Hubert H. Humphrey Building
Room 405A
200 Independence Avenue, SW
Washington, DC 20201
Introduction and Welcome - Dr. Iezzoni
Discussion and Update on Progress for Insular Territories
Committee Discussion of SES Collection and Analysis Implications - Dr. Starfield
Draft Final Report on Medicaid Managed Care Data Collection and Reporting - Ms. Rosenbaum
Status of Medicaid Managed Care Project Discussion
DR. IEZZONI: Why don't we get started then? This afternoon we have a lot of work to do among ourselves basically as a subcommittee, and folks who help us out, staff to the subcommittee.
Why don't we, because there are some new people in the room, just go around the room and introduce ourselves, and say where we are from, and then we can get started on the agenda.
[Introductions were made.]
We have a lot of discussions to get into here. We're going to follow the agenda in only one particular, and that is that Sara Rosenbaum. When she shows up, we have to respect her time, and immediately go to her issue. But other than that, I don't think people should view the times that are listed here on the agenda as set in stone, although hopefully we will try to end by four o'clock, if not earlier, depending on how things go.
MS. GREENBERG: I was just going to ask if you could start with the discussion on the insular territories. I would like to be here for the discussion of SES. If I could be ready to go to it, and I'm not back, just proceed. But would you mind?
DR. IEZZONI: No, absolutely not. What Marjorie was talking about, for those of you around the table who don't know, last year basically we did a two day meeting in Washington where we had representations of insular areas from the Pacific region and U.S. Virgin Islands and Puerto Rico come and speak to us about their health care data needs, barriers to meeting those needs, and recommendations for addressing their concerns.
Dale Hitchcock and Joan Turick have produced a draft primary report for us that will be kind of the beginning working document from which the subcommittee will hopefully make recommendations. So what we're going to do now is just hear from folks around the table, their comments basically for Dale and for Carolyn as lead staff about the subcommittee report. Then after we get comments about the subcommittee report, I think we should talk about our recommendations.
MR. HITCHCOCK: Can I say a few things first? I really want to thank you and Barbara for the comments that we received from the others. Paul said he has some that he's going to give to me today, and that's very much appreciated. I think the comments are right on. Barbara, from the point of view visited the project all that much, there were some real questions that you raised, that I have in my own mind as well, like why do we lump these strange groups together in the first place?
Obviously, we need to address that. Puerto Rico and the Virgin Islands are much different than the Pacific insular islands. It's interesting, in doing some research on this project I even found two more insular areas that we didn't cover. There was a Holland Island and a Compact of Free Association. But they are not listed in the department's report. They do not really have the same stature apparently.
Lisa, in your comments I particularly liked the idea of bringing it to light with some comments and quotations that were made during the meeting. I actually was planning on doing that, but sort of abandoned it in an effort to get something in sort of a structure out to people to comment on.
We're really interested in any help you can give us in bringing some life into this, and having it more accurately reflect what the committee wants to send forward.
DR. IEZZONI: Great. Barbara and I did send comments to Dale. I don't think we need to reproduce these for the whole committee, unless people feel that they want them. It would just be good to hear your comments.
Elizabeth, you had a special interest.
MS. WARD: I had a special concern about what we were doing to these people to bring them all this distance, and whether it was going to be worth it. I think Barbara hit on my concern, and that is, there is nothing special about their concerns. It's not that much different than anyone else in the country is having. It's just when you are living out in the middle of nowhere, it's a hundred times worse.
I think that's the real challenge in this report, is to say if you are having a problem finding programmers, if you are having a problem finding nosologists because you are in Missouri, if you are trying to do it and you're in Guam, it's just a hundred times more difficult to find a programmer and a nosologist.
How detailed do we want to get about possible solutions is a real challenge, because each one of them has some really significantly peculiar problems related to the fact that they are isolated and poor. I think what struck me about the hearing most of all is that there used to be an office that really helped these people, and through the multiple budget cuts in government, that one little office that was worth $1 million, has been eliminated. You could begin to see that with the elimination of that office -- I think it was in the Interior wasn't it?
MR. HITCHCOCK: Yes.
MS. WARD: That that was sort of the beginning of the end of their ability to try to interact with each other, and have an office that try to spread the way to help them deal with their deficits.
MR. HITCHCOCK: That took that office and sort of pulled it into the Interior's ASPE program office as I recall.
MS. WARD: Right. It was sort of like I left the meeting saying, well, if you put that one little office back, you might have some hope.
DR. IEZZONI: So is that a recommendation?
MS. WARD: That was one of the things that struck me about sitting through the hearings; that that was truly, as we heard them give a history of what had happened, and some of the gains they had tried to make, had really been pulled out from under them with the disappearance of that office.
MR. HITCHCOCK: In reading the background too, it's obvious that a lot of these Compacts were really formed because we wanted military bases during the Cold War. Now the Cold War has warmed up, and there is less interest in their problems.
DR. STARFIELD: Maybe that's why we have to put the Sovereign Nations in with Puerto Rico.
DR. MOR: Yes, it's kind of a hodge-podge, isn't it? Mainly, it's a reflection of historical idiosyncracies.
MS. RIMES: There's a quote.
DR. IEZZONI: I think then that we should say things like that. In my comments to you, Dale, I suggested that we not shy away from some of the more pointed comments that were made for example by the woman from Puerto Rico, about how she feels that Medicaid is treated in Puerto Rico. That's where I think a direct quote would actually very much convey the concerns that she had.
MS. WARD: Yes, I think the dramatic problems that these people are living with can only come out by their quotes. Some of the people from those other tiny islands trying to find specialists in any kind of form.
DR. STARFIELD: While that's true though, there are some things that could be put out very clearly, starkly in tables or in lists, what are these different places eligible for. Is Puerto Rico eligible for Medicaid, and if not, why not? Just so that we understand what it is we are dealing with in the political area. Does it require an act of Congress to do something.
DR. IEZZONI: Some of the tables go through that.
DR. MOR: Barbara's right, eligibility is not there. The tables all focus on the data issue side of quality or relations of protectorate states.
MR. HITCHCOCK: I'm not even sure eligibility shows up that much in the transcripts either.
DR. IEZZONI: The woman from Puerto Rico was quite clear. I can't remember enough to articulate exactly. There is a very active Medicaid program in Puerto Rico. It's just that there is something about the federal matching funds or something that is different, that disenfranchises them compared to the 50 states. So that would be a very clear recommendation, I think.
DR. STARFIELD: But that's not a data issue.
DR. MOR: That's a political issue. That relates to whether or not the Commonwealth of Puerto Rico wants to be a state or wants to be commonwealth or wants to be a separate country. It has nothing to do with data.
DR. IEZZONI: I'm slightly at a disadvantage, because I can't remember the direct quote, but this woman made it a data issue. Exactly what was said by her at that meeting was very relevant. So I just think that Dale needs to go back and look at the transcripts and pull that out.
MR. HITCHCOCK: Sure.
MS. WARD: But I think unfortunately, Vince, having sat through the meetings, there is a political-geographic reason for all of their data problems. It's because they are in this peculiar relationship with the rest of the country. So I don't think we can avoid it entirely.
DR. MOR: No, I understand. But as I was sitting there, it was very interesting. I was fascinated to hear all this stuff, but I sort of said for the countries that "sovereign" countries, they are here by historical accident in some sense, because of the relationship the U.S. had with them, but there are other islands that are closer, that were not represented, because there wasn't any U.S. relationship, or is no longer a U.S. relationship. So Panama wasn't here.
DR. IEZZONI: To be honest with you, the exact islands that were invited, were kind of determined by people other than us.
DR. STARFIELD: I think that was the IOM report, if I'm not mistaken.
DR. IEZZONI: Lynnette, the Department of Interior helped us figure out who should be invited, and Puerto Rico was the place that Hortensia was really interested in. But when Hortensia said Puerto Rico, we thought that's not fair to just focus on Puerto Rico, we need to go a little bit broader. So the U.S. Virgin Islands is the other Atlantic property.
MS. ARAKI: Because the U.S. Virgin Islands is also under the -- the Department of Interior has responsibilities there, and so they said that they had similar problems, and them recommended that if we are looking at insular areas, we needed to the U.S. Virgin Islands.
DR. IEZZONI: So we got explicit recommendations from the Department of the Interior about who we should invite. One of the aspects of this was remember that the federal government is giving these places money. Even though they may be sovereign countries, we are still giving them money. So there is an issue about tracking what we get for our dollars too.
MS. ARAKI: Right, because of their compact agreement allows them to receive categorical grants and block grants. Actually, that IOM study was sponsored by HRSA. There are some recommendations there for the government. Some of the recommendations are interdepartmental. It's not just how can you service this, but it would involve coordinating better with Defense. The military hospitals is where a lot of these facilities use as referral for tertiary referrals. It would involve the Department of the Interior, because of their management of the insular areas. It's a cross-cutting thing.
DR. NEWACHECK: I wonder if it would make sense to separate our recommendations for the different jurisdictions based on how far advanced they are in terms of data and that sort of thing. Maybe it could be Atlantic versus Pacific.
MS. ARAKI: Region 2 versus Region 9?
DR. IEZZONI: Well, so far in terms of clear recommendations we have Elizabeth's observation about this specific office that disappeared, and therefore the insular areas lost an identity or a focus for an identity.
MS. WARD: They lost a problem solver.
DR. IEZZONI: Right, and had to address the health data issues that were raised. Were there any other recommendations that popped out to people?
DR. NEWACHECK: I had a thought about the balance or the emphasis given to different sections of the report. It seems like we have a lot of information about existing federal activities in the jurisdictions, but we didn't have even a separate section on that much about what's going on in each of those jurisdictions locally. That is, what is the sort of infrastructure within the areas, combined into a section of the current status of these together into one piece.
I think it would be helpful to have earlier in the report, some more clearer discussion, maybe a table that shows where each of the jurisdictions are with regard to basic data systems, vital statistics, the question of morbidity data.
DR. IEZZONI: That was in the recommendations.
DR. NEWACHECK: So if that could come early on, as opposed to at the end of the report.
DR. STARFIELD: I think that's a good suggestion, to be with combined with Elizabeth's, which seems to me the only reasonable thing we can do is to say there are problems there, and there has to be some office responsible for it.
DR. IEZZONI: If I remember though, there were recommendations that could relate to the use of other technologies, the use of video conferencing, other ways of training people.
MS. WARD: I remember some good testimony from the National Center for Health Statistics about at least the vital statistics system, about things that they felt could be recommended. Things that they, as an agency, could do to make them more capable of doing even those basic systems.
DR. STARFIELD: The question is, does that require tax dollars to do that? That's something we have to worry about, can we recommend our tax dollars being spent for American Samoa?
MS. WARD: I don't remember those particular ones being more money-related.
DR. IEZZONI: It was just another way of using the same money.
MS. WARD: Yes, and how to better organize their training, and how to help them use the resources they had. I think that's what struck me about that office. They weren't getting lots of money for these countries, but they were facilitating what money they were getting, so that there was a better utility of the stuff, and they could help each more. Because if you are out there in the Pacific all by yourself, you don't know that the island next to you is actually trying to do the same thing. You could put your resources together. That's what that office was trying to do.
DR. MILLMAN: I said this at the time. What's helpful is to have a sense of -- there is a laundry list of endless needs here. What is helpful is if you can put some additional money or some additional effort, what is the first thing you want to do, and how does it relate to the most important problems, especially focusing on the data areas, because I think the other issues are complicated and diverse.
Maybe Elizabeth's is the answer, whether it's the focal point plus. But what would be most helpful I think is a clear sense of what is the most important thing you could do to make the data situation better. I think it has something to do with cooperation among those islands, because one of the big problems is getting them share resources given their different situations, and some focal point for negotiating those kinds of agreements and technical assistance.
DR. IEZZONI: I think also, this is just kind of repeating the point that I just made, training people, so they can perpetuate themselves in their local environments. Training the trainers kind of attitude. Some of the stuff that we read was really, really interesting about how you do case finding for drug abuse, and how in different cultures you have to address things in very different ways.
So having a single way of doing these kinds of activities isn't going to work. But if you can train leaders to know basic principles, I think that it would allow you to have the local cultures be able to perpetuate --
DR. MILLMAN: I think somebody did say use our existing educational resources, and build upon that, and link them to what makes sense in terms of technical assistance for them.
DR. STARFIELD: I do have some hesitation about recommending these specific when in fact they are unrealistic given what the reality is in law and regulations.
DR. MOR: I have to think about this from the point of view of opportunity cost. The kinds of training for nosologists, data outreach people, et cetera, the same needs are probably identified on every reservation of Native Americans and the Indian Health Service, and in some Mississippi delta area, and on, and on, and on in local health departments. So I'm not sure how to fit this into some broader framework of a long list.
DR. NEWACHECK: I think that applies to a lot of the work in this subcommittee and the issues. And that means we are neglecting something else.
DR. MOR: I guess I just have to ask the question. I can certainly understand Puerto Rico, and I can certainly understand I guess the Virgin Islands, although I don't really know that much about the political differentiations, but I'm not sure I understand all of the other components.
DR. NEWACHECK: It sort of struck me that what comes across -- I wasn't at the hearings, but from reading the report it said that most of these areas, with maybe the exception of Puerto Rico, are pretty far behind the rest of the world, the rest of the U.S. anyway, in their data collection efforts.
The recommendations that are made here -- there are no real hard, specific things that look expense to me. They are just basically saying you need to help build some infrastructure. I don't think that's a bad set of recommendations to have. If they are part of the U.S. or they are related to the U.S., we should treat them as part of the U.S., and not give them second-class status.
DR. STARFIELD: Can I read something from a British medical journal this week, this ethical principles that people are working on. They met in Tavestok(?), and they came up with something. I just want to read one of them. This is a several country effort.
"Hence, those working in health care delivery may be faced with situations in which it seems that the best course is to manipulate the flow of the system for the specific patient segment of the population, rather than to work to improve the delivery of care for all. Such manipulation produces more flaws and the downward spiral continues."
I think that's something that we have to think about in this subcommittee where we have limited resources. If we can do something that improves things for everybody, it is more equitable than picking out specific population groups, because this is not generalizable.
I'm saying I basically don't agree that --
DR. NEWACHECK: These areas are so far below the rest of the U.S., we could bring them up to a certain level before we could have policies that would apply to everybody.
DR. STARFIELD: I don't think we know that, because we haven't studied the Indian reservations.
DR. IEZZONI: Well, this project is specifically about these areas. I think everybody will understand that our recommendations are specifically about these areas. If we choose to take on the southern Mississippi delta or Indian reservations, we can do that. But this project is specifically about these areas, and I think that we owe it to the people that came in, and the federal agencies that are spending money there to come up with some recommendations that are relevant to these places.
I kind of agree with what Elizabeth is saying. NCHS had some ideas.
MS. WARD: They came up themselves. They said, well we do X in Tennessee. Why couldn't we do that here? It just takes more money. It just takes how we organize ourselves.
DR. IEZZONI: Exactly. So I think that what we are trying to do with this report is put these territories on the radar screen. If we do that, maybe people won't look at the radar screen and notice, but at least that's what we tried to do here.
MS. WARD: I agree with you, Barbara, I think it would be totally unrealistic for us to say all we need is $30 million more to do X for these particular areas. That would be absurd for this committee. But I heard people talking about non-resource, organizing what they are doing now, differently kinds of stuff coming out.
DR. IEZZONI: Also I heard that if we, as a committee, validate for some of the representatives who came and spoke to us in July, validate the need for health data, that in itself is an important enough statement that they can take it to their political leaders and say, look, this committee says that what we are doing is important, political leaders. You heretofore have not been that interested in health data, but these people validate our concerns.
That that might help some of the folks who came to meet with us, convince their political leaders that what they are doing is necessary for improving the public health.
MS. WARD: As I recall, a couple of them said if I could go back to my administrator with a report, that when they are going to decide to spend $10, instead of doing it here, I could get them to understand to spend that $10 on data, you could help me do that. That was one of the reasons they thought it was worthwhile spending 24 hours flying in an airplane to come talk to us for a day.
DR. IEZZONI: Yes, some spent 36. So I think that that's another recommendation. That we support the need for data to improve public health.
MS. WARD: We have done that for states that have been struggling about how to deal with their own resources. We have given them impetus to try to spend it on data rather on something else.
DR. MOR: Is there some way we can differentiate between Puerto Rico and/or X from the others? I hate to put them all in the same tub. If I were on the outside, I'm not sure I would read it, because it's like putting apples and grapes and things like that together. It's very different.
DR. NEWACHECK: Well, we could have a general set of recommendations or principles. Then we could have a subparagraph on Puerto Rico or the Virgin Islands.
DR. MOR: I thought there was some very interesting stuff about Puerto Rico which actually gets a little bit lost in the midst of all of this breadth.
MS. WARD: Yes, in fact I think we have to start out that way. That's what I really liked about your comments, Barbara, is that if you hadn't been there, it didn't start far enough back about the fact that everybody got struck with the fact that we were inviting a whole bunch of people that were not alike.
DR. IEZZONI: That's what I said in my comments to Dale. The diversity has to come across at the very beginning. Then we have to try to come up with the basic principles, as Paul said. You know, Vince, we are going to have to go back to the original transcripts to look at the specifics about recommendations for Puerto Rico, because I'm just blanking on them. Elizabeth, do you remember?
Olivia, you were there. Do you remember for Puerto Rico, specifically what the recommendations were that related directly to Puerto Rico?
DR. CARTER-POKRAS: Actually, in the discussion that we had with the person from -- I think it was Mary Ann Freedman from mortality, she was saying that a lot of the issues that Puerto Rico raised were very similar to what we have heard from some of the states. Their continued battle has been with information, and trying to improve coding on death certifications, and things that were very similar.
MR. HITCHCOCK: The hardware too. They didn't have mainframes. They didn't have PCs.
DR. IEZZONI: The hardware is these four areas too in the Pacific.
DR. MOR: There was an order of magnitude difference.
DR. IEZZONI: Yes, I know, but we could say computers. Puerto Rico needs a really big, fast one, and the others just need a 286.
MR. HITCHCOCK: Remember there was a report that NCVHS and the group that George van Amburg headed up. They went to Puerto Rico. They looked around, and they discussed that at our meeting. We have that for background.
DR. IEZZONI: We're going to have to go back and look at the specifics of that.
DR. MOR: For instance the hardware issue, part of that is that there are different trade arrangements with Puerto Rico than there are with U.S. states in terms of taxes and other kinds of things, which complicate some of these issues, and data exchange. Those are quite different than some of the countries in the middle of the Pacific.
I sort of feel that relative to -- there are a whole bunch of folks there, and they have a uniform ID system. They are part and parcel in many senses of the U.S. health data system. That, I think, should be recognized as quite different, and maybe some greater emphasis should be placed on some recommendations from that. At least that would make sense from my vantage point.
DR. IEZZONI: We would all agree that's what we need to do.
So I would like to just welcome Joan Turick, who has worked with Dale on getting the draft report ready. We're just giving Dale our feedback.
We have had several recommendations around the table. We're going to have to go back and look at the transcripts to come up with some of the more specific ones. The Virgin Islands also has a slightly different set of concerns that were distinct from what the Pacific territories were concerned with.
DR. CARTER-POKRAS: Actually, things have gotten even more strange so to speak in the Virgin Islands. They have had complete reorganization of the department of health. The folks that we talked to aren't there any more.
DR. IEZZONI: Oh, really? A little coup.
DR. CARTER-POKRAS: So we'll have to think about that.
DR. IEZZONI: We're going to want to send this draft report back to the people that came to visit us. So I assume, Olivia, when we send it to the Virgin Islands, we send it to the person in the position that we invited, rather than the individual. I don't know, what do you think? The new person?
MS. ARAKI: She was speaking for the government.
DR. IEZZONI: Any more recommendations?
MR. HITCHCOCK: Let me just ask how the group feels about this recommendation in relation to what Hortensia had talked about earlier. She had three long recommendations in self-determination, inclusion, and social justice and equity.
DR. IEZZONI: Those are basic principles.
MR. HITCHCOCK: Do we frame recommendations within those principles? Or do we just talk about those principles?
DR. NEWACHECK: Couldn't we make the argument that those are principles in general that apply to everybody, and that the following are our specific recommendations for each of the groups.
MR. HITCHCOCK: Although the idea of inclusion sort of struck me as not a bad principle and a recommendation. Basically, she is saying that agencies should attempt to include wherever possible within their legislation, these islands in their programmatic activities. I don't think we do that as much as we could now. It's just really a broad recommendation.
DR. MOR: I guess that's where I have some difficulty. I don't know why we're including the Republic of Palau(?) as opposed to another republic that has nothing to do with the United States from a point of view of "inclusion." There are lots of things that aren't being included. I don't have a rationale for why. Puerto Rico and the Virgin Islands and Guam, that is presumably understandable, American Samoa, but some of the others --
MR. HITCHCOCK: You have to go back and read the Compact of Free Association that we have with these islands. We have some obligations to them.
DR. MOR: I guess then the question is whether those obligations are sufficiently inclusive to be incorporated into a concept of inclusion. I don't know.
DR. NEWACHECK: I think the concept of admission(?) is a very general one here. It's just really saying that we have given them second-class status, and we need to think more appropriately about their role and bringing them up. It's not so specific as to block us into something that we have to do. It's just a general principle as I read it. I don't see it as constraining us or locking us into something.
DR. IEZZONI: Joan, do you have something?
MS. TURICK: I think it makes it hard that you have the three entities that are not really fully associated. I see the Marshall Islands as somewhat different because of the radiation. I also think that --
DR. MOR: Actually, if I remember correctly, the Marshall Islands, all we're going there is there is a sort of case control study. That's just a case control study.
MS. TURICK: It makes it hard to recommend everybody have a -- I didn't understand the recommendation from the IOM report, that there be a regional data system that would be the same for everybody. Clearly, what we have here is in the first place we want them all at the same status, and I don't know how you mandate this on the others. I don't know how this relates to the who coordinates it.
But I guess because of the fact that they have really been dependent upon on us for so long, I think we have some obligation to help them to a sustainable independence, and that may be well be helping them put in the kind of infrastructures they need to function without aid.
DR. MOR: I guess that would help me, that kind of statement up front, a lot more than what currently exists.
DR. IEZZONI: With some historical responsibility?
DR. MOR: Yes, I like that. The sense of historical responsibility, and some rationale for this grouping.
DR. MILLMAN: Well, there are also practical reasons. The Americans give a lot more money to other governments across the country than to these, per capita even. Recommendations are made all the time to states to do interstate compacts and to do regionalization for practical reasons, not because the federal government is mandating it.
So if it makes sense that these should cooperate because it creates a critical mass, then that is rationale enough. And how it gets accomplished may not necessarily be the job of this committee to recommend that actual mechanics and tool, but to say it makes sense from a data and a health standpoint to do some cooperative arrangements.
DR. IEZZONI: Although I wasn't sure that I was hearing the desire to cooperate with each that much.
DR. MILLMAN: No, they're not. That's one of the problems.
DR. MOR: In fact, there's no way to get from one of those to other places without going through a third place. So these are spokes. There's no tire out there.
DR. IEZZONI: I'm not sure that I want to impose cooperation on people.
DR. MOR: The true academic.
DR. IEZZONI: Or committee chair. Those are good comments, and that's very helpful framing.
I must say that Hortensia's language has a very political ring to it that we might need to tone down a little bit, because we can't afford to have this report be marginalized by again making pronouncements that really are not going to resonant with the realities of the current situation.
So I think that if we do have a set of basic principles, that it would be nice to have them more organized around data issue than kind of life in general, which is how these three principles are organized.
DR. MOR: Let me ask the question, what is the applicability of HIPAA to Puerto Rico?
DR. IEZZONI: Yes, we talked about that.
DR. MOR: I didn't see that in here. Is it?
MR. HITCHCOCK: HIPAA will be -- they have to comply with it. They weren't really aware of HIPAA.
MS. TURICK: They already have a unique identifier, which we are having a little trouble getting here. And they are beginning to go to managed care. One of the things that struck me about Puerto Rico is the fact that they are not included in the Census sampling frames. That has affected their ability to get CHIP money.
I don't think it makes sense to try to necessarily include them in the Census sampling frames, but if you could have comparable surveys there which Census would bless. One of the problems is that none of the data that comes out of any of these areas is officially sanctioned by Census. There needs to be some way to tie in.
DR. STARFIELD: Is there a census in Puerto Rico?
MS. TURICK: There is the biennial census, but that's not going to get them the data for CHIP. The states are pooling three years of CPSs to try to get the --
DR. STARFIELD: They don't do CPS in Puerto Rico?
DR. MOR: No.
MS. TURICK: It's not the sampling frame.
DR. MOR: That's a really important issue.
DR. IEZZONI: That needs to come up then. Was it in here?
DR. NEWACHECK: Well, if we're going to do CPS, we might as well do HIS too then.
DR. STARFIELD: That's the whole point.
DR. MOR: That's why I was arguing for some sort of differentiation, because I don't know --
MS. TURICK: I'm not sure it works very well. Maybe what you need to do is have comparable data collection that can be merged with these others.
DR. IEZZONI: But those are questions that we can lay out. We don't need to answer every question in this report. We can just lay out issues that we really cannot resolve at this point, but need to be considered by the powers that be.
MS. TURICK: I thought we had a recommendation in the report. I forgot to bring my copy.
DR. STARFIELD: What are doing that's different than what the IOM did?
DR. IEZZONI: The IOM report mentioned data only in passing. We actually had a presentation from the IOM at the July meeting.
Other comments? Now do you feel that you and Joan have enough that you can work with at this point? I really want Hortensia to be involved as well. I know you've gotten her a copy of the report.
MR. HITCHCOCK: We talked in a conference call.
DR. IEZZONI: I think what we should do is why don't you guys work on another draft, and then circulate it among the subcommittee. We will promise to get comments back to you with a week of receiving it; very quick turn around time. Then let's have you send it out to the participants of the meeting as a draft without the comments on it, and all the agency people like Mike. It would be good to get your comments, and Olivia your comments.
And Roy Linn(?) from the Department of the Interior. It would be good to get her comments, sending it around to all the participants. So what we should do is try to get it out for comments as soon as we can.
I think unfortunately, I'm sorry that Marjorie isn't here. Lynnette maybe you could just take this issue home to her. It looks like we're not going to be able to present our recommendations for the chair's signature until the June meeting, which is kind of shame, because we're going to have this fallow time, but it might take a little while to get the comments back.
MS. TURICK: I have moved to a new job. I've been told that I can assist, but that --
DR. IEZZONI: Dale is going to be the lead. Dale has already admitted to being the lead on this. Thank you very much for your work, Joan.
So are there any other concluding comments on this right now? Anybody? Paul you were going to get some comments to Dale. If any of the other subcommittee members, or any people around the table have private comments that they would like to give to Dale or Joan, please do so.
MR. HITCHCOCK: Thanks for the comments today. This is very useful.
DR. IEZZONI: Thank you very much, Dale, for all of your work on this. The George Washington people used their quotations very effectively I think in their preliminary Medicaid managed care report. So you might want to take a look at what they did on that, to help you with how to bring some of this out.
Let's move on to the next agenda item. Olivia, I see that you're here. Were you going to discuss the discrimination?
DR. CARTER-POKRAS: Actually, there were some other people who really are more highly versed on this. I'm trying to get Pat Russell here, but evidently she is out for an entire month. I tried to get Tatlee Lawman(?) and she has been out of the office. Steve Malog(?) and Tony Mandalin(?), when we originally talked about when we thought it was going to be in the morning, and they already had days planned.
DR. IEZZONI: All right. I just raise it because Marjorie has said that Jim's panel will talk about it tomorrow morning at the department overview, first thing in the morning. So we haven't forgotten about it, Olivia.
DR. CARTER-POKRAS: There were some materials that I passed on to Carolyn. I don't have anything to add to that.
DR. IEZZONI: With that --
DR. CARTER-POKRAS: Hopefully, after SPARKS(?) maybe we'll have a little bit more to add.
DR. MOR: I don't recall getting anything in the mail. Was it e-mail?
MS. RIMES: It was sent to everybody e-mail. We'll check on that. It was work from the Data Council.
DR. IEZZONI: With that, Barbara, you wanted to talk to us about the SES?
DR. STARFIELD: Well, Marjorie wanted to be here, and I think it's important for her to be here. She's been very involved in this.
DR. IEZZONI: I thought she wanted to be here for the discrimination piece.
MS. ARAKI: Do you want me to call her?
DR. IEZZONI: She's still on the conference call.
DR. STARFIELD: Can we do the follow-up discussion form January 22?
DR. IEZZONI: No, because Kathy Coltin needs to be here for that.
DR. MOR: Is she coming?
DR. IEZZONI: I don't think she's coming today. We were hopefully going to do this actually tomorrow during the quality working group. Kathy has to be here. She's the one who has to set the agenda.
We can talk about stuff if you like. What I can do is I can lay out what I was thinking that the full committee would do. Do you want me to do that right now? Sara Rosenbaum isn't here yet.
DR. MOR: This is the subcommittee/quality work group plan you're talking about?
DR. IEZZONI: This is the follow-up from the January 22 meeting of the next steps that the full subcommittee would undertake, and the other is a related piece that is the quality working group would take.
Oh, here is Marjorie. So we can maybe talk about that at the end of the hour. Marjorie, Paul and Barbara wanted to wait until you were here to start.
MS. GREENBERG: Oh, I appreciate that. That's the end of that conference call.
DR. NEWACHECK: I'm not sure why I'm on this list. I think I wrote you an e-mail asking to be put on the agenda for information, but Barbara is the expert, so I'll defer to her.
DR. STARFIELD: This is an issue that seems to have fallen between the cracks of the subcommittees and the committee, and it has to do with collection and standardization of data on various socio-demographic characteristics. It has to do with the fact that we have been concerned all along about whether the core data elements the committee came up with a couple of years ago, were adequately being considered in administrative simplification.
Among those core data elements are a whole bunch of things that were on enrollment. We have been assured all along that in fact, the core data elements, most of them were under consideration or being taken care of. There was an e-mail a couple of months ago from someone in the Data Council that pointed out that socio-demographic data, in particular race/ethnicity data and social position, whatever you want to call it data, were in fact not being dealt with.
The situation has become more and more complicated as various e-mails have flown back and forth, but let me try to describe the situation. It is most common for these kinds of data to be obtained on enrollment, and that in fact is what the committee recommended in the core data elements. But unbeknownst to me, and maybe most of you, I didn't realize that enrollment data is not part of HIPAA.
MS. GREENBERG: Well, that's not true.
DR. STARFIELD: Okay, good.
MS. GREENBERG: The transaction, the 834, is one of the HIPAA transactions, however, employers who initiate this in many cases, are not covered by HIPAA. It's kind of the odd part of it. So the transaction is one of the HIPAA transactions. If some plan provider exchange of this transaction, and the employers can use the transaction, but they are not required to in the way that providers and payers are required by HIPAA.
DR. MOR: How is an 834 transaction to be regulated if the origins of that transaction aren't covered under HIPAA?
MS. GREENBERG: It's strange.
DR. MOR: I'm glad you said that, because I was a little confused how it was possible that could happen.
MS. GREENBERG: See, there might be an exchange of enrollment data between the provider and the payer. Then it must be in this format. But to the extent that the employer initiates it, the employers are not covered. So they can voluntarily choose to follow the directions.
Also, I believe neither Medicare nor Medicaid uses the 834. They have their own enrollment.
DR. MOR: But are they going to have to under HIPAA?
MS. GREENBERG: No, they don't have to use the 834.
DR. MOR: Does the 834 have specs at this point that includes --
MS. GREENBERG: It includes demographic data.
DR. MOR: And specific code sets articulated by the committee in the minimum data set or not?
MS. GREENBERG: It has specific code sets articulated by X12. There are race/ethnicity, marital status, and citizenship, which is actually not like what country are you from, but are you a citizen or not, variants of that. It's a very strange code set, actually. These code sets are not great, but that's another story.
Although they are on the 834, they are situational, and they only can be collected when there is an agreement between the plan and whoever is initiating it that these are needed for enrollment. Actually, the inadequacies with the race/ethnicity coding is something that CDC, with the blessing of the department, is taking to both X12 and to HL7.
DR. MOR: Will they take the new OMB requirement?
MS. GREENBERG: That's what we're working on, getting them in congruence with the new OMB directive, although --
DR. MOR: It's like writing computer code. It perforates the responses. Once you change a gear, it has to change --
MS. GREENBERG: Right. In each case we are trying to make it so that now it only allows a single entry, it will allow more entries. And then getting the code sets congruent also. Actually, I'm told that even if we meet with success with this even in 1999, it would not actually impact on the standards that will be promulgated and have to be adopted 24 months later, because that one has kind of been frozen. I don't know that's true either.
It's complicated. It's on the 834. Another piece is even if you got it on the 834, state health data organizations at least, many of whom are trying to collect race/ethnicity in the hospital discharge data sets have no access to the enrollment transaction, the enrollment data. They have no access to it. They don't usually have a unique identifier. In a few state they do, but in most states they don't.
So what they have been doing is collecting it on the encounter transaction. These same elements -- it's actually a demographic segment that is in everyone of the transactions for that matter, but in the implementation guide for the encounter transactions, this demographic segment is marked "not used."
DR. MOR: And the encounter is marked "not used?"
MS. GREENBERG: It's situational in the enrollment, and not used in the encounter, but it is there. One of the things we are discussing in the department is allowing that to be used. But that is probably something that would have to be worked out.
DR. STARFIELD: You can see how complicated the situation is. There are some people, Hortensia among them, who believe the data should be collected on encounter forms. There are people, John Lumpkin among them, who think not, because they think this maybe enhances the possibility of discrimination if this is part of an encounter. So there are good arguments on both sides of that debate.
There are various things we could do, make recommendations on as a subcommittee and a committee, not necessarily mutually exclusive. One of them was suggested by Bill Braithwaite in a e-mail somewhere in that, who said that well maybe gentle persuasion over a few years will get people to use enrollment and transmission enrollment forms.
MS. GREENBERG: He's saying maybe if the department encouraged this transmission. I should just point out that even those of us who are talking about allowing it to be recorded on the encounter, it would only be where there is already an agreement to do it say in these states. It wouldn't be required of anybody. It wouldn't put a new requirement on anybody. It would only allow the encounter form which is currently being used as a vehicle to collect the information, to continue to be used in that way, which is what a number of state data organizations want.
DR. STARFIELD: There is another alternative, at least for the social positions data, and that is to do things by geo coding. If Dan were here, he would speak to this, because I think he feels rather strongly about that. Vince, you wrote a memo on that too.
DR. MOR: I have used it. It's very, very efficient.
DR. STARFIELD: That could be obtained on an encounter form.
DR. MOR: But the encounter needs to have an address. There is lots of encounter data that doesn't have address.
MS. GREENBERG: Or the address -- they might have zip codes, but they don't have access to the address.
DR. STARFIELD: Well, zip code is fine too, isn't it?
DR. MOR: No.
MS. GREENBERG: Zip code really isn't.
DR. STARFIELD: Well, I know, but what's the alternative? If people don't get address, what is the alternative?
DR. MOR: If one really cares about this for social position classifications, zip code is -- you'll make a lot of errors.
DR. STARFIELD: Well, that's true, but is it better than nothing?
DR. MOR: I don't know. I'm not sure.
MS. WARD: No. Your data can be so erroneous, it's not worth printing.
DR. MOR: The ecological fallacy is really massive. Of course it is differential by zip code, by area.
DR. NEWACHECK: Even by address, isn't it?
DR. MOR: Block coding is down to the GPS slot, and that's literally 200 yards per side. If you've got an address, every address that is non-rural postal in the U.S. can be GPS identified.
DR. STARFIELD: Well, we got an e-mail from Nancy Kreigor(?) yesterday, who I worked with at Kaiser --
DR. MOR: Nancy and Sally are doing this right now for all of Massachusetts and Rhode Island.
DR. MILLMAN: And South Carolina is address mapped, almost 90 percent of it.
DR. MOR: All health data.
DR. STARFIELD: Clinical data?
DR. MOR: Well, they are going to clinical data as well, and reporting data, that is health conditions reported, reportable conditions. So the sacrosanct nature of those files is really, really quite something.
DR. STARFIELD: Yes, they have grant from -- was it AHCPR?
DR. MOR: No, it's NICHD.
DR. STARFIELD: I think maybe we ought to get that project officer. Nancy is on sabbatical.
DR. MOR: Nancy and Sally work, and Sally Zeller(?) is in my department. They are co-PIs on this thing, and it's working with Dan and the folks at the health department in Rhode Island. They are in the process of doing that. A lot of it is available. They could actually solve the whole problem very easily if they sent it out to a commercial vendor, but you can't do that.
MS. RIMES: Well, the cost is fairly exquisite.
DR. MOR: It's going to cost one way or the other, whether they have their own research assistants do it, or a big computer algorithm. But the problem is confidentiality.
DR. STARFIELD: Nancy pointed out that when she worked at Kaiser in Oakland, that they couldn't even do things beyond the zip code level, even though she worked on it for quite a long time.
DR. MOR: Maybe their address.
DR. STARFIELD: But they weren't getting addresses.
DR. MOR: Right, see that's the thing.
DR. STARFIELD: That's in a far advanced system.
DR. MOR: Well, see, because their addresses are on there, and are on their enrollment records. There they are at Harvard, and those enrollment records are irregularly updated. There is no proactive outreach system for updating enrollment records.
MS. COLTIN: Even if a more accurate address comes in on the claim, a more recent one, it doesn't get updated.
DR. MOR: There is no automatic updating. Check the little box on your electric bill.
DR. NEWACHECK: I still wonder how accurate it is to measure a socio-economic status or a class. I'm thinking about my own neighborhood. There are a lot of people in my same block who are urban professional types who are making probably good incomes, but there are also some very poor elderly people that have lived there for years.
DR. MOR: You can do an age gradient. Those things are reasonably well understood. But it's an awful lot better, much, much better than zip code.
DR. IEZZONI: So where are we going with this discussion?
DR. STARFIELD: There is another alternative. We should think about what the committee wants to do. Another alternative is to in fact start with data sets under federal jurisdiction, the Medicare, Medicaid, Bureau of Primary Health Care. To start there, rather than to take on HIPAA.
DR. IEZZONI: Start there and do what?
DR. STARFIELD: And come up with some suggestions, such as geo coding, or those data sets that are under federal jurisdiction.
DR. IEZZONI: For the purpose of who, to analyze what?
DR. STARFIELD: It's not for the purpose of who, to analyze what. It's to get as part of core data.
MS. WARD: But who is looking at the data? It's the purpose to which the data --
DR. STARFIELD: The AMA says we don't need all these data, because we just don't need it. So why collect it? But there are people who think that this is very necessary in terms of looking at quality of care across populations. Some groups with 2010 objectives say it's imperative to do that. We can monitor any of those objectives for disparities if we don't have this information.
MS. WARD: I'm not arguing by pushing. I think that's part of the problem, is that we haven't made the case to the people who have to be trying to do some of the solutions. I think that's true with all the problems we've got with either race/ethnicity to geo coding. No one who has to change behavior to do something differently about coding something somewhere that is not now, now, has not been convinced that it's worth changing behavior. I think that's part of our real problems.
If we could make a better case of why it's needed and who could use it, we might get some light at the end of the tunnel. But I think that's one of the things that our committee could do, that could be very useful. To see if we can make a case that can be substantiated for having that kind of data.
DR. MOR: But I think the question is, what's the case? Should the case be as the original proposal from the full committee was to have demographic and ethnic information on an enrollment record. That is one kind of case, one kind of structure, one kind of problem.
A whole different case needs to be made for whether a recommendation would be that some of this information should be on encounter records? One side or the other. Then a third case has to be made for making available geo-coded information linked to records that could be individually identified in some manner.
MS. WARD: I think there is a real problem saying we can't do it for everybody, so let's just do it for the people who have no choice but to be on Medicaid.
DR. STARFIELD: We have Medicare.
MS. WARD: Medicare would be better if you had it.
DR. STARFIELD: We're not going to solve the problem today. The question --
DR. IEZZONI: Actually, I'm glad that Ann and Sara came in, because I was going to try to come up with a proposal at this point. Barbara, this is obviously an issue that you've thought a lot about, and feel strongly about. Why don't you see whether you can identify some other members of our subcommittee just informally, and come up with a proposal for the subcommittee as to exactly what we could do. Should we hold a day of hearings? Who should be involved? What would be the end result of that? Would it be recommendations? To who, kind of saying what?
Just come up with a tiny little work plan, and so the rest of the subcommittee can look at it. David, you might want to be involved.
DR. TAKEUCHI: Yes, I'd like to. Hortensia is going to be here on Thursday. Maybe the three of us can talk about it.
DR. STARFIELD: I know that Dan is interested. Vince, I think you are.
DR. MOR: Yes, I'm somewhat knowledgeable. I think the big issue is really what's the defined agenda. If it's reasonable to just say Medicare/Medicaid, that's a whole bunch of people. At a minimum, the government should figure out how to do this in some coherent manner.
DR. IEZZONI: Okay, well, why don't you guys talk about it among yourselves and come back to us with a proposal for what this subcommittee could actually do. Because it's obvious that I think people are interested and concerned about it, but it's not obvious to me yet exactly what we should be doing.
MS. WARD: The problem is like this. If we take one piece of it, maybe we can feel successful with one piece.
DR. IEZZONI: Exactly. We need to find that piece to feel successful.
Why don't we move on then. Thank you both for coming. I'll just make the observation at the beginning that we're really glad that you're doing this for us, because it looks like you have really made a great start, and we're really excited about the way that the final report is beginning to look.
I'd like to separate our comments today into two pieces. Sara, could we talk about the draft contract purchasing language. Kathy is hoping to present the raft contract language to the full committee over the next couple of days sometime for them to look at and comment on. What we are hoping is that we can maybe move the recommendations about the draft contract language forward.
We don't have another NCVHS full committee meeting until June. So if we don't present this in the next couple of days, we won't be able to move forward with the draft contract language.
MS. RIMES: What we had handed out at the beginning of the meeting was an updated contract, which all of you should have.
DR. IEZZONI: Right. It has a date on it. It says 1/2/99.
DR. STARFIELD: Do we have a copy of the comments I wrote on the Medicaid project?
MS. RIMES: We handed it out and laid it on the chair before we started. So make sure everybody has a copy.
DR. IEZZONI: All right, so this will be dated January 25th?
MS. ROSENBAUM: This is Draft #4, Working Draft.
DR. IEZZONI: Sara, could you just summarize how it differs from Draft #3? Does it differ a lot, a little?
MS. ROSENBAUM: There is a revised and expanded introduction, a lot of additional commentary, and Part 2 probably has significant revisions. Part 1 is pretty much the way it was by Draft #3 I believe.
DR. IEZZONI: I think actually, the lengthy introduction is very helpful in giving justification for what we were doing.
Kathy, you are going to be presenting this to the full committee over the next couple of days, because you've been the only committee member who has been very involved. Are there any federal agency representatives around the table who have been involved? Mike, you have been. Anybody else? Gail, have you been involved? CDC has been involved. SAMHSA has been involved.
MS. ROSENBAUM: SAMHSA has been involved, HRSA, Kathy, and HCFA of course. At the maximum there were about seven people on the working group that produced this first round. But nobody outside of, at this point, either an agency or essentially the delegated body of an agency like this one. So it's not been publicly vetted at all.
MS. GREENBERG: The Medicaid folks and HCFA have been involved?
MS. ROSENBAUM: Yes, they have been involved. They have been on every call throughout the process, which is actually a departure from the normal specification. They tend to come in later, but with this one they came in at the beginning.
DR. IEZZONI: Great. How has the process worked for you, Sara?
MS. ROSENBAUM: This kind of process is very good, and this group was particularly strong. Sometimes a working group actually is less able to give the kind of direction that one of us really needs. This group was extremely competent. This group was able to take drafts, comment on them, very efficient. We did a series of basically 90 minute phone calls to produce this document.
I started of course with a very early version that you all may remember last fall, and then it went through these rounds. At the point at which you all decide that it's what you want to begin to circulate for public comment, then it's ready to go. You may decide you're ready; you may decide you want to make significant changes or modest changes, whatever.
This was actually quite a difficult draft to produce, because it is very different from our other specifications. Most of the other specifications deal with an illness, or deal with a population. One is sort of talking about the overt parts of managed care buying. In this case, what you are dealing with is the thing that holds the product together essentially, the data information it produces.
There are two sets of issues. One is what data requirements belong in the contract, and the other is what are the duties of the managed care organization, both directly and potentially through its subcontractors to collect and then report or produce the data.
What makes this so difficult is that the contract has to be read against a very complex mess of federal and state laws on data, as you know. Some of the data requirements that are alluded to here either parallel, or in some cases may actually overlap federal or state law. And our recommendation to any purchaser is if you are a purchaser whose law deals with an issue, you don't want to start writing whole new specifications into a contract unless you have a compelling reason to do so.
The biggest area where this is the case has to do with public health data of course. Public health data laws are horrible outmoded. In fact, one of our major recommendations to this committee, and we have already made it to the CDC is that there needs to be a very thorough review of certain kinds of public health laws, the biggest one being the availability of out-of-state laboratory data back to health agencies.
There are a couple of states that have developed what I would call state-of-the-art laws. Whether they are implemented and they work the way they are supposed to is one thing, but the law itself is the right law. The law basically requires licensure for any laboratory that does business in that state, meaning test samples drawn in the state, and then requires as a condition of licensure, the reporting back to the state health agency. That's the way it should be when you are dealing with national companies.
Unfortunately, our sense is that very few states have that kind of law. While this problem is magnified by managed care, as Kathy correctly pointed out, it is a huge issue for any kind of health care service, because of the nationalization of suppliers, particularly lab suppliers.
So there is only so much you can do with your contract. It's a very inefficient thing to have 8 million purchasers in a state all trying to write in specifications regarding data that should be part of state law. We do raise of course, and note that especially on lab data, there is an alternative pathway here, which would be CLIA.
Whether for CLIA certification purposes, a laboratory should be required to report its data back to the state in which the specimen was drawn, and we recommended again to CDC that both issues be explored. Obviously, it will be a recommendation to HCFA and to the committee and to everybody else, that this really is not something that ideally one would handle in this kind of an agreement.
The draft of course attempts to capture classes of data. Some of the classes of data are issues that certainly present challenging problems for you to think through, for example, racial and ethnic identifier data.
DR. IEZZONI: Well, Barbara, I knew we were going to hit this in the Medicaid managed care report. Can you tell us what the issues are from your perspective around that one, Sara, because this is a topic that we're concerned about.
MS. ROSENBAUM: Sure. The central thing that a buyer wants, at least certainly a Medicare and Medicaid buyer, but I would assume based on speaking with colleagues who buy for private sponsors that it's true for private buyers too, and I'm sure Kathy knows this better than I do, the central kind of data that is needed is encounter data. At least as far as access and utilization goes, you want encounter data.
The encounter form would be completed of course by the hands on provider in theory, whoever that happens to be at the moment. I must note parenthetically that if you read this you note that there is no definition of an encounter. We used Kathy Coltin's definition with a lengthy footnote, noting that there are other definitions that one might use. This is a disaster.
If there is no definition of an encounter, then the central element needed for managed purchasing is going to yield you very different outcomes. Of course I was so flabbergasted I went back to the statute on managed Medicaid, and I couldn't believe there was no definition of encounter.
At any rate, the encounter form is filled out by a provider, whoever that provider is. The provider may of course never have face-to-face contact with a client, because if it's say a wheelchair manufacturer, for encounter purposes the supply actually is reported as if it were an encounter, but the manufacturer would fill out the invoice or whatever, and it would be attached to an encounter form.
So you have many kinds of encounters where nobody is ever eyeballing the patient number 1 and number 2. I would tell you that I have great concerns about racial and ethnic data being captured by providers for lots of different reasons as part of their job. I think that they are an exceeding unreliable source of information. And I think that it introduces an element into the encounter that this country, for unfortunate reasons, I think is not sophisticated enough in many places to deal with.
But that's my own personal view. The fact is that in Medicaid the issue may be slightly less insurmountable than in other areas, because the eligibility files that state welfare offices maintain should have data on race, ethnicity, and national origin as part of the profile of the individual at the time of enrollment. So the encounter forms could be matched back up, and that would be true with Medicare as well.
I would assume that the enrollment data would be the place that you would have someone self-assess his or her racial or ethnic status, and it would be recorded. You could, I suppose, instruct providers to provide racial or ethnic data. I have no idea how reliable the information would be. So we give you a long footnote here about racial and ethnic data.
DR. IEZZONI: We heard from the field very clearly that people need to know what is happening to different racial and ethnic minority groups. They need to know that. So somehow I think we were going to have to recommend somewhere that somebody collect it. So we're going to have to deal with this issue for the Medicaid managed care report. Do you have any recommendations?
MS. ROSENBAUM: Well, I can tell you who it can't be. It can't be the company, because often in Medicaid managed care, nobody in the company will ever have any -- the one place the company could collect it, and again, Kathy stop me if I'm wrong, is if it's a state where you get the enrollment tape from the state, both the voluntary enrollees and the auto-assigned people. You have a new member form that has to be sent out to all new members, meaning new in the sense of this is their first enrollment month. You ask them to indicate their background. You probably get a 20 percent return rate on the form.
You could put it at the provider level, and many, many people would never make a contact with their plan during the period of their enrollment. Those who do, you would have only slight confidence in the data. So in the end --
DR. IEZZONI: You get really biased results if you were just looking at --
DR. MOR: That's why these data on the Medicare tapes tend to be a high rate of missing data.
MS. ROSENBAUM: So then in the end where you are left is the one place where there is actually sort of a custom of collecting the data in public system, is the point at which you enroll in a program; not enroll in managed care, but enroll in the program. It is customary actually to collect lots of information about the background of people when they enroll, whether you are signing up for taxation purposes. This is a public program. For private managed care, I don't have a clue.
DR. IEZZONI: We're going to have to deal with this.
DR. STARFIELD: You know that the HIPAA specifications for the enrollment, there is no requirement there.
MS. ROSENBAUM: Right. So you have two problems then. You have the locus of the information, and you have the potential for transmission of that information. Of course you have a third problem, which is if it's two different sources of information, whether anybody is going to have the resources to do the matching if the encounters are in one place, and the profile of the patients are somewhere else.
DR. STARFIELD: When you said it couldn't be the company, you meant the managed care organization or something like that?
MS. ROSENBAUM: Yes. There may never be a contact.
DR. MOR: So if I understood you correctly, if the contract specified, or if we wanted to say the contract should specify subgroup analysis, let's say we won't even specify what subgroups to have reporting go back to the purchaser, in this case Medicaid, you would have grave doubts as to the validity of any racial and ethnic data that the company would have in their encounter data to give it back to the state, unless the state gave them that information?
MS. ROSENBAUM: Yes. In that case I suppose the state could give them the information by extra data runs. Why the state wouldn't just take the raw data at that point and do its own analysis, I don't know, but yes.
DR. MOR: Because your draft contract says you've got to be able to give us reports that look like this every quarter and break out this and this and this characteristic. But they can't break out those characteristics if they don't have the data.
MS. ROSENBAUM: No, either the plan collects it through a member analysis, or instructs its providers to indicate observational data on the encounter forms, or the state is supplying the data, one of the three.
DR. STARFIELD: Well, there is also another way. At the community health centers, for example, they will give aggregate data, not individual data.
DR. NEWACHECK: Sara, do you have a sense about how many states in their Medicaid agencies have that data?
MS. ROSENBAUM: Racial and ethnic data? Well, as I recall, in the old days the 2082 data actually had the table with the racial or ethnic identifiers. Which means that the MIS system certainly has the capacity. I would guess that if we collected and looked at all 50 Medicaid applications, you would see that you are asked to indicate your background.
DR. MOR: I think it's in the MMIS.
DR. NEWACHECK: So your recommendation is to link and send it separately.
DR. IEZZONI: This is actually going to have to be a major topic in the final report, because if you go back over the transcripts, you will see this again and again and again. So all these recommendations and thoughts and kind of the background that Sara just laid I think have to be addressed.
I actually floated this notion of having the race and ethnicity data for the Medicaid managed care reporting at one of our public meetings recently, and was kind of shot down by people being concerned this data question. So we're going to have to justify it and frame it.
DR. CARTER-POKRAS: I just wanted to ask whether you were able to come up with any more information about the role that identifiers play in being able to link or not link the encounter with the enrollment data. Because we have heard on several occasions that this was a real barrier. Either the health care plans don't have identifiers, or those identifiers aren't the same ones that the state would have, so that makes it difficult to link the encounter and the enrollment data.
MS. ROSENBAUM: I think other than instructing the plan to assign an identifier, and being capable of linking that, we didn't explore whether or not the system is working. In this document what you can do is try and make sure you have got the instruction right, putting aside the issue of whether the plan executes its obligation to assign an identifier that can be linked back. What we have to be sure we've done, and we should check to make sure we've done it properly is to make sure that the plan is instructed to assign an identifier that you can link back to the enrollment data.
DR. IEZZONI: So let's have a discussion from the subcommittee, and Kathy, maybe you could lead, because you will be presenting this over the next several days to the full NCVHS. Do you have any questions or comments for Sara as you are thinking about presenting it to the full committee?
MS. COLTIN: Let me just get some clarity. I was not proposing to present this entire document. We have half an hour. What was I trying to get to was the set of underlying principles that we had agreed to as we started going through these positions, that then cut through a lot of the provisions. They had to do with things like where is the locus of accountability for collecting particular data.
I'll give you a couple of examples. We talked about that if plans are required to submit encounter data to the states, complete encounter data, then it would not be reasonable to have provisions that require plans to also submit reports that the state itself could create from those encounter data. In other words, it's kind of a one or the other. You either give them all the data, and they can create the reports, or you would create the reports only if there was a cost reimbursement for it.
MS. ROSENBAUM: Right. We indicated that where you want to buy, is you want to buy runs. You want to buy 100 different analyses. You can buy analyses that the typical contract will some data and some analysis, and of course that's an issue that I assume is a negotiation between the buyer and the seller.
MS. COLTIN: But it wouldn't be part of the standard rate, the per member per month capitation or whatever. You would submit the data to the state, and they could go ahead and do it themselves. But if you had a state -- for instance, we heard from a lot of the states that they didn't have analytic capabilities, they didn't have the analysts, the trained people to actually make use of the data.
They could decide in a make or buy arrangement that they want to buy that analytic service back from the plan, and buy a report. But they would have to buy them, as opposed to suddenly layering on more and more reporting requirements to the plan without adjusting what they are paying. So that's kind of an underlying principle about how we went through this.
We also talked about it in connection with a lot of public health data. There is a temptation, a very real one, to try to make managed care organizations responsible for all the failings in all the state laws, and all the gaps in information systems in the entire health care by saying, well, gee, we could get it from them.
What we tried to do is say where is the right level of accountability? So for instance, do you hold a managed care organization responsible for physicians' complying with notifiable disease reporting? Or is that something where the accountability really resides at the level of the physician. The state should be enforcing it.
They figure because they can't enforce it, we can get the managed care organization to enforce it for us. What we were talking about in these provisions is that the best that you could ask the managed care organization to do would be to have a provision in its subcontract that says that the physician should be in compliance with all applicable state laws regarding reporting.
But then are you going to then make the plan responsible for enforcing that? That was an area of contention, because now you are putting the managed care program in the position of a state regulator with regard to individual physician reporting. So it's that temptation to make the managed care organization responsible. I think we tried to reach a reasonable compromise.
MS. ROSENBAUM: It was also complicated by the fact that with the exception of a few state laws liability, such as the Texas statute that has been enjoined under ERISA, most states do not yet recognize managed care organizations as health care providers. If you opened up the public health reporting statutes for the state, even though a managed care organization is actually a hybrid creature that just like a hospital, both provides services and insures the services it provides, the state could define it as such, that is not the avenue that states have taken.
I don't think it's because they have thought about it and discarded it, as much as there is simply no policy thinking going on at that depth regarding under what circumstances should a managed care organization be characterized under state law as a provider.
And the real problem comes not even so much the physicians, who, as Kathy noted, are under their own legal obligations to report, it is the suppliers. It's out-of-state suppliers where the states have not yet changed everything from their lab laws, to their DME laws, to all the laws on supplying ancillaries or goods and products where people are buying from out-of-state companies, and those companies have knowledge of a notifiable event or critical event under state law, and the state has not closed the loops.
I think one underlying theme is not just the accountability, but that the proper functioning of data in a managed care world is going to depend both on the contract and on the extent to which state legislatures have thought through the implications of moving to this model as they write their public health statutes, their notifiable event laws, their reporting statutes under domestic violence, or other statutes where public health authorities want data reported and can't get it.
MS. COLTIN: We also I think discussed the distinction between when you have this problem without suppliers -- the state where the person who is receiving the service lives requires and can't. The question then comes up, is the patient using a provider outside the state because the managed care plan requires them to do that?
In other words, if we sent all of our labs to a central lab that is outside of Massachusetts, does the responsibility then shift from the state in trying to enforce compliance, to the MCO to enforce compliance. So that in that situation where it's the managed care plan that is creating the situation where there might not otherwise be reporting of the data back to the state of residence.
That then the managed care plan would not only have language in its subcontract about that vendor complying with reporting to the resident's state, but then would also have an enforcement responsibility in that situation as well. And would have to demonstrate that it had monitored in some way, the compliance of the vendor with that subcontract provision.
On the other hand, where they have not created that situation, where it just exists naturally with a supplier in a state, and that supplier isn't complying, then one can argue that it's not the MCO's jurisdiction, it's something that really should fall on the state to take care of.
So it was those kinds of distinctions, how much do you build into this contract language to hold managed care organizations responsible for every possible situation in which data gets generated, or ought to get generated?
MS. ROSENBAUM: Well, the thought that you would hope, particularly data that you need because the entire community needs the data, anything involving a public health event, has of course a compelling connection to the company, and the company's knowing what's going on in itself and its community. But the reason the state needs the data are its police powers. They need the data to protect the public health.
To leave this matter to the contract basically puts public health at risk, because no matter how many people hit our Website and download this and use the model data -- if the Medicaid agency in every state used this form, you would have 10 percent of the purchasing covered. So 90 percent of the purchasing is outside, unless all employers in the world come and visit.
One issue is whether HCFA Medicare is closing the loop on lab data for state public health authorities, on data on notifiable events where they really could. They certainly could single-handedly on the regulatory side, take care the labs problem through CLIA. On the Medicare side, obviously if they add their voice to purchasers who wanted certain data, at least they will have subcontracts' reporting requirements complied would be a good thing.
It seems to me another principle that came out was that while encounter are the central data, actually there was a fair amount of consensus on the need for characteristics data, networks, access. These is not just about counting up the claims forms and doing some outcome studies. You need to know that you are buying from a company that at least on paper, has the basic capabilities to get the job done.
DR. IEZZONI: Even the linguistics. I noticed how many providers with this language live in this neighborhood with this language. I was struck by that, because it seemed like they were asking for a lot. I was really surprised by that. Who felt really strongly that that needed to be included, because it seemed to me, Kathy, a managed care plan might argue that it would be really hard for them to get that kind of information. We heard when we went out in the field that collecting information on language spoken by patients is hard to get, let alone the race and ethnicity.
MS. COLTIN: Most plans are being faced with having to do that anyway. This is one area where it isn't just Medicaid state agencies that are asking for this. There is increasing diversity in work place, and large national companies are applying with increasingly diverse populations. So we're finding that they are asking for it in their purchasing specifications as well.
So a lot of the plans are starting to even put their provider directories up on the Web. I know ours is on the Web, and you can just go in and find a provider. The language spoken is there. It's even something you can search on. It's the cost of doing business these days.
DR. MOR: Some of the information is not just this demand. It's also a marketing price.
DR. NEWACHECK: Sara, have you gotten any feedback from managed care organizations besides Kathy's?
MS. ROSENBAUM: No, this has not been publicly vetted at all yet.
DR. NEWACHECK: So we don't really know the answer to that, other than what Kathy is replied.
MS. ROSENBAUM: Although I can tell you we vetted about seven other sets of specifications, many of which deal with -- they all have their one data and information requirements, which will drop out once these are done. Then when you go to our Website, you see that you go to one unified place for data and information specifications.
But in all the other vettings, the managed care organizations in those vettings have never objected to supplying that data. They are much more concerned with the issue Kathy raised of having to do special utilization or outcome runs in addition the encounter data and certain basic runs that I guess are negotiated as part of the buying and selling.
DR. MILLMAN: I think there is another principle as we talked about this. This is partly a menu of what states could require to support their policies. I think that's sort of a key thing.
MS. ROSENBAUM: Yes. If you look at round four, you will see that in front of every single section --
DR. IEZZONI: Oh, good.
MS. ROSENBAUM: Each section, and many subsections are broken up with notes saying this is an option for purchasing. Now there is one interesting issue though, which while we put this obligatory language in front of every subsection saying if you are interested in a specification relating to X, you might want to look at this language.
The balance budget act does establish minimum requirements. I thought about that after putting in the obligatory language that ironically in the data we have more obligatory requirements for states. You have to produce encounter data.
DR. IEZZONI: We can't define what it is.
MS. ROSENBAUM: Right, but you have to produce something called "encounter" data. That's exactly right. So at some point we will have to note that while all of this is optional in the sense that it's just our way of drafting this, you do have to come up with something like, unless you're going to tell everyone that you're not going to produce the data.
DR. IEZZONI: Kathy, I would like to distribute this to the full committee over the next couple of days. I understand that you're just going to have a half an hour. We're not going to be able to have a full discussion on it. I think it will be good to begin to float this a little bit more publicly, and getting it out to the full committee, Marjorie, means basically that it goes out to whoever is interested in what our committee does.
MS. GREENBERG: Whoever is at the meeting here.
DR. IEZZONI: So we can begin to at least maybe request that committee get feedback back to maybe you, that you can then forward to Sara.
Have you thought, Sara, about what the endgame should be for this?
MS. ROSENBAUM: Yes. Once we get the green light from you, meaning that the committee members, if they have comments, have sent them to us either through Kathy or through another route.
MS. COLTIN: I prefer to get myself out of the loop.
DR. IEZZONI: Who should we have? Who is the e-mail person that they should be forwarded to? Carolyn? So forward all e-mail to Carolyn.
MS. ROSENBAUM: Once we get the comments incorporated, you then need to decide what kind of a public vetting process you want.
DR. IEZZONI: Right, that's what I was asking. Do you have any ideas?
MS. ROSENBAUM: Well, this is not a process that has $25,000 in it for public vetting. So I strongly recommend the inexpensive approach, which is that we send this out to review to 15 or 20 people. We put them on a conference call and mix up agencies, plans, consumers, just like we've done with all of our other specifications. We conduct a 90 minute conference call. We get their comments.
We then write up the comments for you all, because as our clients you need to decide whether and to what extent you want to deal with those comments. Once we have had another discussion with whoever the delegated person is or people are for the committee, we then do the final. Then when it's finalized, with your permission, we then post them. If you have specific people you want.
MS. COLTIN: To the associations, because they will then make sure that get comments from the industry.
DR. STARFIELD: The committee members or the subcommittee members?
DR. IEZZONI: Committee members.
DR. STARFIELD: Are they going to have enough background?
DR. IEZZONI: Kathy has a half an hour tomorrow.
MS. GREENBERG: You're here Thursday as well, Kathy?
MS. COLTIN: I'm here Thursday morning.
MS. GREENBERG: I was going to say we probably have some additional time Thursday afternoon.
DR. IEZZONI: I just think that we ought to get this out to the committee members, Kathy's statements of principle.
MS. GREENBERG: Are you going to be looking for some kind of approval from the committee? Or that would probably be premature?
DR. IEZZONI: No, exactly. I just want them to have it so they can get comments back to us. What we would say to the committee is get comments back to Carolyn within maybe two weeks or so. Then that would mean maybe you could move to the vetting process that you described, Sara, by early March.
MS. ROSENBAUM: Yes, if we get comments by the middle to end of February, and we take another week or so to clean them up, then we can make sure that we've gotten from any of you -- that's the other thing. If you have people or organizations like AAHP or HIAA whom you absolutely want us to send a set to for dissemination to selected people, that's great. If you have individuals you want us to send them out to, then we will send them to those people, as well as the people that we have used all along in our focus groups.
We have a standing group of six or seven Medicaid directors, and then CDC, SAMHSA, and HRSA all nominating a couple of people whom they would want to have look. We would find a couple of consumer advocates, especially from the hearings if there were people who were particularly interested in data, to read and comment. Then we can put them on either one, or a series of conference calls.
MS. GREENBERG: This is just a detail, but it's up to Carolyn. Like with the annual report, people were just asked to send their comments directly to Susan Kannon(?). We could have them send them directly Sara, unless you prefer to get them. I just don't want you overburdened.
DR. IEZZONI: We just have to have a clear e-mail address.
MS. GREENBERG: It's up to Carolyn.
MS. RIMES: Whatever you guys decide is fine. I'm all right. We can do some kind of melding.
MS. ROSENBAUM: Whatever you prefer, because we are used to compiling pages of --
DR. IEZZONI: Why don't we send them to Carolyn, because that way she can have a sense of who on the committee has responded, and all the personalities. So that might be helpful.
MS. COLTIN: But we're saying that we wouldn't put this up as a final product and that our recommendations were good provisions for states to consider and adopt until after the June meeting?
DR. IEZZONI: Yes, at the June meeting.
MS. ROSENBAUM: It's going to take at least that long for us to get the stuff out the door. Maybe there will be a relatively small number of comments, but my guess is for this kind of a document, there would be substantial comments. The most important step is going to be our compiling of the comments, and then some decisions about what comments you want take, how you want us to deal with the comments.
DR. IEZZONI: We are going to have to schedule a meeting for May I think, for this subcommittee to go over the comments and help Sara and Ann finalize the Medicaid managed care report, and maybe also the islands and territories report. Kathy, are you now the person that we turn to, to talk about scheduling? Okay, so let's try to get a date on people's book sometime in May, so that we know we can move to the full committee in June. I guess it could be early June. The full committee meeting is in late June.
MS. GREENBERG: You don't have another meeting planned?
DR. IEZZONI: We don't have any other meetings scheduled. I know that we're going to need them, but we don't have anything on our books. That's why I say we need meetings scheduled. We should try to maybe schedule two.
MS. GREENBERG: That's what I was thinking. If you can do it in a two day meeting, that would be good.
DR. IEZZONI: Let's see if we can find a two day period in May then.
Does the subcommittee have any comments for Sara right now about the draft?
DR. MOR: I'm not sure how familiar you are with the data requirement components of the big consortium of New England states all getting together looking for the dually eligible?
MS. ROSENBAUM: No.
DR. MOR: I just don't know whether you guys have thought about the provisions, whether there is any extra information or extra issues around the dual enrollees. It is the coming issue. All the states now have got mandatory, or are moving toward mandatory managed care for the AFDC populations, and so on and so forth.
But they are grappling now with the dually eligible. The data requirements, issues of functional information, et cetera, might be a little more pertinent there.
MS. ROSENBAUM: What I can do is call one of the Medicaid agencies in the consortium.
DR. MOR: Massachusetts, I don't remember the name, but they actually have a Website as a New England consortium.
MS. RIMES: Sara, we can pull that for you.
MS. ROSENBAUM: That would be great, because actually I think all the basic components are probably here, but the --
DR. MOR: There might be some extra things. I don't know this material.
MS. ROSENBAUM: It's the phenomenon of having two payers needing the same data. It's a very good point.
DR. IEZZONI: Comments from anybody else?
MS. GREENBERG: So this Draft #4, including the cover memo, you want for everybody who wasn't here today to get?
DR. IEZZONI: Yes.
MS. RIMES: And what would be nice is if we could have them mail me comments.
DR. IEZZONI: Maybe write Carolyn's e-mail address on there for comments.
DR. STARFIELD: How different is it from the draft we got and commented on?
MS. ROSENBAUM: Depending on what your comments were, there could be almost no differences or huge differences. So what I suggest is that I get the comments from you, and take a look at them. I could be that I've taken care of a problem, or it could be that I haven't even begun to address an issue.
DR. MOR: May I ask one question on the draft contract items on page 29? It's the section on ownership of data. I know this has actually gotten much longer in this version. Since I don't really understand all the implications, could you just walk us through what some of this might mean to health clinic working for plan, which is working for some umbrella, which then has to respond to the state?
MS. ROSENBAUM: Sure. In the world of public contracts, Medicare and Medicaid, under federal contract regulations, my understanding -- and of course these are the same regulations that bind me as a contractor -- the federal government owns data. I can retain uses in certain data, but the federal government owns the data.
As I understand it, and as Kathy indicated, that is emphatically not the case when private sponsors buy managed care products. In fact, the ownership of the data never shifts to the sponsor. The sponsor can request reports off the data, information off the data, and pay for whatever it wants, but the issue of who owns the data is not governed in the same way that is when the federal and state government is buying data.
Now that means that we have a footnote. There should be a footnote that says -- or I suppose if DOD were to buy these data, the federal government generally owns the things it buys.
DR. MOR: So CHAMPUS is buying from managed care contractors, they own that data?
MS. ROSENBAUM: I presume the DOD has kept the same rights that HCFA would keep in the data, or that a state agency keeps, where it's its own insurer. If you are GW buying insured products, you don't own the data. The company owns the data.
DR. MOR: Of your employees.
DR. STARFIELD: And HRSA owns the data.
MS. ROSENBAUM: Because that's federal grants and contracts, absolutely. This is just in a managed care product. If a person was buying managed care products, HRSA owns all health center data for example, they would own MCH Bureau data.
DR. MOR: So what this then means is that if this is a draft model Medicaid contract, then the Medicaid agency or the government therefore owns the data in that case?
MS. ROSENBAUM: Yes. What we have tried to do here -- it's not easy -- but we've tried to write these specifications with annotations at the bottom to indicate when you come up against a specification that is really unique, either the one buyer or a couple of buyers. Like for example, EPSDT data would not be bought in a privately sponsored contract, and yet a Medicaid agency absolutely has to put that specification in.
So if you read it, many times the commentary at the bottom would say this really only applies to. We assume especially with this kind of a project, that we have people other than Medicaid agencies reading these specifications. Ownership is clearly one that is very special to government buyers.
DR. MOR: So I'm just trying to think of this from let's the managed care company's perspective. They have got this Medicaid data for their Medicaid enrollees that is co-mingled presumably in their system with data that is theirs. Now they have presumably all rights over their data, but not all rights over the co-mingled Medicaid data. Is that correct or not correct?
MS. ROSENBAUM: There are things they could not do with those data. They probably couldn't release them to any third parties without the sponsor's permission. They couldn't publish them potentially without the sponsor's permission. I'm sure that no state agency or HCFA would stop a company from using the data to do comparative analysis.
But of course where there has been a tremendous uproar about this -- I'm sure you've seen this in the committee's work -- is actually in the release of data to employers. The big problem is of course the self-insured companies reaching down into their own plans, and pull data on the health profiles of their patients.
MS. COLTIN: We're facing an interesting situation on that now. Without naming the employer, it is a very large New England employer who is self-insured, who has been conducting audits of how the managed care company is managing the care. They wanted to sit down at a computer terminal and not only pull out detailed claims data on their employees, but wanted at the same time to have open in front of them, detailed case management notes from the case management nurses about how those patients were being managed, which included the kinds of things that are in physician notes about family environment, whether there is an abusive husband.
All of those sorts of things that might be considered personal, and yet are relevant to the management of the care of that patient. They got very upset with us that we did not want to make that sort of information available to them, other than to indicate that they had a case management nurse, and these were the conditions being managed, but not the detailed notes that went along with it.
DR. MOR: So clearly they weren't interested in this in terms of questioning your efficiency.
MS. ROSENBAUM: I have waited for a case involving a state with strong privacy laws attempting to apply the laws to an employer, and having the employer sue to enjoin the law under ERISA, to see where a court would come down. Because I think that this is not so integral to the administration that the plan, that seeking private and confidential information that is identifiable to the individual, as opposed to counting widgets is not justifiable. But no case like this has gelled.
DR. MOR: Actually, I was raising the question not because of the individual focus on who owns the data, because I can't imagine an employer going in and caring about individual Medicaid patients, unless they might do that from the person's history.
MS. COLTIN: I think this is an areas where it would be very helpful to plans to have sort of clear guidance on the state and federal regulation. In this particular situation, we are not the only plan in our market that this particular employer contracts with. The other plans made the data available. We were standing on principle in saying we'll give you this much, but no more. We could lose the business over this.
MS. GREENBERG: The Subcommittee on Privacy and Confidentiality is planning to have some hearings on that.
DR. IEZZONI: This case -- because I know that they were going to have hearing on it.
MS. ROSENBAUM: In fact, one place where this may change you see, is there are proposed standards. There are a limited amount of standards in HIPAA. In theory there would be more proposed standards on this. Then of course you would be expected as a contractual matter, to comply with whatever the standards are in HIPAA. What we would do is do a commentary at the bottom of the page regarding what those standards are.
To the extent that HIPAA drops the bar down, and I'm sure whatever bar there is, is not very effective. It's probably like the rest of HIPAA, which is that by the time it passed, there were so many accommodations that the law has weak or limited protections.
DR. MOR: I guess my point was that if Harvard Pilgrim looked at the pattern of the case management or whatever of pregnant women and children who were Medicaid versus non-Medicaid, they would need to get government permission to do that analysis, is that correct or not?
MS. ROSENBAUM: They would want you to do that.
DR. MOR: I think that's important, because the way this is implied is that it's total ownership of all of the data resides with the purchaser in this case. Unless it's made explicit -- you might want to cross-reference.
MS. ROSENBAUM: We can say notwithstanding these provisions, the contractor shall have free use of the data to perform the utilization and studies required under the contract.
DR. IEZZONI: Other subcommittee members with their comments specifically on the draft language? Sara, it was really an excellent job.
MS. ROSENBAUM: It was a wonderful working group. Everybody was -- we didn't have to spend any time orienting anybody to what the drill was. It was a very good group to work with.
DR. IEZZONI: Why don't spend a few minutes talking about the draft Medicaid managed care report. Barbara, do you want to lead on this?
MS. ROSENBAUM: I have to teach, so I'm going to leave. My class starts in about 20 minutes.
DR. STARFIELD: I thought it was a good report. I just have some technical things.
DR. IEZZONI: Sara and Anne, thank you.
Are there any overall comments to begin with from anybody?
DR. NEWACHECK: I think it's a good report.
DR. MOR: It's a lot more concise overall than what I read last summer.
DR. IEZZONI: Well, we won't even talk about what happened before.
DR. MOR: Also, it's really great that it is six months later, because of the BBA, which really changes things.
DR. IEZZONI: I thought it was really well done too. I think that my overall comment though is that it's going to need to be kind of reorganized and remotivated slightly. I think it's going to have to lead more with the fact that our subcommittee was doing this, and why we were interested in doing it, and then some of the background about the different waivers are. The Medicaid background I actually found very refreshing too, because it laid things out in a way that those of us who aren't intimately involved with it, actually understood, which is very, very helpful.
But then I think it would be good to before going into the results of your survey, to go back out into the field to hear from some of the people that we hear from during our hearing. You do that in the last section. To kind of maybe bring that up a little earlier to get a sense of where people's concerns were, what the gaps were.
Then your piece would come in after that, to kind of say, okay, what's actually going on out there in the community, in the states with managed care plans. So kind of be motivated by specific comments. Have your piece come as what's the reality out there, and then go to the recommendations at the end.
DR. NEWACHECK: I think that's a good idea. It allows us to use our comments to define the problem.
DR. IEZZONI: And having the quotes was great. I would even encourage you to do a little bit more of that.
DR. MOR: I remember one of those on the MMIS system.
DR. IEZZONI: They were really good, and the woman from Families USA in Arizona. They were just really good, and they really motivated.
MS. MARKUS: That whole part in front of everything else.
DR. IEZZONI: I think it's going to have to be recast slightly so it's clearer that there is something coming later. Basically, the way the report reads right now is it's three separate sections. It needs to be integrated, and kind of flow as a whole. So it is word processing, but maybe starting with kind of an outline as to where you want to go with it, and then just taking the paragraphs from different places and moving them around.
I like the way that you dealt with the recommendations. That was good. That put accountability where it needs to be. I think we need to start talking about this. We have kind of been pussy footing around the issue of that we need to come up with recommendations for a while. I think we finally have to belly up to the bar and do it.
I think that we do have some time right now. Ann, if you've got a few minutes to stick around.
MS. MARKUS: Sure.
DR. IEZZONI: I have some thoughts, but I'd like to hear some others. Elizabeth, you are from a state. You and George were among the folks that motivated this whole endeavor back a couple of summers ago. Do you have any thoughts about where we should head with the recommendation part?
MS. WARD: No, I haven't thought about it. Sorry.
DR. IEZZONI: Maybe David, you weren't really involved. Vince?
DR. MOR: I think about this in relation to the data, so I guess in some sense the recommendation is the contract language. When Sara was here I was actually sorely tempted, because I was on these conference calls with these folks from Massachusetts, Vermont, when they were talking about the dually eligible contract specifications.
There we tried to sort of talk about the whole issue of a minimum data set. Actually requiring that common data elements be used, rather than just say the report should be available, and it should cover this type of topic and this type of topic. Because you can't make comparative statements unless you have uniform information.
I guess that's where I would start. Although I recognize that having said that, the level of detail becomes excessively proscriptive.
DR. IEZZONI: That was actually a clear recommendation that I think we have to field, that people wanted consistent data elements. They wanted the same definitions. They wanted to kind of be told this is the way that it's going to be done. It goes back to the core data elements way back when.
MS. GREENBERG: Or back to McData. At that time, when that effort was sort of disbanded, they said we hadn't made the case that this data has to be collected. Now the case has been made I guess with encounter data and the BBA. But what is likely to come out is to be some minimal version of the 837, unless something else is done. Right now that is the transaction/equivalent encounter data. We know what its deficiencies are.
I think for the hospital data that HCFA is asking for, it is like UB92 data, right?
MS. MARKUS: Yes.
MS. GREENBERG: So it would be that for the hospital and 1500 data for the -- actually, in a sense it is better than that in that the 837 really standardizes the UB92 and the HCFA 1500, rather than allowing 400 different versions of it.
DR. NEWACHECK: That doesn't get at issues of satisfaction.
MS. GREENBERG: No, it doesn't get at any of that.
DR. MOR: Something broader than just the event codes in order to understand who is in, who is out, who is treated, who is not treated, something like that. I guess as I looked at the implicit report requirements under the contract, it sort of calls up evaluative and comparative kinds of tables. As soon as you put those evaluative and comparative tables out I want to say, oh, this MCO's tables match up with this MCO's tables. That of course requires that there be a common set of elements.
DR. IEZZONI: I think that's a recommendation. We have to recommend that. We can't recommend exactly what the details are, but I think --
DR. NEWACHECK: Do we want to talk about maybe not the detailed level of individual variables, but do we want to talk about domains?
MS. GREENBERG: Do you want to suggest that somebody actually specifies -- Medicaid or somebody -- specifies the elements? It doesn't mean the committee has to.
DR. MOR: In fact, it might even be a recommendation for the contract, that the contracts say that in some way, or maybe the recommendation should say that if states are going to proceed with this, that they should sit down with "federal" guidance, that together with the participating plans, develop a uniform data set.
I guarantee you, as soon as you say with federal guidance and overlay, because of the number of plans that operate in multiple states, the plans themselves will insist that it be uniform, because you can't do business in 16 states having 16 different specifications. But on the other hand, if the federal government says this is the way you do it, everyone will say no.
DR. NEWACHECK: Well, maybe another set of standards could be applied.
DR. TAKEUCHI: Related to that, even if you have comparable data, I was struck that the definitions of encounter varies. So do we want to tackle that whole issue.
DR. IEZZONI: How do we define encounters?
DR. TAKEUCHI: Yes, encounters vary.
MS. COLTIN: I think it would be a very helpful contribution to do that.
DR. IEZZONI: Can we do it?
MS. COLTIN: I think it can be done. If we can get agreement on whatever we come up with outside of our -- I think we could come to some consensus about what we think it ought to mean, but whether or not we could get buy in to that.
DR. IEZZONI: It's going have relevance to our post-acute care activities. That's when you think about long-terms stays at nursing homes. Is that a single encounter? And home health -- if you have three weeks of home health care, is that a single encounter?
DR. MOR: An encounter only makes sense as a function of the billing unit or the reporting unit.
MS. COLTIN: Internally, and I can only speak for our plan, because I know there are other plans have tried to do this, but everybody does it differently, and that's where the problem lies. We link multiple units. The billing is the claim. The encounter is an aggregation of claims that stemmed from a patient contact.
So the patient comes in and sees their primary care physician, and that primary care physician orders some constellation of services for that patient, some of those services will come in on separate claims, but they would all get linked back.
DR. MOR: Even though they happened on different days, because there is an order number.
MS. COLTIN: We have algorithmic logic. But also the referral.
DR. MOR: The descriptor or whatever the order was written that day.
MS. COLTIN: So even if the test gets performed three days later, the order for the test or the referral goes down to the day.
DR. MOR: But the only way you can do that is to have in the system, a link across claims that is common up and above a date. In there system it has such a thing. In a commercial world, there is no such thing. So when HCFA does things like that, it just sort is either by the day or other bundles that have a trail, and no more than six days after, or something like that.
MS. COLTIN: The areas that becomes contentious has to do with whether you want to go further than the direct supportive services that come out of that encounter. Let's say that primary care physician says you need to go see a dermatologist. Now they go see another provider, and that provider orders services. Do those get bundled into the original, because they were initiated --
DR. IEZZONI: What I'm hearing is that it would be awfully hard for us to come up with a definition for encounter. Maybe we can do it.
MS. GREENBERG: I can bring for tomorrow, what we have for definition for basically an ambulatory care encounter.
DR. MOR: But the ambulatory care encounter doesn't try to attribute a medical device that actually is delivered three weeks later that emerged from that particular ambulatory care encounter. There is no effort to try to do that. It may or may not be important for cost tracking. It is probably important for individual management of an individual company for its quality assurance.
MS. GREENBERG: I'll bring you what we have just to throw in your hopper.
DR. IEZZONI: I guess what I'm saying is that we recommend that there be agreement on definitions and data elements. And that there should be some mechanism put out to do that, that involves the states. We might be more specific. Maybe with Sara's help, she could think about how to craft language around that and we could look at it, that wouldn't put too much of a federal overlay on it.
I think that we did hear from the field that they --
MS. COLTIN: There always is that tension that Medicaid is telling us how to do it one more time, but the same entities will also say, well, if they don't do it, it's never going to get done. We are wasting money dealing with the problems, because they are not going to do it. So sort of bypassing the complaints and getting to the solution, which they will probably be glad for.
DR. IEZZONI: Another recommendation that I heard clearly from the field that we talked about, about a half an hour ago was the need for race and ethnicity data. I can't remember anybody who argued that point, but I remember quite a few that you pull some quotes out of, of people who very eloquently argued in favor of having that information.
DR. NEWACHECK: Do we need a recommendation on vulnerable populations?
DR. IEZZONI: Everybody on Medicaid has to be at a certain income level, so I think that income issue is a little bit less pressing. I think that reason for enrollment was another issue, because of being a dually eligible person. But I do think that we have to be specific about race and ethnicity, because that was brought up very, very clearly, and Hortensia was especially eloquent in talking about that.
So I think the general overview is vulnerable populations, but within that is reason for Medicaid eligibility and language. That's a really good one too, language and race and ethnicity.
DR. STARFIELD: We ought to look at the core data elements. You have that don't you, Marjorie?
MS. GREENBERG: Oh, yes. Do you want me to bring that tomorrow? I'll send it to Ann.
DR. MOR: How about this for language? We would encourage states to develop specific data element definitions covering all data reported to the states under this contract, which would be incorporated into their Medicaid managed care contracts, which in turn are passed on to subcontractors.
DR. IEZZONI: For this report, we are going to talk in English. These recommendations have to do with quality. We haven't even gotten to the quality piece.
DR. MOR: But the quality stuff is embedded in the contracts.
DR. IEZZONI: But right now we're talking about recommendations for the report, not for the draft contract language. The report is going to be very reader friendly, regular English, no six syllable words, and no wherefores, or whereas. It's really nicely written.
Kathy, you haven't had the advantage of having read it, but from just having heard in the field, because basically this is the recommendations, so there wasn't anything to read about them. Do you have any recommendations for Ann?
MS. COLTIN: Well, one of the things that I heard in the field had to do with the lack of trained analytic staff. Even if we could agree on the definition of encounter and data on it, and make all that data available to the states, so that they could then apply standardized criteria for creating quality measures from the data, do they have the people to do that.
DR. IEZZONI: What would a recommendation be from that? We had talked about like remember how they used public health service grants to send people to health schools in the seventies. Would that be a recommendation? We can observe that that's a problem, but what would our recommendation be?
MS. COLTIN: We recommend that HCFA think about in terms of the kinds of technical support that they provide states, think about some sort of training programs.
DR. IEZZONI: Regional workshops, regional training programs?
DR. MOR: No, you don't learn this stuff in a three day workshop. There is no way you learn this stuff in a three day workshop.
DR. IEZZONI: Okay, what would you recommend?
MS. COLTIN: With some of the ASTI courses, you can learn an awful lot in three days.
DR. NEWACHECK: I think so too. If people are talented to begin with, picking up the specifics for doing this kind of work is not that difficult.
MS. WARD: ASTI?
MS. COLTIN: The Applied Statistics Training Institute courses.
DR. MOR: What are those? I know that HCFA now has data centers, and people go to these data centers for three or four day sessions, and they spend time doing SAS(?) and working on data issue for file construction. But I just had some people go to one of those in our group, and they walked back and said, well, we had a few examples. But that's not sufficient, I don't believe.
MS. WARD: As I recall, we heard from the states, the states that had gotten the most training, had spent their own state dollars, adding to the federal dollars to go do it right. What the states who didn't have that ability said was why does Medicaid demand we do these kinds of reports, but don't give us any dollars or an ability to deal with their indirect or their administrative costs to include this.
Unfortunately, we're going back and telling a federal agency to find more money, which nobody wants to hear.
DR. MOR: I recall hearing that there is an absence of people. There aren't live bodies.
MS. WARD: But some of the states who had no bodies, had state dollars to buy the bodies from a university. The other states said, we don't have the dollars to buy it. We would go buy it if we had the dollars, but we can't get our state general fund to allow us those dollars. I wish the feds would give us some flexibility about our administrative costs, to go buy it. We don't necessarily want to hire it, but we could buy it. But they don't have the dollars to buy it.
DR. MOR: So I heard both the money and a human resources issue.
DR. IEZZONI: So if we make an observation. I'm still a little unclear that we've got a really good recommendation about this.
MS. WARD: I think we have to talk to HCFA about what recommendation around that problem is something that they think is doable.
DR. MOR: Does HCFA do training for this kind of stuff?
MS. RIMES: They have held large group training sessions. There is a technical assistance arm that does work through this.
DR. IEZZONI: I agree, Vince, that something like that, you want people to go to one year masters in public health degrees. That's what you want. You want training fellowships.
DR. MOR: Yes, public health service or HCFA fellowships, or maybe even AHCPR fellowships, focusing on this whole issue of the analysis of these kinds of --
DR. IEZZONI: We are going to need more analysts. Jim, you are looking skeptical. We heard this everywhere, that there just simply aren't the people who have the training, who can help analyze the data and look at quality of care.
MS. WARD: I'm not sure that they have to go through a full MPH training program to be able to do this sort of thing. I think that there are combinations of more short-term kinds of training institutes. Even like the summer program at Hopkins.
DR. IEZZONI: All I'm saying is that it's more than a day. It's more than three days. We want to have a menu of options. Somebody is going to have to make an investment.
MS. COLTIN: Private industry has tuition refund programs or programs that if you go back to school, they pay half, you pay half, whatever. Something like that, that creates an incentive for employees in state government maybe to be able to go back to school and improve their skills. It's not so much that you can't recruit people who are trainable, it's that you can't recruit people who are already trained.
So we can recruit very smart, college educated people who have degrees in quantitative areas, but don't necessarily know how to do this sort of thing.
DR. MOR: I guess I'm going to have to float my public health notion here. This is different than just running numbers on production of how many units of aspirin went down the pike. It's slightly different, because there are issues about health, and knowing what these codes mean, which maybe requires teams of clinical people and analyst people. But it's been my experience that without the people having some notion of the substance, it is more difficult. I agree, you don't need an MPH kind of thing.
MS. COLTIN: I think you have to provide some notion of the substance, but I could name dozens of people that we have trained with the kind of background that I'm talking about, who can do an excellent job really in creating the kinds of quality measures and reports that we use.
MR. HITCHCOCK: It seems like in the back of my mind we heard something about the CDC model at one point, where CDC folks go out in the field, they do their epidemiological stuff, and end up getting hired there. Maybe this could sort of parallel that.
DR. MOR: What would a whole other interesting model altogether.
MS. COLTIN: Assignees, you mean?
DR. HENDERSHOT: There are probably three CDC programs that are similar to that, the Epidemiological Intelligence Service, the Public Health Informatics Fellowship, and there is a new fellowship that is also in that area too.
DR. MOR: There is a new one that actually relates to HCFA, basically with public health data stuff.
DR. IEZZONI: Let's move on from this recommendation. I think we are getting a flavor of where this is heading.
Our next category of recommendations, Paul?
DR. NEWACHECK: Well, I feel like we haven't really talked about how we're going to get information from the plan enrollees about access to care, and satisfaction about care. We talked a lot about encounter data.
DR. IEZZONI: The non-users we don't do.
DR. NEWACHECK: I think this gets back to recommendations of using population-based surveys to measure access and satisfaction modeled after the Medicaid CAPS in particular. That's a good survey, and I would like to see it used in managed care plans to provide comparable data across the country within the states.
DR. IEZZONI: Okay. And we're going to need to also recommend that it's going to need to be translated.
DR. STARFIELD: I'm not sure I'm willing to endorse CAPS just like that. Do have to see CAPS?
DR. NEWACHECK: I don't think it should be a separate one for every state. I think CAPS is the leading contender right now, unless you have an alternative.
DR. STARFIELD: No, but I'm not, but CAPS is really not tailored for Medicaid.
DR. NEWACHECK: Well, there is a Medicaid CAPS that is separate from the commercial CAPS.
DR. STARFIELD: It would be nice before we recommended it, if we looked at it.
DR. NEWACHECK: It is now a requirement in many states. It's becoming the standard. I think if we are going to recommend something else, there is nothing.
DR. STARFIELD: I'd like to look at it before I buy in to the recommendation. Is that a problem? Carolyn, we're talking about getting copies of the Medicaid CAPS, because we're thinking of making a recommendation.
DR. MOR: What did we get at the last meeting? That was just the Medicare one.
DR. IEZZONI: Okay, I think that's a really, really good point.
MS. COLTIN: But to build on Paul's point, it's in HEDIS, but only if the state requires the plan to do it. So it's the state's decision about whether to require the plan to do it.
DR. IEZZONI: So we are recommending that states do.
MS. COLTIN: Or the state themselves. HCFA for instance, is drawing the administration of CAPS itself. They didn't ask the plans. They drew the samples and they did it. Some states are doing it that way. Massachusetts is doing it that way. So the Division of Medical Assistance is drawing the samples and doing the survey. I don't think it matters.
DR. MOR: I would actually prefer a state to do it, because I would like to give it one time, given flopping and the lousiness of those administrative records across as they are distributed, from my experience, across different sites. I'm not sure some of the plans could actually identify their people. One arm knows who their people should be, and the other arm knows who they are serving.
MS. COLTIN: It should be who you are getting paid for.
DR. MOR: That's right.
MS. COLTIN: If you are getting a capitation, you are responsible whether you're serving or not.
DR. MOR: They don't have those updated addresses for the people they are getting paid for. They are just getting paid.
DR. IEZZONI: There was another category of recommendations I think I remember hearing from the field -- people help me with this -- about carve out services, and it was essentially mental health and substance abuse. That we need to have some way to link the experience around carve out services to what is happening elsewhere. Do people remember exactly what it was?
MS. COLTIN: There were a lot of issues around that. It was highly variable by state in terms of what they carve out. It was even variable sometimes within a state in terms of whether they carved it out for every plan, or left it in for some, and carved it out for others based on evaluations they had done about how well that plan managed that particular aspect of care. So pharmacy and mental health and dental are the three most common services the states carve out.
But there is also something called a carve in, which is when the plan subcontracts it out. I think that most states, although I can't really speak for all of them, treat the carve in stuff as if the plan provided it. The plan has the same responsibilities, the same obligations. They have to provide the encounter data, so they have to make sure that their subcontractor supplies it to them, so that they can in return, supply it to the state. That's the way HEDIS treats it, and that's the way NCQA treats it.
When the state carves it out, then I think it makes sense to make a recommendation that the state require the same encounter data from those vendors, that they require from the managed care organization, so that they can link the information.
DR. MOR: We might want to actually go further, to say if we would recommend perhaps that states consider the carve in strategy unless they had the capacity to link the data from the carved out data records with the non-carved out to the same population. Because otherwise --
MS. COLTIN: Well, part of the reason that some states have moved to a carve out is that they have a benefit package with the services that the plan has either provided or arranged for through a carve in.
DR. MOR: I understand. It makes a lot of sense in some particular areas. But that does leave the state with two sources of information. They have no way to actually rationalize --
MS. COLTIN: Well, worse than that, it leaves the managed care organizations without often some very important information to use to really understand the patient's care. If you don't have the pharmacy data, you really -- you can't tell whether patients are getting beta blockers after heart attacks, or whether people with congestive heart failure are getting ACE inhibitors.
You just don't know whether they are getting managed appropriately or not, because you don't have the data. The only place the data could come together would be at the state level, and that's why I'm saying we write a recommendation that at least it come together at that point, and that they require that data from their carve out vendors, so that they can look at those sorts of things.
DR. IEZZONI: Great. Are there any other broad categories of types of recommendations?
DR. NEWACHECK: I'd like to add one on data dissemination and availability, and having to do with I think this data be available in the public domain. And also having summary data be available to prospective enrollees, so that there is some kind of a report card that consumers would get, just like we're trying to bring in commercial managed care, that would allow prospective enrollees to make choices. It's really more about data use.
MS. WARD: Would that be standardized in terms of the kinds of report cards or process?
DR. NEWACHECK: It could be.
MS. COLTIN: Do states now have to report the encounter data to HCFA on Medicaid?
DR. IEZZONI: Under the BBA, and through the MMIS.
DR. MOR: All states are now liable under that, and that goes into effect sometime after 2000.
MS. RIMES: I think it started January 1.
DR. IEZZONI: We need to look at that.
DR. MOR: Well, the MMIS is already set. That was in the law. That's defined.
MS. GREENBERG: That's individual level data?
DR. MOR: Yes.
MS. MARKUS: It's the Medicaid information system, but it's going to be totally electronic with the BBA that changed. So the transactions become the state and HCFA are going to electronic, and it gives the option for HCFA to set standards for encounter data.
MR. SUMME: [Remarks off mike.]
DR. IEZZONI: This is not directly on that point, but it is somewhat related.
MR. SUMME: There is some language in here on confidentiality of data, but it's rather narrowly drawn, I think, perhaps reflecting current law. I don't know if you all would like to go beyond that or not.
DR. IEZZONI: Because what I was going to say is this is not exactly on your point, but it's somewhat related. That is that one of the things that we heard especially from the woman that I think you quoted Ann in Arizona, was individual recipients, especially kids, having to report the same information over and over again. And having the people coming and asking them X million different ways for the same types of information from different data gathering mandates.
One of the recommendations that I would like to make is that states look at the data that they are requesting from individuals and consolidate it. So it is done in as nonintrusive a way as possible for the individual. And then we could perhaps add the privacy and confidentiality section around it.
I think especially when you are talking about people with disabilities, and children with disabilities, basically this mom told us that she kept getting asked every month whether the kid had reached his milestone or not. It just got to be absolutely overwhelmingly burdensome.
DR. STARFIELD: There is the quality care issue. If the kid's got a good primary care provider, that shouldn't happen, unless the primary care provider thinks it needs to be asked.
DR. IEZZONI: The point is that there are various regulations that require for eligibility, that these questions be asked at least in Arizona and other places. That's what we were told.
DR. MOR: Medicaid, under the medically needy requirements.
DR. STARFIELD: They are being asked about development milestones?
DR. MOR: It's every month. It's like being homebound. Someone has got to attest to being homebound.
DR. IEZZONI: We try and limit the intrusiveness of the information collected from individuals.
Then were there any recommendations having to do with providers? We heard from some doctors, and I don't remember that they actually gave the most compelling presentations, with one exception in terms of recommendations. I remember in Massachusetts one was saying, well, they don't even money to buy computers. How can they expect us to report all this information.
I don't know whether there are any recommendations that people remember.
MS. COLTIN: I think in particular it was the community health centers.
DR. MOR: Even the community health centers have more potential capacity than individual practitioners.
DR. IEZZONI: It was actually a private practitioner who was directing this, who was talking about it. So I don't know how we would frame a recommendation around that. Buy computers. Somehow being sensitive to infrastructure demands of providers who traditionally care for vulnerable populations, that are not going to have the ability to meet some of these reporting requirements, because they do not have the infrastructure capabilities.
So they have some sort of program that would allow a pay back, or some sort of financial assistance to allow these providers to continue giving this kind of care.
MS. GREENBERG: Another meeting.
DR. IEZZONI: Even though they just need to beef up their information infrastructure.
DR. MOR: So you're contemplating sort of a subsidized loan program for health care providers?
DR. IEZZONI: These are people who give a lot of care to poor people in communities, and would have to drop out.
MS. COLTIN: Safety net providers.
DR. MOR: I don't understand. Safety net providers tend to be agencies or institutions.
DR. IEZZONI: There was an individual provider who was here the first morning of the Massachusetts testimony. Ann, you can go and dig that up. It's a woman. She was talking about how she cared for huge numbers of poor people, and she didn't have --
DR. MOR: I thought she was from a community health center.
DR. IEZZONI: She was actually somebody I had known 20 years ago.
DR. MOR: She worked for the community health center.
DR. IEZZONI: But she said they didn't even have a computer there. So the basic point is being sensitive to the fact that there might be high volume providers in the community, who may not be able to meet some of the data demands, and working with them around infrastructure issues. I'm not sure how to do it. Does anybody have any thoughts or suggestions?
Well, it's getting late. We're a little bit after the time. Are there any other comments or questions?
Ann, what would be your time table, given the rest of your work.
MS. MARKUS: Probably a couple of weeks.
DR. IEZZONI: Okay, why don't you get us another draft in a couple of weeks. We'll get comments back on that. What we'd like to do is maybe after that circulate it to the full committee. I can tell them that this is going to be coming when I give my subcommittee report over the next couple of days, so they're just aware of it. So have the full committee get the next round. Maybe we could have the same meeting in May where we talk about Sara's draft.
Any final comments from committee members?
MS. COLTIN: Just because Hortensia is not here, I remember one other thing that she wanted, which was with the non-traditional type of service providers. I wasn't always sure exactly who she was referring to, but she would bring it up, and then people would comment on it. That was an issue.
DR. IEZZONI: Let's see if there is something in the transcript.
DR. MOR: What kinds of providers was she talking about?
MS. COLTIN: Little ones. Little, specialized kinds of services.
DR. MOR: Like drug treatment kinds of programs?
MS. COLTIN: Drug counseling for pregnant women.
DR. MILLMAN: Prenatal services, that whole area is usually not collected. Encounter data doesn't really quite fit it.
MS. COLTIN: So it creates these gaps.
DR. MILLMAN: Transportation, case management.
DR. MOR: This is what we talked about when we started the craziness about the cab drivers having their own unique provider ID number for doing transportation.
DR. IEZZONI: For the subcommittee, tomorrow afternoon at Kathy's working group we are going to talk about the post-acute care initiative, so I hope that everybody can come.
So thank you everybody for coming. This was actually quite productive.
Ann, thank you very much. It was great. A good start.
[Whereupon, the meeting was adjourned at 4:10 p.m.]