule the Work Group stuff yet, except for November 3rd. And so we have had a terrible time trying to schedule the full committee stuff with these two-day meetings that track out to July of Year 2000. So now if you want to append on Work Group stuff, given that we all overlap except for John--
BARBARA STARFIELD: But, you know, John, with the NHII, we've done work for the last four months all by e-mail. We haven't had one meeting, and a lot has been done.
CHAIRMAN IEZZONI: Well the functional status stuff cannot happen by e-mail.
BARBARA STARFIELD: Well, you know when you stop to think about it, conceptualizing what an NHII ought to look like and getting to the point where you not only know that there are three things you want to do, but what the contents of those are, that's a major problem.
PAUL NEWACHECK: I think the initial problem is further that these are two major activities that are starting up at the same time, so they require--if one were sort of phasing down and the other were starting up, that would be fine, we could balance them, but they are two major initiatives. And it could consume a huge amount of resources.
JOHN LUMPKIN: Right. And that's really why I'm suggesting that may be the initial focus be on trying to merge in with the NHII piece, which I don't see as trying to produce a 25-page document. I really see it as a couple of pages of conceptual thought that can go in, and that may be able to be done through e-mail, you know, through a list serve where people can actually discuss and try to pull those pieces together. But one of the things that I've come to understand about the functioning, and even though there's an overlap, and maybe this is the missing piece, is that what we've seen on the standards and the Work Group on CPR is that we actually have a totally different staff with the Department for those two groups, and having a departmental staff that are separate who are committed to that particular issue allows more work to be done than if it were all merged together. So that is just something to consider.
CHAIRMAN IEZZONI: Well we--I was talking before you came in that Paul Newacheck and Jerry Hendershot from NCHS are staffing the Functional Status effort. And by the way, Functional Status is going to be defined broadly to include things like depression and mental health reporting as well as physical functioning and sensory function. John, if at this point we did not pursue the Functional Status work, I frankly would probably not want to participate anymore because I'm so invested in it. And so, you know, by stopping at this point and doing NHII stuff, you know, given that the ball has actually been started, it might not be perceptible to you but, you know, I came down and did a teleconference with Interagency Subcommittee on Disability Statistics in September, the first week in September. I did that in July. So things have been happening without the subcommittee being aware of it because we haven't been able to schedule a meeting.
JOHN LUMPKIN: I don't think I was suggesting that. What I was suggesting is that the face-to-face time be spent on Functional Status.
CHAIRMAN IEZZONI: Oh.
JOHN LUMPKIN: And that the quality piece for the NHII be done by e-mail. And that would keep from having to try to get additional meeting time to do the quality stuff.
BARBARA STARFIELD: So what it took was somebody taking responsibility for organizing it and putting it out there and they'll react to it.
PAUL NEWACHECK: A draft.
BARBARA STARFIELD: Yes. I mean you could do that already because we've already done that, basically.
CHAIRMAN IEZZONI: Well I'll tell you, I won't name the people who I have e-mailed about the Function Status stuff, but there are people around the table I haven't heard back from. You know, I've asked for input and I haven't heard back from people. But the people I have wanted input from haven't given it to me. So I think that if we do agree to go in one direction with an e-mail approach, that there needs to be a consensus that this is our modality for communication and that people will respond to it.
PAUL NEWACHECK: I think also John is suggesting that Functional Status not be the e-mail version; right?
JOHN LUMPKIN: Right.
CHAIRMAN IEZZONI: Right. That's what I have now understood. I thought that he was suggesting that we start with the NHII and not do the Functional Status.
ELIZABETH WARD: Well I think as a subcommittee we have to decide one more time whether we have the time to have a work group in addition to the subcommittee.
CHAIRMAN IEZZONI: Right. Exactly.
CAROLYN RIMES: Two work groups.
ELIZABETH WARD: What I've heard is that--
CAROLYN RIMES: Two work groups.
CHAIRMAN IEZZONI: Two work groups, because Stan Freedman is on both--
CAROLYN RIMES: They are both work groups, Quality and Division.
ELIZABETH WARD: What I've heard from the subcommittee that's sitting here is that they are not going to show up for an additional work group regardless of the descriptions of the work that we could get done. So I think if we really don't have the time and that the only time is going to be, then I think, Kathy, you have to decide whether you would be willing to work by e-mail.
KATHRYN COLTIN: Well I have no objection to e-mail, actually. I kind of like that.
ELIZABETH WARD: So it would be your responsibility to manage that work group, if everyone around the table says I'm not coming to another meeting--
KATHRYN COLTIN: Right.
ELIZABETH WARD: --and I'm not going to add another three-hour meeting to when I do come here, then we have to decide whether you are willing to manage an e-mail system for doing that.
KATHRYN COLTIN: Well I'm comfortable with managing an e-mail system, but I am concerned about the point Lisa made about sending e-mail out there and never hearing anything back. I mean if we are going to decide to go that route, then it has to be with some level of commitment that people are going to participate and are going to get back to me. And maybe one of the first things we do is agree on some reasonable turnaround times; that I don't ask you to get back to me in three days on anything; that we agree what's a reasonable turnaround time, and that people commit to meeting those turnaround times.
CHAIRMAN IEZZONI: That's fair.
KATHRYN COLTIN: And if that can be set up and people can agree to that as a ground rule for how to work, then I would be very happy to proceed along that kind of a track. I think that is perfectly fine.
BARBARA STARFIELD: Well it's worked very well in the NHII, but it takes good staff, also.
KATHRYN COLTIN: But I would like to continue to also do what the NHII Work Group has done, and that is carve out a small piece of time at full committee meetings to get together and go over sort of where we are.
CHAIRMAN IEZZONI: How many people are on the NHII thing?
JOHN LUMPKIN: Gee, I don't know.
CHAIRMAN IEZZONI: You don't know?
JOHN LUMPKIN: I would say that there are probably four or five committee members and--
CHAIRMAN IEZZONI: And who's been your staff on that?
JOHN LUMPKIN: Mary Jo Deering has been the lead staff person on that, and she's pulled in-- we probably have now about, I would say there's about five or six department staff who participate with some regularity with the--
CHAIRMAN IEZZONI: See, that's the difference. That's the difference.
JOHN LUMPKIN: Right.
KATHRYN COLTIN: Yes.
JOHN LUMPKIN: And I think that if we know that there's staff on the--I forget the name of that thing that John talked about, the--
CHAIRMAN IEZZONI: The QUIC?
JOHN LUMPKIN: The QUIC, to what extent can we pull some of that staff in to work with us.
CHAIRMAN IEZZONI: That's a good suggestion.
JOHN LUMPKIN: And really it's just a matter of if we can find someone who will play that kind of lead role. And the other piece is is that what Mary Jo set up was a List Serve, which we're just now starting to use, but I think that makes it easier. Because sometimes when you want to initiate something it's hard to pull all the names together, but if you've got a List Serve you can just send it to the List Serve and you can start a whole process.
STANLEY EDINGER: Well some of the staff on the QUIC are on this, because the INFOMATICS one, Rob Kolidner is on the staff, and the other person is from DoD but isn't on it; the other one is Gregg Meyer and Steve Klauser, which is some of the Risk Adjustment Measures, but they're not on either one of these, although how much time Gregg and Steve will have to devote is hard to say, too, so it's difficult. But in some cases, like I said in Rob's case, he's on both.
JOHN LUMPKIN: And I think the e-mail piece of that is that some of them you may want to have just to respond to the e-mail.
CHAIRMAN IEZZONI: So it's a staffing issue. Where did Marjorie go?
CAROLYN RIMES: She's right there.
KATHRYN COLTIN: There she is. You're lost. We're talking about--
MARJORIE GREENBERG: Yes. I really came in in the middle of that.
KATHRYN COLTIN: We're talking about the fact that scheduling time for, face time for meetings of the Quality Work Group is probably unrealistic in the sense that launching a major initiative on Functional Status through the Subcommittee on Populations is going to consume a fair amount of face-to-face meeting time over the next year, and it's all the same people here.
MARJORIE GREENBERG: Right.
KATHRYN COLTIN: And so while we may be able to preserve a couple hours of meeting time at every full committee meeting, in between for the actual work to get done it's probably very unlikely that we could muster enough support for committee members to--
MARJORIE GREENBERG: For separate meetings.
KATHRYN COLTIN: --to attend separate meetings. So we were looking at using the kind of approach that the NHII Group has done, which is to use e-mail. But that has required a significant amount of staff support to keep going.
MARJORIE GREENBERG: Yes.
KATHRYN COLTIN: And we as a Work Group have stamina, but--
CAROLYN RIMES: Well, and there are a couple of people. I mean there really are a couple of people that have been assigned to this Work Group, too. Well, I mean you might also want to revisit some of the resources that are already allocated, too, because they are there.
KATHRYN COLTIN: Well, they're there, but a good many of them aren't even here now.
STANLEY EDINGER: Well, but it may be that there needs to be a coalescence.
CAROLYN RIMES: That we can work on.
VINCENT MOR: I think that the NHII is a frame. I think to really start a productive e-mail exchange, a frame is probably necessary.
BARBARA STARFIELD: That came out of e- mail, though--well, largely. We had a statement paper, but it wasn't...
VINCENT MOR: Yes, but the paper provided a frame I think more than what is currently, what work currently exists because my interpretation actually ended up--I was operating from a different interpretation of our brief charge than you clearly had thought about. So some more detail on that will help focus people's thinking.
KATHRYN COLTIN: I can try to do that. If we're going to try to weave this into the framework that's already been developed by the NHII Work Group, I think, only to get familiar with that and see where it overlaps. Clearly I will want to weave it in with Functional Status. To me I see that as sort of coming from, if I took the road map that NCQA had developed, they have these charts with concentric circles and they show where the gaps are, you know, where it is that we don't have the data that we need, and they're big categories of information. I think one of the first things we would want to do is take a look at those gaps, take a look at what you've developed, see where the overlaps are and aren't, and maybe there should be more overlap and there may be opportunity to input into what you've developed, if there are key areas or kinds of data that they have identified as being important for quality measurement and they don't show up somewhere in what you've come up with at this point, the types of records you've developed we can certainly bring that to your attention. But there are going to be more gap areas than we can address in the short term. If we were going on for two years, maybe we could get through them, but at least for the first year we will still have to prioritize. Functional Status will be clearly on that list of gap areas, and we can agree right now that that will be the number one priority and that we will address that through the subcommittee. But there are other key gaps I think that we could proceed to address through some e-mail discussions, and some small amount of face time at regular meetings.
MARJORIE GREENBERG: I think, too, if you identify specific areas where you need staff support beyond the staff that are currently named to the Work Group, then maybe we can have two to get that ongoing kind of dialogue with the Data Council as to need for staff. I mean I think we would have to kind of identify specifically where that was going.
KATHRYN COLTIN: Well it may be that if we prioritize among those gap areas and pick just a couple of them, that the ones that we choose might in fact overlap with the work of one or more of the other committees, like Simon's group, or even the Full Committee on Standards. I mean, what is the role, for instance, of the Claim Attachment Transaction Standard in being able to capture additional types of data? That may be something we talk about with John's committee. So, you know--
MARJORIE GREENBERG: Simon's pretty into that--
KATHRYN COLTIN: That's Simon's too, right. The other one is Jeff's. Okay, Simon's and Jeff's. So, you know, it really depends on what we select. It may be that there are other synergies that we can develop and not have to pursue everything on our own, but rather work through some of the other avenues.
MARJORIE GREENBERG: And you're planning to have the meeting on November 3rd?
KATHRYN COLTIN: Yes. And I think that will help with prioritizing.
MARJORIE GREENBERG: Yes, that's what I was going to say.
KATHRYN COLTIN: That was my main purpose for that meeting, is to have people who have been working on developing quality measures tell us what the major limitations are that they're running up against, what are the key frustrations? What's preventing them from developing the kinds of measures that they think we need? Where are they running into the data limitations? What's forcing them to surveys?
MARJORIE GREENBERG: That's what I was thinking. And then maybe some e-mail exchange after that meeting.
KATHRYN COLTIN: And I actually talked to Simon to see whether there were people from that group that might want--
MARJORIE GREENBERG: Yes, he and Jeff I think will both want to participate.
KATHRYN COLTIN: And he thought maybe Clem might. I don't know if he's planning to come on the 3rd, but he thought that might be an area that would interest him, as well.
VINCENT MOR: Kathy, in addition to data limitations, there are sort of measurement and conceptual limitations that need to be grappled with beyond even Risk Adjustment. Risk Adjustment is very important, but there are big, big conceptual issues. I will see if I can send you a draft of sections of something that I put together for Contract which is sort of dealing with conceptual issues about quality measurer kinds of things that are focused on Long Term Care, but I think will have applications beyond that.
CHAIRMAN IEZZONI: Could you send that to me, too, Vince? Because it may be very relevant--
VINCENT MOR: I first have to get permission.
CHAIRMAN IEZZONI: --on Functional Status work.
VINCENT MOR: Yes, clearly very relevant.
CHAIRMAN IEZZONI: Very relevant, yes.
KATHRYN COLTIN: So it sounds like that's the route we want to take. I think what we should do is, after that November 3rd evening meeting, in between now and then I will take a look at the frameworks of the NCQA one and what the NHII Work Group has come up with and see if I can come up with some sort of a draft framework that we can plug into for at least examining priorities. And then after we have participated in the November 3rd evening and heard, you know, from the community of people who are struggling with performance measures about what they think are some of the key priority areas, we can start an e-mail exchange to discuss that information and pick, hopefully, you know, maybe two or three areas that we want to try to focus on and some strategy for how we would do that in kind of a low effort/hopefully high yield approach. One being a high-effort approach, which is the Functional Status, but I think a couple of the other areas may not need to be such high- effort areas.
CHAIRMAN IEZZONI: Well, and I don't have any illusion that our Functional Status effort is going to come up with specific recommendations at the end of it. Because I actually think it's too big an area, and that what we will come up with is recommendations for a process that should be put into place. So if you want to do something specific, Kathy, you should probably bite off a narrower piece of this. We can talk about it tomorrow morning.
STANLEY EDINGER: John mentioned the QI taxonomy. We're working on it with HCFA. Actually Eric Schneider is the one who is writing a good part of it. You know Eric really well.
KATHRYN COLTIN: Yes, I know Eric.
CHAIRMAN IEZZONI: Well Eric wrote the NCQA Report on the Information Infrastructure thing.
STANLEY EDINGER: That's being played with right now, but some parts of that might help because it's a little vague at the moment, to be honest about it.
CHAIRMAN IEZZONI: Eric is also the lead author on the piece that's going to be appearing in JAMA, if it hasn't already appeared, from the NCQA Report. So he's been thinking about it lately.
KATHRYN COLTIN: Okay. Well, I think we have a plan, or at least a pre-plan.
CHAIRMAN IEZZONI: Well I think it's a great idea for the NHII to think about quality, don't you?
CAROLYN RIMES: I think it's necessary.
CHAIRMAN IEZZONI: Yes. I mean, I'm surprised that we didn't come up with that before.
JOHN LUMPKIN: Well, and where it fits is that part of what we're trying to--we're trying not to reinvent the wheel, and so there are pieces of the technology at NHII which relates to the National Data Model, which we're not going to necessarily reinvent because I think other countries have done that piece, but it may be that the two pieces in our contribution are what we're calling Information Organizers, or Views of the data, and the other piece being the quality piece that we can sort of add to what's been done by other countries.
CHAIRMAN IEZZONI: That's a great idea.
MARJORIE GREENBERG: What about the CPR who has its focus on quality? Is that primarily the quality of data, or the data in the CPR?
JOHN LUMPKIN: Well I think conceptually what we want from the computerized--the patient's medical record and electronic transmission of same provides the data. And here's the piece that's missing of, you know, where do you put those little functional tags or, I've forgotten the term that's used, but I like that term--
KATHRYN COLTIN: Status?
JOHN LUMPKIN: No, CPT codes--
KATHRYN COLTIN: Oh, the tracking codes.
JOHN LUMPKIN: The tracking codes, which are the higher level than the natural medical language that can drive decisional support. That's where the quality piece I think comes in, as well as the ability of the electronic computerized medical record to provide the raw data that's then used for measuring quality.
KATHRYN COLTIN: My sense of the tracking codes is that they may be an interim solution that works until you have electronic medical record data available. Because there isn't anything right now that those tracking codes represent that one couldn't get from an electronic medical record if you have it. So my sense is they may have, you know, a shelf life of ten years or so, but in the meantime they meet a critical shelf life.
CAROLYN RIMES: Ten years is a long shelf life these days.
(Laughter.)
CHAIRMAN IEZZONI: Well, this was productive.
KATHRYN COLTIN: So I don't have anything else.
CHAIRMAN IEZZONI: You don't have anything else?
KATHRYN COLTIN: That was really it.
CHAIRMAN IEZZONI: Because I think I'm supposed to end with something, but I'm not sure that I have anything to end with, like "overview and assignments"?
CAROLYN RIMES: Ending is good.
CHAIRMAN IEZZONI: Ending is good, yes. Ending--except, John, I need to speak with you.
JOHN LUMPKIN: Yes.
CHAIRMAN IEZZONI: And tomorrow morning, 8:30, but we will be a little flexible, 8:35, 8:40, for some of you from the West Coast, and those of us who got up at 4:00 this morning on Eastern Time, like me.
PAUL NEWACHECK: That makes us even.
CHAIRMAN IEZZONI: What?
PAUL NEWACHECK: Makes us even.
CHAIRMAN IEZZONI: Yes, it does. It makes us completely even. So we'll see you tomorrow morning and we'll talk about our next initiative, which actually I think will be interesting from a conceptual point of view. VINCENT MOR: You're going to talk about the Functional stuff?
CHAIRMAN IEZZONI: The Functional stuff, yes. And it's going to be "FUNCTIONAL", writ large. So people's ability to function in their daily lives, including the ability to see, hear, move their bodies, think, react emotionally, et cetera. So it should be fun, including kids.
(Whereupon, at 4:45 p.m., the meeting was recessed, to reconvene at 8:30 p.m., Tuesday, September 28, 1999.)