[This Transcript is Unedited]
Quality Hotel
1200 North Courthouse Road
Arlington, Virginia 22201
Proceedings By:
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
(703) 352-0091
DR. MAYS: I think in your packet on the subcommittee should be the outline. What I would like to do is to just have some introductions because we do have some new people here.
I am Vickie Mays. I chair the Committee on Subpopulations.
MR. HITCHCOCK: I am Dale Hitchcock. I am a staff person on the subcommittee. We have got two other important staff people who aren't here today; Edna Paisano from the Indian Health Service and Audrey Burwell, who is with the Office of Minority Health. They are in Denver at a conference.
DR. NEWACHECK: I am Paul Newacheck. I am with the University of California at San Francisco and a member of the subcommittee.
DR. LENGERICH: Also a member, Gene Lengerich from Penn State University.
DR. CAIN: I am Virginia Cain from the Office of Behavioral and Social Sciences Research at NIH.
DR. WEED: Jim Weed, National Center for Health Statistics.
DR. FRIEDMAN: Dan Friedman, Massachusetts Department of Public Health.
MS. COLTIN: Kathryn Coltin, Harvard Pilgrim Health Care.
DR. EDINGER: Stan Edinger, AHRQ, and I work with the Quality Workgroup.
MS. CRUTE: I am Sherry Crute.
MS. JACKSON: Debbie Jackson, NCHS staff.
MS. WHITE: NCHS staff.
DR. MAYS: Welcome to those of you who are new. We will definitely welcome your input, your insight.
Let's just quickly get the big picture of what we have going in terms of the agenda and then our actual guest is here to talk with us. As you know, we are going to be continuing our hearings in terms of looking at the issue of the collection of data on race and ethnicity as it relates to -- in terms of the determination of health disparities in racial and ethnic groups.
Part of our next hearing, which is on Friday, is to look at the issue of data collection for Native Americans and Alaska Natives, the focus for that particular hearing. Our next hearing, which is scheduled for November the 8th, if I am correct -- the 7th and the 8th, but I think it is the 8th for sure, will be to actually look at the states' needs relative to the collection of this data. That is part of why Jim is here to talk with us today. We will follow that with talking about the measurement of health disparities, which we had a presentation last time.
I want to check in and find out where we are and what you want to do and bring you up to date on that. We will talk a little bit about the executive committee retreat in that it did focus quite a bit on population and population and health. So, I think this group should talk a little bit about that. That all should help us in terms of rethinking where we are with our work plans since our work plan seems to be a little bit out of date.
Then we will move to talking with Sherry about the report that I just gave you, which I -- I think I am going to actually change the order to talk with Sherry first and then the work plan, in fact, after that and then we will do some updates on the other things.
Is there any other agenda item that anybody else wants to put on our list today?
Jim Weed is here and this is not Jim's first time. We now know that you have actually done presentations several time prior to my chairship and during my chairship. We are thankful for your returning to talk with us again.
Part of what you are going to focus on today is really to help us think about what some of the issues are that we should be raising questions at our hearing with our state folks and introduce to us some of the efforts that are going on in terms of the collection of race and ethnicity data in the vital record.
Jim.
DR. WEED: I am happy to be here. I like to do this every once in awhile just to keep my portfolio up to date. Things change on a monthly basis almost. I was expecting to have power point here, but since I don't, I made, happily, copies of the slides. I made 20 and I think that will cover it fortunately. That is the nice thing about power point.
DR. MAYS: That is the nice thing about being prepared. We thank you.
DR. WEED: I haven't written anything out other than what is here. So, I will just talk from the little slides, like I was going to do anyway. I am deputy director of the Division of Vital Statistics and I have been assigned the division director, Mary Ann Friedman, to sort of be the division's liaison coordinator for race and ethnicity data within the division as a result of this OMB revised standard of 1997, which we all knew it was going to be quite a job to implement it and it is a job. It is rather complicated and particularly for many data systems that may have a little trouble getting racial identity to begin with.
It is even more difficult sometimes getting multiple race information. Anyway, we are making the right effort, I think. So, I will try to give you an overview of what we are up to, particularly at NCHS and in relationship to the states that provide us birth and death information.
Basically, we are in the midst of a re-engineering project to re-engineer all of the state vital statistics systems. I put down on the first slide or the second one what the characteristics of the current system are. Not every state is like this, but generally we have problems that are noted here. Many of the systems are fragile and inflexible, hard to change or update.
This was brought home to us when we tried to get the states to implement the multiple race question and found that just changing their electronic birth certificates might cost as much as a half million a state, looking at maybe 30, 40 million dollars just to change that item, which is sort of crazy and it wasn't too long -- also, we had revised our standard certificates for birth and death and getting the new certificates implemented turned out to be very expensive also, largely due to the lack of flexibility in the system, but more than that, most of the systems, particularly the electronic birth certificates are using old technology and many of them with stand-alone systems in hospitals, developed in the seventies or eighties. They tend to be rather subject to down time. There are many, many versions of the electronic birth certificate out there and states have each done their own thing pretty much. Many of those systems are still DOS-based and the death registration system is non-existent electronically, except for some prototypes in a few states.
It is still paper-based pretty much. So, what we are proposing is to re-engineer state vital statistics systems and the characteristics that are proposed for it in the third slide are that we would like to have comparable systems across all the states, reasonably comparable systems. They should be driven by national standards and guidelines. It would be very useful for them to use Internet technology in the transmission of data either from the source or to federal agencies.
It would be good for them to be iterated with other public health systems and they should support national security and privacy requirements. The re-engineering plan that we have been working on recommends that every state should re-engineer their system as soon as possible, whenever they can get the money to do it or whenever money can be found to do it, whatever might happen.
DR. MAYS: -- states bearing the whole burden of the --
DR. WEED: No. We would like to find money from any source that we could find that could be put to use for this purpose. The states are, of course, the first responsible for getting whatever money they can. NCHS has tried to get money. Social Security Administration, as I will be pointing out, is involved in this and they have been putting out a little money.
If we could find money from someplace else in some way shape or form, it would be also useful. The re-engineering plan should implement revised standard certificates of birth and death when the state re-engineers its system. In other words, we are not asking them to implement revised certificates until they re-engineer. The idea was that it doesn't make much sense to implement a new certificate in an old system. It doesn't make any sense at all.
So, the idea is re-engineer first and as you do it, implement the new certificates. So, basically that means we wait until they re-engineer.
DR. NEWACHECK: So, that could go for years and years.
DR. WEED: Okay. NCHS will operate a dual data collection system until all states reengineer and implement revised certificates. That means that we have got to be able to accept data from both the current certificate and the revised certificate. Obviously, when it comes to race, that is also true. We have to be able to accept a single race item from the states, as well as multiple race data from states and somehow put them together. There is not too much that we can do to get around that.
If we are going to tell the states to re-engineer as soon as they can, we have to be able to accept both kinds of data.
DR. CAIN: But you are not getting both kinds from one state.
DR. WEED: No, I wouldn't think so.
DR. NEWACHECK: Will the re-engineering permit a lot of cases in the future that will be easier to incorporate?
DR. WEED: They planned this to make the re-engineered systems as flexible as possible so that changes could be made with a lot less trouble and a lot less expense.
DR. MAYS: Let me make sure I understand because there is an OMB requirement of the data that is reported to the Federal Government. So, if a state has an old system and the state, for example, doesn't report ethnicity, do you then get the ethnicity or you are waiting until they re-engineer to make sure that you get like the newer reported.
DR. WEED: Well, when you say ethnicity, according to OMB you are talking Hispanic origin and we get that anyway. That is not being changed. Well, it is changing a little bit in the way we are collecting it, but it is Hispanic origin. Race is the thing that is the question.
MR. HITCHCOCK: OMB has little control over how they collect their data. OMB would probably encourage states to collect data that --
DR. MAYS: I thought any data that was reported to and used by the Federal Government from the states or from anyone that was collecting data on behalf of the Federal Government had to comply.
DR. WEED: No, I don't think that is the way it works.
DR. MAYS: This is even more of a shame then.
DR. WEED: No. This is a cooperative system with the states and we cannot force them or require them to do that.
DR. FRIEDMAN: Jim and I could give each other's speech. He has heard the tape and I have heard his.
One or two points. The model certificate is based upon the revised federal standard. The revised federal standard, my interpretation, has one standard for census and another standard, actually two standards for other than census. In the revised federal standards, NCHS births or births and deaths got what I saw -- he has heard my speech
-- got a free walk and the free walk was NCHS walks with census. NCHS does not walk with the other -- with the rest of the revised standards.
So, what we have is a situation, my interpretation is that basically we have got for some good reasons the birth and death certificates matching census. So, theoretically the numerators match the denominators, but theoretically other non-vital systems do not match vitals and census.
There are also some implications of that --
[Multiple discussions.]
DR. MAYS: I just wanted to make sure I understood that.
DR. LENGERICH: Those new standards don't have to be implemented until you change systems. Is that right?
DR. FRIEDMAN: For vitals, but that is not true of all of the other scores of data collection systems that any state will have and is eagerly looking to CDC for guidance.
DR. MAYS: We will bring Jim back now to his presentation. Sorry. We are getting ahead of you.
DR. WEED: Okay. Well, let me just continue what I was doing and we can get -- I think this will come back up in a minute. Even if Dan weren't here, it would come up.
I guess I am on the second page now. Yes.
The NAPHSIS-NCHS-SSA collaboration process, those three groups have formed a partnership that aims at collaborating on the re-engineering effort. NAPHSIS, is everybody familiar with that acronym? National Association for Public Health Statistics and Information Systems. It is a mouthful. It is the old association of vital statistics executives. It dates back, oh, probably to the thirties in various incarnations. It is currently -- it is the Association of State Vital Registration and Statistics Executives and Other Staff, who combine in this association -- the association represents them in various negotiations with the Federal Government, in particular with social security and within NCHS.
So, for example, our cost formula that we work out to pay the states money in our contract for vital statistics every year, the contractual amount for every state is determined by a formula that is worked out with NAPHSIS. It is not a union, but it gives us somebody to negotiate with, to talk to, to coordinate with and to work with, rather than 50 states.
So, it is a focal point. So, we have basically formed a partnership with NAPHSIS and SSA, Social Security, to do this. The goal is to develop a base model electronic vital statistics system that the states could adopt outright or adapted to their own needs. The system would include components to automate birth and death registration and record management activities, such as mending records, correcting them, certifying copies, doing interstate exchange, transmitting information to the federal agencies and so on and so on.
In the next slide, I have got NAPHSIS, NCHS and SSA are all full partners and participants in the project and they are all represented on the oversight committee that has now met once back in August and will be meeting rather regularly. There are going to be some committees of that oversight committee that will work on several components, including the electronic birth certificate, the electronic death registration system, business functions and so on.
They are going to begin with use case modeling, which basically says what is this new system supposed to do and let's put it down in black and white. Let's describe the system we want in a systematic way and after that you move to prototype programming. I suspect the work on this whole thing is going to take several years.
They are just starting in the oversight committee work. So, it is going to take a little while for all this to move on, but there is now a movement in this direction to really coordinate it, get Social Security into the act. They have been in the act, but now we are trying to get everybody together into a unified approach to the thing.
Okay. That is I think about all I want to say, unless you have questions about re-engineering because at your November 8th meeting, Delton Atkinson is going to speak, I understand. He is currently working with us as a contractor to basically be our staff person to work on the organization of all this.
I think he will give you a very nice overview of what the re-engineering effort entails, all the details, where we are going, time frames and everything. So, I won't encroach on his territory.
So, I thought maybe from here on out, unless you have questions that are burning through, I will go on and talk about race and ethnicity.
MR. HITCHCOCK: I just have one, Jim. You mentioned that there are a couple of states that are sort of out in front of this doing pilot projects or demonstrations or something.
DR. WEED: There have been several initiatives. I don't know if you are aware of it, but right now there is already out there an electronic birth certificate in every state and something like 90 to 93 percent of all birth records are done by electronic birth certificates. But the point is we have got to get those systems changed in the hospitals where they are done and we want to have them done with revised standard certificates.
Electronic death certificates, I think one of the early prototypes was done in New Hampshire by Chuck Cerk(?) about in the middle of the 1980s when he was registrar there. Pam Atkinson from New York has been hired by NAPHSIS to put together standards and a protocol for it. She has done that. There are some other states that have started to tackle the problem of doing it on their own. I won't go Internet the details but the concern that we have -- there are several things I guess I should point out.
One is we would like to avoid the situation that went on with birth certificates where every state contracted with a contractor to develop a certificate system for them. It was not coordinated and there are many systems out there, some, they differ in their properties and characteristics and their uses and all that sort of thing. We would like to avoid that. So, we are trying to unify and coordinate the whole re-engineering effort.
The other point is that if every state tries to do it by itself, it is going to cost a small fortune. You are looking at maybe several million dollars per state if they were each to do this. We hope that by doing a coordinated effort for a much smaller amount of money, we could create a prototype system anyway that could be the basis for a system -- it is probably still going to require adaptations to be put into practice in a state, but if we can do 70, 80 percent of what they need in this procedure, then I think we have got a big leg up. At least that is what NAPHSIS and everybody seems to agree to.
DR. NEWACHECK: On the death records, they are done by hand now basically or -- then who tabulates them, the states and then they send that to you or --
DR. WEED: Well, every state is sending us the demographic information on tape or in some -- I will do our bulletin board on the Internet or in some way communicating it. Many of the states are doing their own cause of death coding, using our electronic systems, MICAR, Super MICAR, which are basically input to ICD coding and coming up with the underlying cause of death. We do the coding for some states, a small number. Although we are trying to get them all into Super MICAR, which is basically a literal entry. You literally enter in what the doctor writes down on the certificate. It is entered in in the order in which it is written down so that the whole thing can be processed, but still there are rejects from the system that are not codable by the electronics. So, we have to handle the rejects and there is something like 10 percent. That is still 200,000 by hand.
But some states are doing their own rejects. They have to have a nosologist to do that. We keep refining the thing and as you probably know, we implemented ICD-10 in 1999. So, the entire system had to be redone and we have been working on that for the last seven or eight years.
MR. HITCHCOCK: Just one question. In the old NCHS -- where is the coding done these days?
DR. WEED: Any coding we do is done either in North Carolina at our office down there or we have got about -- at HHS, it is about 120 people down there right there. In our division we have about 55 people plus about 30 contractors. The contractors are mostly revising ACME MICAR, Super MICAR, but there are a few -- we also contract out to private contractors to do the coding for us, some small amount.
DR. MAYS: Is there any sense that as you lean towards this electronic system both for the birth and the death that you are going to for racial and ethnical groups cut down the amount of missed classification? Is there something built into this as you move to this new system for quality checks to see if a person says they come from two parents and the two parents are white and they say they are black, can you say is there a problem?
Is there any --
DR. WEED: You are dealing with a moving target on race all the time. People will give you today one race and in a month they will tell you something else. The census is already finding that with the quality survey. They will answer whatever seems to fit the occasion. They won't say the same thing always later. Sometimes --
DR. MAYS: The problem we have is like say, for instance, with Native Americans, there is a tendency to be
-- death certificate. So, what I am trying to figure out is that if, at least, you have it at one point and it says that is there any reason to think that you will flag it or are we going to assume that it is people that are -- who have given this to you. The person themselves gave it to you maybe at birth and then at death, it was the funeral director who gave it to you and do we have any way since we are going to go to an electronic system of someone flagging those things to do something --
DR. WEED: The electronic system doesn't solve the kind of problem that you are talking about, I don't think. The funeral director still is going to be a key element in this whole thing and it is going to require some training in the states to get funeral directors to actually elicit or make sure that people have an opportunity to report more than one race.
DR. FRIEDMAN: A couple things here, one of which is I completely agree that the issue with the birth certificate in which parents are generally asked is completely different than with the death certificate in which funeral directors typically will not ask a son or a daughter or a father or mother what the race is of the person who has died. Generally, it is purely by guesswork on the part of the funeral director of looking at a child or an adult child.
Secondly, for at least -- I can't speak for any other systems here. For us, the train has left the station. The horse has left the barn if you are writing code in our state. And I doubt that we are the only state that is doing that.
We can't stop because of our funding stream or we will lose our funding stream. But we are trying to build a person-based system not a, quote, certificate-based system. So, what we would be doing is having not a system for birth certificates, a system for death certificates, but a system that has an entering for me, for Vickie, for Kathy, et cetera, et cetera, a person-based system. While we would not go back and change what the funeral director says, at least we will have some basis for making evident what is -- you know, the joke is you are born Hispanic and you die white. At least we would be able to have that essentially as a data element, race at different times, race at marriage, race at birth, race at death, et cetera.
MR. HITCHCOCK: Do you have a unique identifier for --
DR. FRIEDMAN: I am not sure how we are going to do it yet. We have legal issues with --
[Multiple discussions.]
We can only capture people who were residents or who were born or dying in the state. So, we are going to have quite a bit of --
DR. NEWACHECK: It seems like it would make a lot of sense if this could be national.
[Multiple discussions.]
DR. MAYS: We can have Massachusetts at the hearings. We will make sure --
DR. WEED: That is an interesting -- I hadn't heard that before. So, that is an interesting thought, interesting approach. Something is changing every month. I can see the problem of people moving into the state or leaving the state and doing stuff.
I guess I wanted to mention one thing. Along with the revised certificates this time, we have made a real effort to try to improve data collection by creating work sheets, model work sheets that would elicit the data for the birth certificate and that would be given to the mother to fill out in the hospital. Those work sheets have been researched rather thoroughly. We have gone to quite a bit of trouble to actually try them out in hospitals and give them a workover with cognitive interviewing and all kinds of stuff. Those work sheets will hopefully elicit multiple race on the birth records, but probably not much different from the way the census did it in 2000. I mean, we still are relying on some person to fill out information.
Mothers will probably fill it out for the fathers, I would imagine. Of course we don't ask for race of the baby. That is not on the certificates, but we do ask for race of the mother. So, I don't know. Is Massachusetts going to ask for race of the baby? I didn't think so.
DR. MAYS: Wait, wait, wait. Could I ask the question why nobody does that? You ask for the race of the mother and the race of the father and then how do you determine the race of the baby?
DR. WEED: Well, the way we do it in our publications is we publish all the information by race of mother. Most of the information on the birth certificate relates to the mother. So, it is the mother's race that seems relevant. The race of the baby is not there because
-- well, only one state has race of the infant on its certificate. That is Washington State.
I don't know why they did it. They did it. They code it. But it is generally probably not popular because
-- I am sort of guessing I think here -- it doesn't seem quite ethical in a sense to specify for somebody else what their race is and then it is on their record forever. The baby should be allowed to specify his or her own race.
DR. FRIEDMAN: There used to be an algorithm -- NCHS used to have an algorithm that had mother -- you know, mother, father and then had all the different race and ethnicities. So, some race and ethnicities would essentially trump others. So, for example, if you were Native Hawaiian, that trumped everything else. So, if your father was Native Hawaiian and your mother was whatever, Samoan, the baby -- because the matrix was to determine -- the baby then became ipso facto Native Hawaiian, et cetera, et cetera.
DR. WEED: Yes, and if the mother was white and the father was black, the baby became black. But remember now, this is not on the certificate. This is only for tabulation purposes that we did that. It didn't go on the certificates.
This goes back 30 years --
DR. MAYS: We are in such a different age now that in terms of what --
PARTICIPANT: That is why we don't do it anymore.
DR. MAYS: -- in terms of what we think about what is important as data points, it is kind of like how the paper is raised in terms of their race and ethnicity classification is important. So, if we wait until the baby is like -- at what stage -- I guess when you are 18 and you fill out your own materials at that point and they get to say what they are, but yet throughout the baby's life, the parents at schools and all this other stuff decide to say what the baby is.
Let's just say it is a mixed race baby and the parents say the baby -- and if the mother is white and the father is black, then in the birth certificate you see the mother being white. The kid is being raised as black. We don't know anything about that, I guess, then until the kid is 18, right?
MS. COLTIN: -- point though. Now I thought in Massachusetts that the immunization registry was populated through the birth certificate and so does that mean that there would be no race and ethnicity data in the -- registry if there is not race, ethnicity --
DR. FRIEDMAN: If there were, it would be maternal. I agree. I agree. I mean, my kids are ipso facto white. Now, you look at my kids, they are not white kids. There is no question. They are not. But according to the birth certificate, they are white.
DR. WEED: You don't have race of the child on the birth certificate.
DR. FRIEDMAN: On the statistical -- no. According to the tabulations that we -- you are right -- the tabulations that we would do for a low birth weight for infant mortality.
DR. MAYS: I think these are important issues for us to think about. I think we need to -- after the hearing, either build some of these as questions into the hearing and then after the hearing, I think, without having people needing to talk with us, we as a subcommittee I think should talk about these issues and make some -- and give some thought to what we would want to do in terms of recommendations. I think maybe we should take a look at what you are proposing for Massachusetts and think about -- we can make recommendations but think about that and think about -- you know, this is what you do. The question of what we are beginning to understand of the relationship between race and ethnicity to both health status as well as how one is treated in a health care encounter may raise the question more as to whether in some place we need to capture something.
I mean, I am not saying we recommend we mess with, you know, the birth certificate, but maybe that we need to think about at some tabulation point of having such information available.
DR. WEED: I think we have something of a solution to the problem already. When and if we are obtaining multiple race information from mothers and fathers from birth records, you have got the race of the baby. It is all the races of the mothers and fathers combined. You don't need anything more.
DR. MAYS: No, it was different.
DR. WEED: Remember now, the multiple race is the name of the game. If you have parents with more than -- that are of different races, that child theoretically has both those races, if there are two of them. If they are different, why doesn't the child have them both? It is multiple race. The child can have multiple race, too.
DR. MAYS: I would still want to know what the parents are determining that they are going to tell the child, I think, because you could have a child that comes with two races --
DR. WEED: Sure. That is another matter.
DR. MAYS: -- determined what -- when that child goes to school and they said what that kid is and what the kid says to the other kids that they are and the kids says it all -- and the parents at every encounter, that is what then I think, at least from a kind of social construction point of view really is going to impact the kids access and a lot of other things that may --
DR. LENGERICH: That concept would also apply to what is happening at birth. So, how do the mother's race and the father's race get recorded? Is that the mother responding for the father as well?
DR. WEED: In many cases.
DR. LENGERICH: So, we have got that perception happening there as well.
DR. WEED: I think it is not untypical at all for the mother to fill out the form and she puts the mother's race down. In fact, there are cases where the father is not stipulated, you know, unmarried mother and the father's race may not be recorded.
DR. NEWACHECK: What do you do for the child then?
DR. WEED: Pardon?
DR. NEWACHECK: What would you do for the child then?
DR. WEED: Well, you may not have a complete race for the child. If you are really following that pattern of using combined races, but I don't foresee us very likely using birth records to assign a race to the child. For research purposes, I could see that you can -- there are various ways you could go at it. We would probably hope to be able to provide multiple race information for mothers and fathers on our multiple -- on our birth record data files.
Confidentiality will prevail, of course. So, I don't know how much detail can be provided, but that would be, obviously, a whole new area of research with multiple race information. This is an area nobody knows anything about. It is totally new in terms of research. We can't find hardly anything that we can use to say what to expect about people who are more than one race. It doesn't exist.
DR. MAYS: We tried to recruit -- David Harris, I think is his name. He is a sociologist at the University of Michigan -- has actually been looking at this issue of multiple race and ethnicity, multiple race and the implications of it in terms of when they reach adolescence.
Again, I kind of only have heard one talk of his.
DR. WEED: There just aren't very many data sets out there on this.
DR. MAYS: Let us make sure because we are throwing a lot of questions at you that if you have some very specific things for the rest of the presentation that you wanted to tell us, that we let you do that.
DR. WEED: I can cover as much of the presentation as you want. I don't have to cover it all. It sort of stands by itself if you prefer. Actually, the next two slides show what has been proposed or what is contained on the birth and death records for Hispanic origin and race and the slide that follows those two --
DR. MAYS: Could I just ask, is this the work sheet that is given to the parent?
DR. WEED: No, this is the actual parents on the birth record and death record. The work sheet would look very similar.
DR. NEWACHECK: So, these two slides, the funeral director would fill these out? Is that correct?
DR. WEED: Well, actually it depends. Funeral directors sometimes ask the informant to fill it out, but -- I don't know. I think it is probably more common for a funeral director to fill it out and hand the certificate to the informant and say does this look okay. I don't know. In a situation like that -- maybe Dan has got more experience with working with them than I have.
DR. FRIEDMAN: My experience was from being an informant.
DR. WEED: They know what the race is. They don't have to ask.
DR. FRIEDMAN: They assume they know.
[Multiple discussions.]
I think one of the reasons why we have such a low Hispanic mortality rate, one of the reasons, is because of misclassification. Let me just say all states are not going to be using this schema.
DR. WEED: Actually, the next slide that I have on the --
[Multiple discussions.]
I will just say this. The next slide shows what was on the decennial census of 2000 and if you compare it to the proposed race question for birth and death records, you will see that the same check boxes are there basically. So, you will see that we have, indeed, mirrored the decennial census format.
Okay. Dan, go ahead.
DR. FRIEDMAN: I think there are two reasons. One, some states, like California, has, I think, a literal effects entry for race less ethnicity. So, they just say hey, what is your race and then they copy it down. Then there is coding where they basically look up tables. Some states, like us, with all due respect to the U.S. Bureau of the Census, we just think these are not helpful questions. Certainly in -- again, I can't speak for any other states, but in Massachusetts, we have people of many ethnicities beyond Mexican, Puerto Rican and Cuban and also, you know, the data that are asked, it is sort of valuable real estate on forms. We don't have a lot of Samoans or Chimorans(?) in Massachusetts. I am sure there are some, but we don't have a lot of them.
We are not going to take up real estate asking about that. Then, in addition, you know, we don't regard Chinese, Filipino, Japanese, Korean, Indonese as races. We think of them as ethnicities. So, we just basically we have a somewhat different schema and when we report it to Jim, we will make it conform to those -- you know, whatever NCHS wants.
DR. CAIN: So, what do you ask for Hispanic origin, for example?
DR. FRIEDMAN: We are asking individual ethnicities.
[Multiple discussions.]
DR. NEWACHECK: So, you can lump them up and then compare --
DR. FRIEDMAN: We separate out the revised federal -- you know, whatever it is, five races from ethnicities as two separate data items.
Jim has probably spent weeks talking to us about this and, you know --
[Multiple discussions.]
DR. WEED: It has changed over time. Actually, we had sort of reached an agreement to disagree in a way because the last format that I saw, which I am not sure is still the right one anymore, but they had a format that basically was an Hispanic question -- if I recall correctly, there was an Hispanic question. Then there was an ethnicity question and then there was a race question. That race question had the five basic race groups of the revised OMB standard.
That is correct. Those are the races. There are according to OMB five race groups. Chinese is not a race. It is a subgroup of Asian. Okay. So, what he is saying makes sense if you look at it this way. I am not disagreeing with him. Five race groups are required by OMB to be collected and he is collecting those. But he has another question called ethnicity. At least he did have. I don't know if they have still got it. But ethnicity had something like 15 to 20 check boxes and each check box reflected some ethnic group in the state of Massachusetts of interest to them. Capernian(?) may be one. Nong(?) might be one. I don't know what all of them are. But those ethnic groups included Chinese, Japanese, Korean, those.
So, in effect what he has done is he has just simply broken out a lot of what we call race items and put them into another question called ethnicity, which, okay, if somebody is Chinese, they will give us that check box as Chinese and if they then go to the race side and check Asian, that is okay, too. We will make those people Chinese.
Now, where it is going to get interesting is if somebody checks Chinese in the ethnicity box but not Asian in the race box. So, in other words, it is sort of a contradiction in a way. It is not redundant. It is just not -- well, you get the idea.
So, that the edit that we have created to handle these 15 check boxes and the write-ins is designed so that it will handle the data he is producing. I don't care how he does it. If he has those check boxes, we can handle it with our edit because we can literally assign data that he sends to one or more of our race items and process it. So, we can use what he is creating and I don't think he is going to be out of line, at least the last version or two I have seen looked like to me we can both have our cake and eat it too, I think.
That is why we are smiling at each other.
DR. MAYS: As you say, you have agreed to disagree.
DR. WEED: You have to understand that NCHS has to abide by the OMB regulation and that says there are five race groups, but you can have more if you want to. You can get more detailed if you want to. Ethnicity is Hispanic origin or not Hispanic. That is what ethnicity is to OMB. We have to follow OMB.
He is in a state that doesn't have to follow OMB, but it is in his own interest and in our interest to collect data so that they can be combined with other states to make national numbers. I think they are doing a balancing act of what they want, plus what we need.
MR. HITCHCOCK: Accommodations are made here. You are accommodating --
DR. WEED: We can't dictate but they do. It is a federal system and of course it is in every state's interest to produce data that are comparable to other states so that they can compare themselves with them and with the national average. Without that, you sort of don't know where you are. That is, I think, the glue that holds the whole thing together frankly.
DR. MAYS: I think that one of the things we may want to think about is to ask if we can have copies of these forms just so we could get a sense of how they are doing what they do and the range of permutations that exist.
MR. HITCHCOCK: The re-engineering plans and --
DR. MAYS: Yes, I think we should do that.
DR. LENGERICH: Will that be represented by Delton Atkinson then or --
DR. MAYS: I think we should ask -- no, we are having individual states to come and I think we should ask them because I think, again, you can see the difference in the sense that they all get to this same point of meeting the OMB, but the states -- if we want to understand what the states data collection needs are and also because there are some states the really need some help, I think it would be useful for us if we make recommendations and talk about the range of possibilities of what can be done or is being done so that other states who are just starting to re-engineer can consider that.
They will consider the cost to them also in the sense of, you know, eventually having to make it fit into the federal request, but at the same time, there is a rationale by a state as to why -- I know in California there is as to why we collect what we collect that helps them to be able to serve the community, the subgroups.
DR. LENGERICH: Then the selection of those states will become critical to represent the needs, as well as the resources available to them to do what they are going to do.
DR. MAYS: When we get to that, we will talk about which of those states have been proposed before.
Okay. I am going to bring this to a close soon but I want to make sure if there are other questions from the group that they -- I mean, are there other specific parts of your presentation that -- we just keep having all these questions for you. You will have to keep coming back and visiting. Maybe we will get it -- we just keeping asking you lots of questions because the landscape keeps changing.
Are there other parts of the presentation you want to make sure that you talk with us about?
DR. WEED: We have sort of gone through the first half of what I was going to talk about. The second half, the remaining slides, focused primarily on the bridging problem, the problem of the fact that the decennial census already collected multiple race and beat us to it. We are still getting single race and the result is -- from the states -- and the result is that there is lack of comparability between our numerators and denominators.
Very shortly, I think, within another week or two or three, there should be a MARS file available to the public based on the April census 2000. That MARS file, I understand, should be at the state, county, national levels. It will have for every county 31 race groups. That states are really trying to figure out -- we are, too -- what to do with 31 race groups. What is white anymore. What is black anymore. What do these mean.
When you deal with Asian, what are you going to do with the people that are both white and Asian. How many of them are there? Well, I got the information in one of the slides here that refers to that.
DR. MAYS: Which one?
DR. WEED: It shows the complexity of the problem. If you want to dodge right down to Table 1, that shows the nature of the problem. Table 1 shows for the decennial census of 2000, the first column is the number of people of each race group, who reported that race alone. That is all they reported. The second column is the number of people who reported that specified race either alone or in combination with some other race or races.
The third column is the increase that you get between the reported alone and the alone or in combination. And the final column is the percent increase, percent increase over the reported alone number. You will notice that the American Indian, Alaska Native group, the AIAN, you have got a 66 percent increase. The actual number of American Indians, therefore, in the United States, is someplace between 2.475 million and 4.1 million, someplace in here.
If you hadn't asked multiple race, the number of Indians probably would be bracketed in there, but where? And it makes a big difference you see, what you do for your race. Do you take the smaller number, the larger number and you have to think about your numerator. I mean, we don't have multiple race on birth and death records. So, the number of indians that we get is, who knows.
Finally, the really rough one, of course, is the
-- and other Pacific Islanders, got anything from 400,000 to 874,000 of NHOPIs in the United States, 119 percent increase. So, we recognize very early on that even though the number of multiple race people is small, it is only 6 million, yet they make a big difference in the three smallest race groups. They even make a 5 percent difference for the black and African American group, which is enough to kind of knock your birth and death rates around.
So, we recognized early on and so as OMB that a bridging was going to be needed. So, what we have done is we have done some research to create an algorithm and the algorithm basically allows -- it specifies how multiple race people should be allocated to a single race. We have given that algorithm to the Census Bureau and they have applied it to the MARS file of April 2000 and I believe they will shortly deliver to us a bridged MARS file. Expect to get it in the next couple of weeks, I hope, for the national level only, not state or county.
That file will give us, I think, something that we can use to calculate birth and death rates with. Otherwise, you don't have much that you can do with race.
DR. MAYS: -- the question that data is going to be released so that -- I am just going to clear up this one level to say so that any researcher can get it with a 31 rate. Is there guidance or caution or guidelines about how to use --
DR. WEED: You mean from the census standpoint, the decennial census MARS file you are talking about. By the way, MARS stands for modified and race sex. A little fine point here. The decennial census itself has a question as a category as we do in the vital records, some other race. You can check that and write in something else, like you can write it a religion, Buddhist. You could put down something that is not there, like -- it could be anything. People write down anything for race. Anything under the sun can be put down as race.
I have seen zorro asprene(?) as a race. Any religion, Christian, Buddhist, Moslem, you name it. Anything can be put down for race. What we have to do is have to code things like that to something and the census does it, too. They get all this sort of thing also. But they have some other race.
If you look at this Table 1 in my handout, there are 15 million some other race people that reported by itself. Now, when the MARS file was created, those 15 million disappeared. They converted some other race to one of the basic race groups, through an allocation process. Now, as it happens, 97 percent of those 15 million are Hispanics. So, what has happened is they have probably written on the race side and they have written in Mexican, Puerto Rican, something.
What happens in the census is it goes through an allocation process, gets computed and most of them go to white. That is why the 211 million white population is about 15 million short of what the projected number was for 2000. That will be made up by those some other race people in bar spot.
DR. FRIEDMAN: Just a couple of points. One specific to the some other face, which is, you know, again, just as a point of information, the revised federal standard, the other part of it, not the census/vitals part, the other part. So, if you are -- if there is an other federal data collection system, some other race is not an allowable entry for race. You can be multiple race or you can be one of the five races, but you can't be other.
Then generally I just want to quickly pick up on something that Jim mentioned a few minutes ago. I think that one thing that, you know, with all of the attention focused upon, you know, just the specifics of the revised federal standards. I think one thing that at least I have seen less attention focused on is a research agenda for looking at the health implications of being multiple race. I have seen a few things in terms of infant outcomes in terms of stress for children and, you know, the few things I have seen that seem to indicate that, well, there are some real health implications and, you know, far from trivial health implications.
I think at some point, Vickie, that might be a good thing for the subcommittee to take some testimony on because maybe starting a research agenda for that, I think, would be something that people would appreciate.
DR. WEED: Actually, Jackie Lucas, who is a staffer for this subcommittee, has been doing some research on that herself with the Health Interview Survey and can talk to you about -- she knows all the research. Just get her to give you a little presentation.
DR. NEWACHECK: I had a couple questions about the bridging part of this. Is the method you are talking about with bridging, is that is what is used in the NHIS now because they do report it both ways on their data sets?
DR. WEED: Yes, it is all in here.
DR. NEWACHECK: Then in this algorithm in your slide it says that you assign people to the race that you hypothesize the individual would be most likely to have reported if they only chose one race. Does that mean that there are a set of decision rules that are based on your empirical analysis data that says that they pick this and that, that they would have picked this based on --
DR. WEED: No. I didn't get into the details too much. It is here in a sense. We use multi-nomial logistic regression on an individual case basis in the Health Interview Survey, combining four years of data. There is a regression for each race group. Black/white, for example, there is a regression for that.
What it does basically is it has independent variables that are characteristics of age, race, sex, Hispanic origin -- not race -- age, sex, Hispanic origin and characteristics of the county in which the HIS respondent lived. We use a percent of the county that is the particular minority race in a concentration level and then you throw all these people into a regression equation where the independent variable -- the dependent variable is the main race. What you are able to do then is get an estimate of the probability or proportion of the population in that county at that age and that sex, Hispanic origin, that would have reported a single race if that was all they did and used the main race in the Health Interview Survey to do that.
So, you take all the black/white people in HIS and use the main race they report as the dependent variable. So, you get probabilities, all kinds of probabilities. You get percentages or whatever you want to call them. So, the algorithm is a massive file of percentages.
DR. NEWACHECK: That is different from the old style method that Dan was talking about earlier, where you had a set of decisions that --
DR. WEED: That was very, very simple-minded.
[Multiple discussions.]
They don't have a data set to do it with. They are going to apply the percentages that -- they have already done it. We have given them the percentages. We have done the regression. I think there are about seven or eight regressions. We have given them to them last April or May. They have already applied it to the census and to the 2000 July 1 estimates and the July 1, 2001 estimates
At the county level what they do is they apply it at the aggregated -- to aggregated numbers. So, if in a particular county at age, sex, Hispanic group, you have got three black/white people, the algorithm applies to three and what you get then is like it might be for that county that you could -- black/white in the ratio of 2 to 1, the black or white. Two of them go to black, one was the white and of course, that is two-thirds and one-third and you multiply those times three, you would get even numbers. But if you had five people and you had crazy fractions, you are going to get lots of -- it just produces a whole bunch of fractions that have to be added up. You can envisage how complicated it is.
Then, of course, the census has a rounding algorithm that they employ to round all those fractions. It is complicated.
The alternatives are less attractive, I think. Things like assigning everybody who is black/white automatically to the smaller group, black, or assigning everybody who is Indian/black to the smaller group, which would be Indian. There are rules like that. There are all kinds of allocation methods that are out there. We didn't like any of them. We felt that the Health Interview Survey data was much more justifiable.
Now, mind you, I should make sure that you understand that the Census Bureau is not releasing these as official bureau data. They are going to be basically NCHS data. They are doing it for us as an arrangement to us at our request. They are not in the business of releasing official Census Bureau bridged population numbers. They are doing it for us and we will give the data -- make it available to the states and other people. The National Cancer Institute is very interested in it.
DR. NEWACHECK: So, in their time series that they publish, they will just have the new category. They will not try to do any bridging?
DR. WEED: That is the current policy and I don't think I see any chance that it is going to change. The only thing about the OMB standard that is really specific to NCHS or the census, as Dan was referring to, is that business about other race. Census was given permission to use some other race in the collection and they did so. It was pretty much done because they had done it in 1990 and 1980 and so on. Then we argued that, well, vital statistics should do the same because, you know, it is -- the system should be compatible, should be collected in the same way. So, OMB said okay, vital statistics can do it, too. But I don't think they have given it permission generally to all NCHS systems so far.
That was done because of the interlocking relationship between vital statistics and population data. But other than that, what the OMB standard says is basically not unique to census in any way shape or form.
DR. MAYS: I think this has been really useful to us as we head into our state hearing.
Before I thank Jim, are there any last questions?
DR. WEED: You can read, I think, enough of what I have written here so that I don't have to read it to you.
DR. MAYS: We focused planning on being at -- at a couple of the hearings?
DR. WEED: I am not. Delton sits right next to me in my office and he will do a good job.
DR. MAYS: Thank you very much. I really appreciate your putting together this presentation and sharing as much as you have with us. It has been very useful.
MR. HITCHCOCK: Our meetings are open, Jim, and you can always send somebody from your group to sit in. It might really be a good idea.
DR. WEED: Well, I am happy to keep you updated. If we get these bridged numbers in the next couple of weeks, we are going to probably be employing them to revise our final reports for 2000 and 2001. There will be more stuff coming out as it moves on.
It will be interesting to watch Massachusetts, too.
DR. MAYS: I will put a bid in to say watch California, but I don't know if we would match Dan at all.
DR. WEED: Very creative thinkers in that state.
DR. MAYS: Thank you.
All right. Let's move on to our next agenda item, which is really kind of a follow-up of a presentation that we had at our last breakout in which we had Jeff Percy talk to us and Victor Shornbach(?), about the measuring health disparities. You received from me his last version of that particular document with a request to comment. Where we are is -- and I said what I would do is pull those comments together.
I had one person ask me some questions, but we really don't a lot of comments back, but it turned out that they decided to put it on hold doing any further work on the document until after there was a meeting of a group that was being pulled together by -- called the Interagency Work Group on Managing Health Disparities, which met in mid-August to deliberate further on the acceptance of this disparity index because they wanted to talk -- I think what has happened is they wanted -- more people wanted to actually talk about it and talk about its implications, think again about its implications.
I have some notes here because Audrey Burwell, who is one of the co-lead staff actually attended that meeting, which was in mid-August. She said that their focus of the meeting was to decide on a reference population for developing statistics. Extensive discussions took place on whether to use the best group or total population as a reference point.
The group -- that many pros and cons regarding the use of those terms and it was decided that inpatient staff would need to develop information for the use of both terms that would meet the following criteria; reliability, stability, ease of interpretation and continuity across all measures, time, domains and indicators and finally that of precision, meaning statistics and standard error.
NCHS staff further explained the development of the disparity index. This index was developed from a Department of Education model that looks at the expenditures per student, per school district. The disparity index would not measure disparities and population size jointly.
This is a summary measure of disparity that will display the mean of absolute differences of an indicator. It averages the differences relative to a reference point. The group wanted to review research literature to determine if there was a better model to -- agree to do the disparity index.
Finally it says the Healthy People 2010 Steering Committee is meeting this afternoon. This is why Audrey is not here because she is actually there. I just got this today, which is why you don't have it -- is meeting this afternoon to receive an update on the work group deliberations. So, Audrey will actually bring us up to date.
So, I think the bottom line is that right now we are not the only group who is commenting, I guess, on that disparity index that they wanted to use and raising a lot of questions about it. So, unless you have like put in hours of work or have a document that you received, my understanding is that we probably should wait and see what is going to happen in terms of the interagency recommendations and then Ken Kappel(?), I think, is going to go back to the document and make whatever changes based on the feedback that he is getting and then send that out again.
Unless you feel the need to go ahead and comment, which I would say we probably shouldn't do since things seem to be in change with, you know -- that they are being asked to refine the definitions, that we should just hold off at this point, but stay very involved in keeping an eye out on the direction they are taking.
DR. NEWACHECK: [Comment off microphone.]
DR. MAYS: The presentation from -- that was at -- what was the meeting? June. No.
[Multiple discussions.]
PARTICIPANT: It was the one at the Windham Center.
DR. MAYS: It was in the June meeting.
DR. NEWACHECK: That clarified the other question.
DR. MAYS: Okay. Did you get --
DR. FRIEDMAN: I don't remember getting it.
DR. MAYS: Yes, I sent it out by e-mail with a request for comments. Jeff sent it to me right after the meeting. So then it would have been in probably end of June, beginning of July. I think I sent it out in the beginning of July before I left town and then asked for comments I think once I came back from out of town and then right after that, got this e-mail of this meeting.
So, does anybody disagree with that? I just wanted to say where we were on that.
Okay. Let's talk a little bit about the Executive Subcommittee retreat. The reason I put that on the agenda is because it really was a discussion about population-based health, kind of, some of the work we are doing, which is on populations. I am going to ask Dan to actually -- I shocked him a little bit, but -- the reason I am doing this is because I am going to ask Dan to talk just a little bit more about what he sees as the way in which we can approach the wider committee having a sense of what population health is about. I think once we do that, we were going to move into some re-engineering to some extent of trying to ensure that population health is kept as a focus.
Given that Dan was the person that was assigned to pull some panels together, I just want him to share that and then what I want to talk about is where we are with population and population health for the Subcommittee on Populations.
Dan.
DR. FRIEDMAN: Thanks, Vickie.
There was a discussion at the meeting about, well, exactly what Vickie said, sort of the difference between populations and population. Some of it is -- my feeling and I will look at Kathy and Paul in particular -- sort of the derivation of the work for the last several years of the Subcommittee on Populations has been essentially not -- well, has focused on subpopulations issues rather than population health more generally.
Some of the discussion at the Executive Subcommittee centered around should we and assuming we should, how we should try to have more of a generic focus for the entire committee using a population health framework. Part of that discussion was of the desirability of trying to relate some of the more specific issues, standards, privacies and so forth to the broader population health construct.
So, one of the ideas that came out of that was trying to have some panels on an ongoing basis of full committee panels that would try to bring people -- try to introduce a common vocabulary in a common set of concepts around population health, knowing that it is extremely unlikely that we would ever have full committee agreement on even some very, very basic parts of a population health framework.
So, one thing that we discussed in terms of these panels would be, for example, starting with a panel on different models of the determinants of population health. Perhaps having one of those folks come and discuss it, one of the things -- another possibility that I discussed this morning and -- I was trying to bring somebody in -- other Canadians. A lot of this has occurred in Canada, possibly John Millar from Canadians for Health Information talking about the impact of models of population health on Canadian health policy. Maybe even bringing in somebody from World Health Organization, who has worked with the World Health Organization, such as Ivana Kickbush(?), who used to be with WHO and is now at Yale or Harvard to talk about the impact on WHO policies.
Then having another panel focusing on some of the empirical work that has gone on. There are a lot of issues that are sometimes implicit, sometimes explicit in these population health discussions that are not always -- well, that are far from -- to which the answers are far from concentric. So, for example, one of the issues that gets repeatedly brought up in Canada and particularly by Les Rouse(?) at the University of Manitoba is the need to focus on the extent to which health care affects health.
Another issue that, you know, I think particularly among clinicians that generates some controversy is, in fact, whether or not population health is just individual health writ large and whether the same things that affect population health affect individual health. We have had bits and pieces of some of these discussions occasionally at the full committee and the reactions have been varied.
Those are the thoughts. Starting out with one or two panels, but spread across several full committee meetings.
DR. MAYS: The goal, as I understand it, and why I think that the committee needs to think about where it wants to go is, as Dan said, to kind of raise the consciousness of the entire full committee about population health issues and to ensure that as we look at something like I guess -- that we are thinking about the population problem.
At the same time, there are pieces of the population health and population agenda that we are also talking about whether we want to disseminate in other places. If I understood that correctly, it would be that NHII was talking about, for example, beginning to focus on population health specifically and then it would be putting on the table the issue of the continuation of the 21st Century also continuing and then this group also continuing. So, I think to some extent, we have to kind of figure out, you know, kind of what the matrix is and how to work at the level of the full committee, as smaller committees and what the organizational piece might be.
I think one thing this committee needs to think about is whether or not it becomes populations, which I don't think is necessarily where we want to go, but we have to think about how to balance all these things out.
Kathryn, you were at the meeting, Dale and -- so, your sense of kind of the matrix and there was a big piece in terms of quality coming in, too, and ensuring that with quality there was a notion of population also. I can't remember if it was populations or population.
MS. COLTIN: It is a little bit of each and I think sometimes these are difficult distinctions. I mean, clearly, I think the discussion of the Executive Committee in terms of what was recommended for the full committee is as Dan outlined it. It was population health, singular, and models of population determinants, of population health, all these things so that we would have kind of a broad construct of a model in which to say here is where what we are doing on patient medical record standards might tie into this understanding this model better. What data might be needed as part of a standard medical record that would help to illuminate these kinds of relationships?
So that if we started playing around in different sandboxes with our different subgroups, we would have some common framework in which to say our work relates here, here and here, you know, under these different models. I think that was really what the Executive Committee was trying to get at, kind of an umbrella that everyone would have some common understanding and framework that was both population-based framework and then that we wouldn't see our work as so stovepipe the way I think we currently do. Is that fair?
DR. NEWACHECK: It seems like a real laudable goal. It is hard for me to see, though, how -- I mean I think it is a great idea even if it doesn't work because -- but it is hard for me to see how having these discussions of models and having -- talk about their models and how that is going to influence the discussions about code sets and, you know, thinks like that within the committee and standards and that.
It stretched me, but I think it is a great idea and we ought to try it and see what happens.
MS. COLTIN: Well, you know, if it happens successfully and everybody at least gains a common understanding of the model, it may make it a little bit easier when we are making recommendations for something in this committee, that another -- in this subcommittee, then another subcommittee should do something, that we tie it back --
DR. MAYS: A value that I think will happen from this in terms of these presentations is to bring a public health perspective as opposed to either medicine or just statistics, but a public health perspective. So, I think that is going to be the most successful part, but for me sitting here with this committee, I am trying to figure out, well, then, how does this relate to this and this and this and this and I don't have a clue kind of thing.
Then I also say, okay, if we also now get back to the population, in the field there is a big debate by some populations that those models have really been examined relative to their particular population issues. So, I also at the same time sit and want to say now don't make those the ultimate models, but to think of them as guideposts that you think, you know, within those terms and that you may have to then step back and kind of specialize them sometimes.
But I think that the issue of having a common public health perspective is worth it. If nothing else happens, that is worth it. So, I am not opposed to it at all. I am more on the kind of management organizational side of -- I don't know how all this stuff is going to go. Already we have problems when I come up with an issue of seeing if I can get, you know, standards to like consider it or standards wants us to consider something and we never have time. So, I agree.
So, it is almost like a -- it will be a great academic exercise that -- all opposed to. I just wanted kind of for us to think about the whole picture of that and to be involved in it because I think it is important -- if any other committee can make the link between the population and populations, it probably will be us. So, at any point in there where we can do that, I think it is going to be a service. So, that is why I thought we should put it on the agenda to discuss a bit.
DR. LENGERICH: I agree with you completely that just raising it at the committee, full committee, level would be worthwhile and educational for all of us. I guess as a result of some of the conversation and what we heard this morning, I just think that if in the panels that come through subcommittees, I guess I was wanting to hear about, well, what is the expected impact upon this on the health of the people they are serving, on -- or DOD.
I mean, we can get really caught up in the standards or the systems or whatever and it should be resulting in health, improved health, for example, and that should be part of that discussion as well.
MS. COLTIN: It is interesting. I had some of the same thoughts, but I looked at my watch and said that we are a half hour late and I had better not ask this question. When they were presenting on the various community health or whatever the -- CHI, whatever it stands for -- Consolidated Health Informatics or whatever, but all of them had an outcome goal which had to do with improving patient safety, improving quality, improving -- I mean, those were their outcome goals and so those clearly relate to a lot of the work we are doing, but the connections weren't always explicit. I mean, they had the outcome goals btu it wasn't clear to me what all the connections were and I looked at it and I thought to myself, well, I can see that they have to start with the basics and communication standards and vocabularies are really, you know, building blocks at the basics.
But my experience with three different medical -- electronic medical records that I have worked with is that population-based query tools are an afterthought and that the records are designed so that you can go -- if you know who the patient is, Mary Jones, you can go in and cut Mary Jones data six different ways to look at what is going on with Mary Jones and her care and her risk factors and whatever, but to be able to go into these electronic medical records and say find me all the patients, who haven't had a mammogram in the last two years, who are between age this and this. They don't automatically come with that capability.
So, I thought, well, are they going to develop some minimum functionality standards for things that these systems need to be able to do or functions that they need to be able to support and population-based queries are one such thing. We ended up having to build our own because the products didn't come with them.
DR. LENGERICH: Having that kind of perspective being part of the panels that come through from subcommittees would, I think, tie us together and we could all --
MS. COLTIN: But perhaps having this broad conceptual framework --
[Multiple discussions.]
DR. LENGERICH: I am fully in favor of that.
DR. MAYS: The suggestion I was going to make though is that maybe what we should do -- because remember the discussion was that first we went from we will have panels come and talk to us and I was like, no, let's see if we can find what the issues are. It might be helpful to say that every time there is a presentation like this at the full committee level, that you give the person the question or that you ask whoever the committee is that is presenting if they could tell us a little bit about its impact in terms of populations or is kind of population health implications. So that we could ask for that -- that relevance is important. So, it forces them to listen to these presentations coming up and to think through if they are going to be asked that when they present codes that -- the 22-ABZ --
MS. COLTIN: That is a real good example. I mean, that is a real good example because I mean I have been on this committee a long time and I used to be on a subcommittee called the Medical Classification Subcommittee and, I mean, we have over the history of my being on this committee, this is the third time that we are now addressing the issue of procedure coding and twice this committee has come out in favor of a single, common procedure coding system. Now we are going to be faced with this again and, you know, the recommendation as I was reading the handwriting on the wall in this letter was to, again, defer that issue and say let's go with, you know, what we need to do to have a coding system for hospitals.
You know, when you start thinking of it from a population and a population health perspective, it is really important because right now there is no cross walk between the data that are captured in the outpatient setting that use CPT procedures and the data in the inpatient setting that use right now ICD-9 CM following three procedure coding when, in fact, as we are moving more and more care into the outpatient setting, procedures that used to be done in the inpatient setting are now -- we can't get overall numbers of how many people are getting these procedures without, you know, writing codes that -- and programming logic that runs on two different procedure coding systems.
We have to collect all of these and then collect all of these and then add them together and it is -- and it is not easy because you are getting a bill from the doctor and a bill from the hospital and you don't want to double count the same patient, who has, you know, two different goals. So, it is not like you can do that really easily.
DR. CAIN: This may just unclear to me or it may unclear to everyone. Since I am new here, I am not sure. But you have said -- first, you had said -- what I was thinking about was that you had mentioned that this group would be involved in the population health. In my mind since this was the group on population, I sort of thought there might be -- it might be directing it or having more than just involvement.
But now as you are talking, is this -- this is -- the idea that this would be an organizing framework that each of the subcommittees would need to somehow incorporate into their activities or their questions?
DR. MAYS: Well, let's put it this way. At the retreat a lot of these issues were raised but the real implementation wasn't. So, for example, Dan's group, which is the 21st Century, does something and then NHII does something and then we do something. Those groups are kind of stand alone. You know, this group doesn't direct them. They stand alone. So, I mean, they are still -- I mean, and that is actually one of the questions. I think there were at least, you know, a couple of times the comment came of it coming back to populations.
So, I think right now what we have done is we have thrown out that -- you know, this laudable effort, but the real nitty-gritty of how is this going to work, whether or not with the Subcommittee on Populations, this all is under that or whether they are separate and, you know, we have committees that come from the boards like your Quality Work Group, even though it is related to us, still has a relationship and does all of its work to the full committee.
MS. COLTIN: It is a work group of the Executive Committee.
[Multiple discussions.]
DR. MAYS: So, those things didn't quite get worked out and, again, I am just, you know, putting these out there as I think that a critical piece for us, regardless of structure is we are probably the people who can make the connections and if we make the connections, it really makes our life in general on the committee better, because our issues that are everywhere -- I mean, I don't know if that is good because we might get asked to come to everybody's meetings, which we don't have time to clone ourselves to be able to do, but structurally I don't know and I think for me kind of in the role as the chair, I do worry about the structural part because it is a management piece and then also the coordination piece.
I know at least last year, I tried to coordinate with the other groups and kept sending e-mails out, which we would like to meet with you if you have any issues during our -- every time I would set the agenda, I did it twice and told them if you have issues that overlap with us, we would like you to come in and if you want us to come in let us know.
It hasn't gone anywhere but it will, I think, this year.
MR. HITCHCOCK: Are we being asked to be a public health conscience?
DR. MAYS: I don't know if we are being asked. I think we are taking it on. I think they want to have a conscience and, you know, we are saying, well, okay, we will continue to be the conscience.
DR. NEWACHECK: I think this is great. Whatever we can get out of it, it would be terrific. The problem that I see at a more generic level is that we don't have a HIPAA population site. HIPAA creates a structure, creates a framework. It creates an agenda of things that need to be done, an audience for the work that is being done, a set of expectations. It is like a whole package for that part of the full committee.
MS. COLTIN: We could get in the game in that, though.
DR. NEWACHECK: Pardon me?
MS. COLTIN: We could certainly get in the game.
DR. NEWACHECK: I wonder about that because --
MS. COLTIN: -- how the administrative transactions should be revised to incorporate race. I mean, they don't have a monopoly on that.
DR. MAYS: I am going to tell you because that is part of what we are going to do today is -- actually, if you go through, which is what Sheree has done -- these things have come up several times and I don't want to say whether we dropped the bomb. All I know is that it was raised and I don't know where it went and I am kind of thinking maybe we should try and answer some of these things.
DR. NEWACHECK: Historically --
[Multiple discussions.]
It has been idiosyncratic to the membership. The issues that are raised are like somebody's pet issue and they want to work on it. So, we do it. We don't really have a client. We don't really have an audience. We create one and sometimes that works and sometimes it doesn't. It seems like if we had more of a framework and we had more -- somehow built in an audience to what we are doing, we could have far more impact than we do if we are just kind of floating trying to come up with ideas that we think are interesting.
DR. MAYS: Let me make a comment about that. One, I think that maybe we have arrived and have more of an audience in the sense of it you just talk purely about race and ethnicity with the new National Center on Minority Health and Health Disparities and also with OMH now having an advisory board.
I think those are going to change the landscape a little bit, which is why, for example, we now have people who also -- staff members from NIH, who are a part of those groups because to some extent that is to be -- that should be one -- we should consider them as a client. But the other is maybe we ought to think about this issue of the clients are also the other subcommittees and that -- at least this is how I have begun to think about some of this when I look at the fact that standards probably is not going to do the race and ethnicity piece, but, yet, somebody might pipe up and say something about it and that maybe and that maybe part of our consciousness is working with them to some extent to make sure that that piece is handled there.
I was asking Kathy in terms of her report whether or not -- I said, oh, are you all going to do a thing in terms of equality on race and ethnicity. Again, the problem is -- and we are really few now. As you can see, Barbara is not here.
I keep hearing you all telling how many years you have been on.
[Multiple discussions.]
How dare you think about it.
MS. COLTIN: I said I would come back around getting the report through the review process.
[Multiple discussions.]
I was going to actually lip something you just said and that is that, you know, in my mind I actually instead of thinking about us or -- I think that we are in some ways the client for the standards committee. We should be telling them what we need. They are hearing from the public at large about what is needed. We should be part of that process, too. They should be hearing from us about what is needed.
You know, on the quality side, I am certain, I mean, I can tell you -- we haven't voted yet or anything but a lot of the recommendations that we are going to make around being to improve the measurement of quality will involve the ability to support the diffusion of electronic medical records in there for the adoption of standards and common terminologies and things like that.
I mean, I was at a meeting at AHRQ the beginning of this month were we were reviewing the ChuSpan(?) initiative and they were putting together their next RFP and we kept hearing over and over again that because of the burden involved in quality measurement that the idea of filling in a lot of those gaps in the National Health Care Quality Report were taking that measurement below the national level to the states and to the levels where action can really begin to occur is going to be dependent on electronic data collection or data that is collected as part of the routine of delivering care, you know, not that is layered, extra burden kind of process.
That means that we are going to have to make recommendations to that subcommittee about the kinds of data that need to become part of the standard electronic medical record to support that.
You know, three years ago now, the AMA proposed and about two years ago, they actually sent out an announcement about the establishment of Category 2 CPT codes that were going to be codes that would enable you to do performance measurement. In other words, you would submit them on a claim but they wouldn't be paid. They were for information purposes. Nothing has happened with those. Nothing. I know that we will be making recommendations about that. Now, that is a code set issue. So, that is, again, going to land right back in that committee.
If they don't make those things a priority, if they don't hear from us as a client in the same way they are hearing from software vendors and payers and others who are clients for what they are doing in terms of defining HIPAA standards or making recommendations about HIPAA standards, who is going to do it? You know what I mean?
I think we have to get them to consider us as a client or the recommendations aren't going to go anywhere. We are going to recommend them and then the committee is just not going to address them.
DR. MAYS: Do you think it is possible that we ask for an hour in which -- and it means suspending something but that an hour when the Population Subcommittee can meet with the chairs and the staff of the other committees. We have this kind of discussion and we identify -- I think, you know, I can even from this report --
MS. COLTIN: Isn't that what the Executive Committee is pretty much? I mean, it is the chairs of all the subcommittees.
DR. MAYS: -- call for --
MS. COLTIN: for an Executive Subcommittee session at the next meeting.
[Multiple discussions.]
DR. MAYS: So, can you get us an hour at the November meeting, where it is the Executive Subcommittee meeting?
PARTICIPANT: [Comment off microphone.]
DR. MAYS: -- like in November where we meet again -- so, it would be November. I mean, it would take us just as long to get a call together as it would be to be able to do it in person at the November meeting. Is that possible?
PARTICIPANT: Who are you trying not to have at this meeting.
DR. MAYS: I guess I would just want the chairs and the staff that staff those committees. The lead staff
-- I am sorry, not all the staff, the lead staff and the chairs.
MS. JACKSON: In closed session?
DR. MAYS: It doesn't have to be closed -- it doesn't have to be closed, but it is just that it is acting -- to some extent it gets too diluted when it is everybody because what the real goal of the session is is a commitment for work. I think the chairs can go back and -- see I don't want to try and work it out on the floor of the whole meeting as much as I think first the notion of exactly what Kathy said, is the discussion of the quiet relationship is one -- and I think we have to prepare a presentation to talk abut here are a variety of issues that concern you, that concern us, but they are issues that it would be very useful to have you addressing in your hearings. We would be more than happy to consult with you on it, but that these are important and they should form a part of your work portfolio.
It is one of the ways we kind of overwhelm ourselves with trying to be experts on every issue. Instead, they have their hearings and when they set up the questions, it is the -- it is almost like it is the population issue and the populations issue, that they also bring into their hearings that we are asking and that then when they discuss their recommendations at a full committee level, my hope would be that we would get recommendations that are specific to the kinds of issues that usually come to us and that we could discuss them then as opposed to being -- I mean we shouldn't have to go out and have hearings on HIPAA. HIPAA is having -- they are having the hearings on HIPAA. So, have them include these issues there.
So, I don't think that is something you want to start with with a full group as much as I think you will want to start with the Executive.
MS. JACKSON: It is sounding to me like a steering Executive Subcommittee meeting or call or component of time. I was trying to make expeditious use of the full meeting in the fall -- I mean in November with the full committee and it is pretty jam packed. We are just looking logistically how to get it --
PARTICIPANT: It is too bad we didn't try to do it this meeting.
MS. JACKSON: Or the way John was envisioning how this meeting would work, the September meeting itself. Now, he was mentioning today how hence forward you can use assistance in the meetings, even have a Executive Subcommittee retreat, an interim call with this kind of steering focus and in the September meeting everyone talking to each other not just at each other and throwing ideas.
DR. MAYS: See, I just hate to do this on a phone call. I really would prefer to do this face to face. So, I am just trying to figure out what the possibility is. What do other people think?
DR. NEWACHECK: I think the phone call is a mistake.
DR. MAYS: -- and you want them to do something. It is like you hear beep and you go who signed off.
MS. COLTIN: Of course, maybe part of it is our fault in a way, too, and in not learning enough about their agenda to know where to plug in because they hold three sets of hearings on code sets and we could have said we would like to testify at one of those hearings about the limitations of code sets for measuring quality or for some other purpose that we are concerned with, functional status. What is happening with the recommendations that came out of that group and code sets for functional status?
Part of it, you know, it is not all --
DR. MAYS: Exactly. That is why I say I have gone through this stuff and you can see there are places where it has been raised and I don't know if -- for lack of whatever and I think it is exactly what you said is that, you know, maybe the committee has kind of done what -- whoever is on here and what they are interested in as opposed to you kind of still have this other -- nobody is interested in it but you still need to be doing it in order to kind of, you know, build these -- I am just trying to put that back on the table.
Can we maybe walk through this --
MS. JACKSON: -- logistically to see how we can accomplish it an hour doing what at the time that maybe another subcommittee would be meeting, we can use that time or --
DR. MAYS: But remember no other subcommittee can meet because we want all the chairs or --
MS. COLTIN: Or if the regular meeting starts at 10:00 instead of 9:00 and we meet from 8:30 to 10:00 or something like that.
DR. MAYS: I was going to say, you know, if there is a way in which to do that or if there is a way in which for us to achieve this, next time if it is a three hour meeting that, you know, we will cut back and let the other group do two hours and then take the third hour and have -- you know, ask that everybody come in or something. We just need to know ahead of time.
DR. LENGERICH: And possible mechanics -- there is this work plan, which goes into the future.
DR. MAYS: It is a little outdated but yes.
DR. LENGERICH: Well, but as it gets updated maybe there should be a column about population health implications for each of these activities. I guess I am sort of feeling like this conversation is good. It needs to happen before you get down the road of all the hearings so that they can set it up --
DR. MAYS: Oh, I agree.
DR. LENGERICH: So, if it is a goal of the committee to have more population health focus, can it be woven into -- somehow into the work plan, which is -- what is going to go on in the future.
DR. MAYS: Debbie, when do the other groups do -- we apparently have been sitting with kind of an outdated one, but when do the other groups do their work plan?
MS. JACKSON: This is the work plan I developed for kind of keeping up with what is going on with all of the committees and subcommittees --
DR. MAYS: Do we change each time?
MS. JACKSON: Each meeting it is updated.
DR. MAYS: Oh, so we do a work plan like that. So, they could -- we could go back tomorrow and ask them.
MS. JACKSON: What I do is I kind of select what is happening within the group, the subcommittees, and working groups and put them in this work plan and it is updated each meeting. This is what I am hearing and trying to reflect so everyone will see in one big matrix what is happening for the full committee and all of its components. So, that is a great suggestion that all of the subcommittees and components need to know that that is an integral part of where they have to go and what they have to address, all the way through in their work.
You were talking about the subcommittee work plan that hasn't been -- for populations that --
[Multiple discussions.]
DR. MAYS: Yes, that is what I was talking about in terms of when the other groups do theirs. We really have to talk to John because you can see this taps right back into the bigger notion about the discussion of population because I think it should be about population and populations. Let's not lose it. That we are not in there, you know, just getting on there about populations but that it is population and that that is what we want to see reflected a little bit more.
Okay. Could I get you to turn to --
[Multiple discussions.]
Oh, sorry. The one that says "Summary -- Activities of the Subcommittee on Population Specific Issues." One of the things that I asked Sheree to do was to try and go through all of our notes from way back when and see what we have been doing in terms of race and ethnicity because part of what I was concerned about as I come in and I hear you all say, well, two years ago we did that already, but it did seem like sometimes we remember, sometimes we didn't. It is like I don't want to be embarrassed in these hearings and have somebody later say I have been to that committee three times within three years and talked about the same things. So, I wanted to find out who has been to us, what they have talked about.
Also, I thought that the recommendation that was made -- and I think it was Kathy that actually started us on a path of let's begin to plan what we want a report to look like.
Part of that planning is, well, what do we already have? If we don't need to have a hearing about something because we have asked somebody about it, like the year before, then we need to go get those recommendations because part of what I think is a problem is that this committee has made recommendations several times and they haven't been picked up. Some have. Some haven't. So, what I am trying to do is to pull all this material together so that that can be in the report and we can say in 1992 we asked for this and we asked again in 1996. We think it is critical that someone, you know, take a look at this.
You know, I think that is a lot stronger than -- 2000 said something and so we are just asking them, but to say this has been a long standing problem that really needs attention I think will get us much further with the administration and I think will help us in terms of partnerships with organizations like APHA, et cetera, who will chime in on some of these things.
So, what I have done is the copy that you have, I have put some little arrows where there are areas that I saw things that, you know, maybe we should try and find them. The rest is -- you can have it to read. It is not a big deal, but I want to just go through the specifics of where I think there are things which either we were asked to do and I am not sure we have or there is a report we should be trying to get and have a discussion of those because I think it is going to help us in terms of our work plan. That will bring us back to, you know, kind of the very issue we were just talking about is how do we work with other groups.
MS. CRUTE: -- but one of the things I noticed is that every now and then someone would allude to a conference call or a very specific format of an OMB letter that I couldn't find on line anywhere in the year it was supposed to be completed or things of that nature.
So, what I wanted most -- you were there. If you remember, please say something so that I can search it or -- be revised or edit it because all those side conversations have not been included.
DR. MAYS: There are times at which there are -- I was just going to say there are things that maybe are alluded to that we haven't been able to necessarily find. So, as I point to some of these if you think you have these, we would be happy to receive them.
Go first to page -- thank you for reminding me of that because that is part of where we need help and then we are also going to ask some of you with institutional members if you remember something happening. The first place I thought, if you go to page 6, you see I have an arrow up there about Barbara talked about the concern that race and ethnicity data not being dealt in the work on administrative simplifications.
Marjorie talked about what was going on in terms of -- if you look down below, you will see that Barbara kind of gave her opinion and then what you see is Dr. Iezzoni closed the discussion by asking Dr. Starfield to head an ad hoc group to develop a proposal for subcommittee action on these topics. Dr. Takeuchi, Moore, Amaro and Friedman joined the group as well.
If you all have memories about exactly what went on and whether -- we would love to find out.
PARTICIPANT: The minutes are edited.
DR. MAYS: So, we can't find out exactly what you were specifically going to do and we can't find -- we don't know if there is a product from this, right?
MS. CRUTE: No. The only product -- I tried to match the information -- it was a matter of reconstructing.
[Multiple discussions.]
MS. COLTIN: When you are able in the minutes to say this is one --
DR. MAYS: I think she has been trying to go back and forth between things. So, part of putting it out there is that just in case you remember something -- because this will be an example where if you all really did work on this problem of administrative simplification, particularly if you looked at -- report transaction. It would be great because I think it has kind of dropped out to some extent from the discussion by the HIPAA people.
DR. FRIEDMAN: Vickie, looking at the people who were --
[Multiple discussions.]
-- the only thing I can think of is that being put on the group must have driven David Moore --
[Multiple discussions.]
-- off the committee and he retired.
DR. MAYS: But, see, that may be what happened then is that you started losing people and so the issue kind of got lost and it is kind of like as Paul said, whoever was there wasn't there --
MS. COLTIN: Honestly, I think what happened is reflected in the problem that we were talking about earlier in that this subcommittee, populations has for each period of time until it finishes its work, picked one issue that might be unique to a particular subpopulation, maybe more than one in some cases, but, you know, I would say that, you know, the issue of functional status is more crosscutting than some of the others that we have looked at, but Medicaid managed care, again, a very -- a subpopulation from an economic substrata standpoint and insurance standpoint and that each of those people have had an interest and it wasn't their interest that was the one that was going to soak up all the time over the next, you know, year or two and I think people lose interest -- a lot of the people who were part of the Subcommittee on Populations were there because they had an interest in the issues of a particular subpopulation or a particular cross cutting issue, but a very narrow, cross cutting issue, like post acute care or mental health or substance abuse services, I think, was a -- issue.
So, when you got off that topic and they were spending a lot of time meeting on broader issues, people did lose interest. I mean, I think there aren't a lot of people whose interests are very broad and really at this whole level of population health. You know, I am interested and can plug in almost anywhere on those kinds of things. So, I think that is part of what you observe when you look at the history of this committee and I think that it is something that we kind of need to address as we go forward.
How do we pick things to work on that do engage everybody and fit in the broader framework? Or you are going to have those same --
DR. MAYS: I know that, for example, when Marjorie talked to me about the committee, she made it clear that you can't do just a little population. I couldn't do just a little minority group and that was it. You know, I said that is just kind of not me anyway.
I think it is a committee issue though as to -- and I think that maybe that is why if you look at Sheree's notes, which are excellent, every couple of years there is another name for this committee and another, you know, direction it is going in and that is my experience and I said this at the retreat was like, well, we had that discussion about populations last year. We are back having another discussion about populations again. I really would like not getting it. Why do you keep talking about this committee and not the others?
So, we do have, I think, an issue and it is one that, you know, it would be helpful for us to figure out how to do this so that we finish things. We have, you know, the products. We do this kind of working with other groups because otherwise I do think we become like the flavor of the month and it is like, well, you know, the administration was concerned about that for about three months and it is fixed and then it is who you have next and what they are concerned about.
Okay. We will try and see -- if you all have anything about what happened -- and poor Dan just drove them all off and he won't tell us why but --
MS. COLTIN: I can tell you that one thing related to a couple of these issues that came up is that sometimes if an issue like this came up in this committee and the issue had to do with like Barbara's issue of the fact that HIPAA wasn't adequately in any of the standard transactions addressing the need to collect data on race and ethnicity.
You are not going to find anything in the subcommittee notes about that. There was a panel then scheduled at a full committee to address it. So, you have to go back to the --
[Multiple discussions.]
MS. CRUTE: [Comment off microphone.]
MS. COLTIN: That was an excellent panel on HIPAA and race/ethnicity data collection at the full committee because this issue was surfaced here and we asked to have a panel. I remember that panel was an excellent panel. It was a long one. I mean, we had a good amount of time on the full committee agenda where we heard both from providers who might be asked to report it on claims transactions, as well as from representatives of various minority populations advocacy groups and what some of their concerns were about having the data collected.
So, from a privacy standpoint, we kind of heard some of their concerns as well and I thought it was, you know, one of the things that came out of those -- that panel or those discussions that we had was that there really wasn't a clear consensus in either constituency. I mean, there were some providers who said, oh, yeah, we collect it all the time and it would be simple. We could just put it on the claim and, you know, it is part of in the patient management system and it is just like name and address and it would just go automatically with the data and others who said this would be a real burden and whatever.
Then, interestingly, on the population side some of the various minority groups did not -- some thought it was a good thing because you can't identify discriminatory practices if you don't collect the data. So, they wanted to see the data collected and then there were others that said, no, we don't want to see the data collected because it is an invasion of privacy. So, we really didn't see a consensus on either side and didn't know how to deal with it.
[Multiple discussions.]
MR. HITCHCOCK: I still think, though, that one of the things we were told was that you had to make a business case for collecting these data. The other variables that were -- to collect these data. We got as far as having race/ethnicity placed on sort of a second tier level where these data could be collected. They were sort of optional, but not required or something to that effect.
[Multiple discussions.]
Yes, but what it did do was spawn this group that Marjorie Greenberg and people at NCHS are involved in. As I understand it, they are trying to work through the standards of development organizations that HIPAA worked with to convince these SBOs that it is important to collect racial and ethnic data. I think we have got that effort going on and I don't know the status of it.
MS. COLTIN: So, we should hear from that effort.
DR. MAYS: I was just going to say then maybe the way to deal with that piece is to ask Marjorie if she could give an update and then to check with the HIPAA people to say are you going to discuss this at any point in your -- in the materials that you --
MS. COLTIN: Maybe you should.
DR. MAYS: Drop down below that to the second point on here about Ms. Coltin prepared a final draft of the report on Medicaid managed care, noting that representatives of HCFA, SAMHSA, HRSA contributed to the document. Dr. Iezzoni asked that the issues related to racial and ethnic identifier be reviewed, saying the subcommittee will eventually have to deal with how this data is captured.
She noted that the subject came up repeatedly in Medicaid managed care hearings. Now, my recollection of this is that there is a section, that there is a Medicaid manage report that is on data collection for race and ethnicity. So, that is there and I don't know whether or not again we need to highlight that in our report.
MS. COLTIN: I think any time we achieved consensus on a recommendation we should remind people about the recommendation because it is hard enough getting to that point and where we have gotten to that point in any of our reports that were adopted by the full committee, we need to remind everybody that we all agreed to this recommendation.
DR. MAYS: So, what we will do is go to the Medicaid managed care report and what we want to do is pull out of that the section on data collection issues about race and ethnicity -- this is something that was recommended and blah, blah, blah, blah, blah. And, again, as we begin to think about where we are going report-wise, that is work that is done. We have it and it has been agreed to and we can --
MS. COLTIN: -- a little further up, after Barbara's comments, it says the conversation included brief references to geocoding as opposed to collecting data on encounter forms or as part of enrollment. Several years ago we had somebody come and testify. Again, I think it was at a full committee meeting on geocoding as a possibility. You know, that was a long time ago and certainly it may be worse -- there are a lot of committee members who weren't here then because it was a long time ago. It might be worth revisiting that in terms of looking at its pros and cons limitations for being able to look at disparities, for instance. You know, if you don't have data on HIPAA transactions, but you do have data on address and you could do block level coding, what are the biases in doing that? How accurate is it? How inaccurate?
DR. MAYS: When you say "revisit," do you mean revisit in the sense of having someone come in and talk or revisit in the sense of pulling the materials out?
MS. COLTIN: Both, actually. I think we ought to know what we heard and then we ought to hear it again.
[Multiple discussions.]
I am not sure if the testimony of the full committee came around this time when Barbara brought this up or whether it came after it or before it. I thought it came before it. I thought it was longer, further back in history, but I don't know. I have been on here so long that sometimes --
DR. MAYS: Again, that is one of the things --
[Multiple discussions.]
Okay. Great. Let's go to page 7 at the bottom. The primary questions posed about race and ethnicity data were who should collect it, why and whether it should be required of all patients. Concerns about protecting patients from discrimination were raised as well. I think this is what you were just talking about.
It says the subcommittee agrees that the data should be collected on everyone. I don't know what it means by agree -- on everyone with proper privacy protection, given that the information clearly is not just focused on race and ethnicity but language and other factors. They decided to call it enrollment data.
Now, this one, I really didn't understand kind of how people got to that and if you agree at the subcommittee level, we can't quite tell whether or not it went to the full committee. If so, then whose report would enrollment data be in? Would it be in the HIPAA? The question is like where does this go? They agreed on something but did it go anywhere?
MS. COLTIN: It went into the Medicaid managed care report because that was the context. I think that the issue was how it got framed and phrased in that report. So, you may see something there. That is really the context in which they were making --
DR. NEWACHECK: -- like we have different data sets. We were talking enrollment data versus claims records, things like that.
MS. COLTIN: Because they wanted it collected on everybody. Well, we didn't say not in claims data. We just said if you did it in claims data, then you wouldn't know anything about people who weren't getting services and what biases there might be, you know, in terms of who gets into the system and who doesn't. But on the other hand, if you don't also collect it on claims data, you don't have any information about the uninsured.
So, my -- I mean, my personal bent is it should be collected both places.
DR. NEWACHECK: But it should be in our report as to where we specified where it should be collected and it is enrollment.
MS. COLTIN: I think it is enrollment.
DR. MAYS: That would be in the Medicaid managed care report.
MS. COLTIN: Well, it was easy there because it was an insured population. So, we were able to say if you put it in the enrollment, you knew about that whole population, but when you are talking about -- or subpopulation. When you are talking about population health, then you have to think is that the recommendation you would make. It may work for Medicaid because they all have insurance. So, you put it on enrollment. You have got it on everybody, but a lot of people don't have Medicaid. They don't have anything. They are uninsured. How do you get data about that population and any potential discrimination if you don't have it on claims.
DR. MAYS: See, I think that is an important thing that should be in the next report.
MS. COLTIN: Even if those claims are not -- you know, even if they are free care, they generate a claim that goes into a statewide population -- data set as free care. It exists at least. It is not a claim in the true sense.
DR. MAYS: Let us make sure we capture that in our report.
Go to page 10. Afterwards I will ask you to look through this whole thing, but these are the things I flagged.
Tony DeAngelo reported on HHS's plan to improve the collection of race and ethnicity data. He discussed in some detail a report that included testimony from the Healthy People 2010 meeting. He mentioned six other reports, ten recommendations in the meeting. They were develop a ten to twenty year schedule for the periodic targeting of racial and ethnic groups and surveys -- you can see all this.
My question is did this make an impact? This is the testimony of Healthy People 2010. I mean, if they did this, I want to know. Develop a ten to twenty year schedule for periodic targeting of racial and ethnic --
MR. HITCHCOCK: This report reviewed what the Department had essentially done internally. I don't think we looked at the NCVS work when this particular report was developed. It did go back to testimony that Healthy People 2000 people or 2010 people had received at various times when they were holding hearings applicable to certain populations and subgroups. This report reviewed recommendations that came out of --
DR. MAYS: But these are good recommendations that we would want to in some way, I think, resurrect or say something about and reference the fact that, you know, this was testimony in Healthy People 2010, produce state level racial and ethnic data. Expand or establish -- I think we have to think about this one -- new registries for chronic conditions, disseminate relevant findings back to community where data are collected.
MS. COLTIN: I think No. 2 is particularly --
DR. MAYS: I know. See, that is what I said. These are really great. I was just curious as to -- we should dig them up again.
MR. HITCHCOCK: I think that this report was -- let me say there was -- not unlike the islands report, which there was never any real formal response to. Both reports have been influential and the work is moving towards a lot of these recommendations, sometimes without really even acknowledging them but just knowing that they are there, that people are aware of them.
HRSA really boosted the work that they were doing in the --- based on their involvement with developing their report more than just the report I would say and a lot of it -- a lot of -- that sort of thing goes on in this, too. These people are working on these things maybe through their involvement with the committee that did this in developing the report than the actual report.
DR. MAYS: But I think it might be good for us to highlight these as recommendations again. I mean, as we kind of say where are we going, this is an instance where we don't need to not give this testimony. We need to reference this and to some extent if we could say what has gone on, that is great. But, again, here we have insure that HIPAA standards include racial and ethnic identifiers. That is being called for by testimony to Healthy People 2010. That is being called for --
MR. HITCHCOCK: That is only a minor part of all this. Maybe an appendix. I don't know. It had little role in this.
MS. COLTIN: We need a copy of this report.
MR. HITCHCOCK: I helped write this report. I had the contractor that edited it and -- I mean, we just don't want to get hung up on --
[Multiple discussions.]
They are an important part of this but there were several other work groups, reports and letters and memos that were of equal weight.
DR. MAYS: Okay. We hear. Can we at least look at the report if we can find this report?
MR. HITCHCOCK: Yes. It is on the Data Council Web Site.
DR. MAYS: Okay. Is there a name to it so she can find it easier with a name --
MR. HITCHCOCK: It is pretty self-explanatory.
DR. MAYS: All right. Perfect. We are not going to have hearings forever if I can help it because we have got a lot of stuff already it looks like.
Go to page 11, the next page and we have Jim Weed, who came and saw us. He made a presentation on the possible impact of the OMB standards on race and ethnicity, data gathered for vital statistics.
So, he has updated us. So, I think we ought to find his previous material and use his material from today, but it also said that the subcommittee also decided to write to the Data Council in support of the NCHS recommendations on age.
MS. CRUTE: I have to admit that completely confused me. It hadn't come up before and it never came up again. It was like this one sentence and I could not figure out what it was related to. It didn't come back.
DR. MAYS: Okay. Can you find this letter, Debbie?
MS. CRUTE: I mean, I can pull the minutes.
DR. MAYS: We just want to see what it says. The subcommittee also decided to write a letter to the Data Council in support of the NCHS recommendations on age. Oh, you know what it is about?
DR. FRIEDMAN: No, I don't, but the only thing I am wondering is if perhaps that was a reference to the -- talk about arcane -- the NCHS standard for the new standard million for age adjusted mortality rates, that somehow or another it may have gotten translated in the notes, the committee minutes.
MS. CRUTE: The word "mortality" was not there, but I could go back and look and see --
[Multiple discussions.]
DR. MAYS: If it is, then it has been done because they have put a little -- I can see it visually, the little special thing they put out. Okay. That is done. We don't need to worry about that, too.
Page 12, this is all about the insular areas report. I am going to handle that separately in a sense. I am going to send that to you because we have about a 20 page report that she did on that. I think it should serve as the basis for the planning for the API hearing. Before we even do anything with the API hearing, we should draw upon what we already know and either -- if we ask any of the same people, it is to update us or to figure out what the new issues are.
But also here there is this issue of IPRO, quality improvement programs, IPRO role in analyzing race and ethnicity data. I was going to ask Kathy what that was about.
MS. COLTIN: The first year that -- what was then HCFA, collected Medicare HEDIS data from health plans. They asked the health plans not simply to report to them the numbers. In other words, how many people were in the denominator for a measure and how many people were in the numerator. We had to do that. But we also had to provide them with a record for each person in the denominator and each person in the numerator that included the HCFA identifier number. In other words, they wanted to know among our Medicare Plus Choice enrollees in each health plan, who was counted in that denominator and who was counted in that numerator.
We had to submit that data all to HCFA at the time. They then matched it up to their files where they had data on race and ethnicity from Social Security, I believe, because I think that is where HCFA got it, because they wanted to look at the numbers from the health plans by race/ethnicity.
The health plans didn't know the race/ethnicity. HCFA did. If we told them who was in the denominator and who was in the numerator, they could break it out by race/ethnicity and look at that. So, they had a pilot project the first year to do that and they contracted that with IPRO, which is one of the, you know, peer review organizations that is I think out on Long Island. I think it stands for Island Peer Review. So, it is in New York. So, we had them come in and talk about what they learned about the quality of the data. They did an audit on it or something.
What happened is we didn't get what we asked for. We got a presentation, but it didn't really focus on the issue that we had asked them to come in and talk about. They talked about their audit of the Medicare HEDIS data generally, but not as much about what they learned about the race/ethnicity stuff.
DR. MAYS: Okay. The question would be, one, is this something then that is routinized that it continues to do this or is it something that --
MS. COLTIN: You know, I don't even know. I can check with you. I know that for several years at least we did continue to have to report the patient level, the member level data. I think we still do, but I am not involved in that at that end of the process in my health plan anymore. So, I really have to go back and check with people and say does CMS continue to require us to provide not only the numerator and denominator data, but also the individual patient level data.
If, in fact, they do and we are providing it, then the next question is are they continuing to link it to race/ethnicity and are they doing anything with that and if so, what.
DR. NEWACHECK: Can I interrupt just for a second. It is 4:15 now and we are kind of late on the agenda. I have some important things, like what are our products going to be in the November meeting. Is it possible since Kathy is the sort of institutional memory here that maybe like this could be off line about clarifying things that go back before any of us --
DR. MAYS: Okay. Kathy, maybe you and I can have a conversation.
MS. COLTIN: I can look through my files and see what I can find on some of these things, too.
MS. CRUTE: If anyone reads this and finds glaring errors, mistakes, misinterpretations, wants to explain or give me a note, I can pass out my cards and if somebody could just set me straight on whatever it is as I go forward and backward.
MR. HITCHCOCK: Why don't you pass out your cards.
DR. NEWACHECK: So, how will you get -- Kathy has been on for how long, Kathy? Eight years? So, that goes back to like 1995 or 1994.
[Multiple discussions.]
DR. MAYS: I think Marjorie is -- I assume Marjorie and Jim are also like --
[Multiple discussions.]
Well, some of these things I really want to know what happened. So, for those early ones we will do that. I think that that is fine.
Let's see where we are. Let us pull that stuff together and I think that is going to help us to determine where we need to go in terms of a report. Sheree has just been -- as you can see, I just got this about a week ago and then I just got the insular areas report like Monday night. So, I think -- that I will send you to look at and I think that will then flesh out -- we actually have quite a bit and so I am not real prone to wanting to keep running around doing hearings.
As a matter of fact, what I would like to suggest is that what we think about is that we rather than waiting until we get to the end, it might be useful for us to think about if we are going to impact the administration, I would to find that we do this and it is like, well, who is listening. Nobody.
But that we see if we can't, where we have had hearings already, pull together something for that particular group, like we could talk about population-based surveys. That is very clear that that goes to NCHS. That is very clear that that goes to NCHS. It is very clear that that goes somewhat to NIH. It is very clear that that goes to -- we should, I think, see if we can't get a move on with getting commentary out and then at the end have one big report that gets put together in some little fancy looking something or other for distribution, but, otherwise, nobody is going to, you know, care, other than us.
DR. NEWACHECK: So, we don't actually have an outline of it. We are working for that big report --
DR. MAYS: No, because we are trying to find all the stuff that we have and I think she is almost -- we will finish finding out what else we have that will then help us to know, you know, what we have and then I think to -- and then that is what I will ask her to do is to work with us on outlining what we have. Then I think we can talk about what we want to do and we should be able to do that at the next meeting in November.
MS. COLTIN: During most of those early years, the committee did an annual report and I am wondering if --
DR. MAYS: Subcommittee or the committee?
MS. COLTIN: No, the full committee. But it summarized the work of the subcommittees in the annual report and it might be a good starting point to at least get a feel for it. You know, in that year was there anything much going on around the topic.
DR. MAYS: I think we looked at that and I think we looked at the 50 year thing. So, I think that that is helpful. What you find is that it is really summarized, really decided to look at the issue of bottom line managed care.
MS. COLTIN: It is over a lot of minutes.
DR. MAYS: Okay. Thank you for bringing us back. I think that is a good suggestion. I assumed that you all had been here a long time.
Let's talk -- I think we could skip all the way down to the discussions on -- and then kind of come back to a work plan if necessary. Let's talk about a hearing first, which should be down here for Friday, so that know kind of where we are and where we are going. I think everybody got one of these whether they are going to the hearing or not? Okay. Great.
So, whether you are going to the hearing or not, this is the preparation material for the hearing. Now, let me ask one question, Gracie, as I do this quickly because I also got this today. Where are the questions? Do you have them? Someone have them? Unfortunately, Audrey is not here, but there is a set of questions that we have generated that serve as guidance for people to respond to and that looked like we don't have that in here. I think that is probably critical. We also put a notice on the Federal Registry that indicated people could go to a web site and give us written comments that we did that in this instance because there are 550 -- and we know that we are only going to get a small number at the actual meeting.
Edna Paisano has been just a real gem in working with us and, you know, she knows many of the people and has called them directly and has gotten them to agree to present. As you might remember, for Denver, the reason we are going to Denver is because the Office of Minority Health is having a meeting on Native American and American Indian health. We hope to have culled from that group individuals who could give us some insight into their data needs and health disparity issues. We were able to call some but we also had to pull other people in to travel to the meeting to do presentations.
So, as you can see from the agenda, we, I think, have a fairly good group of individuals. The issues, let me just run through the issues so that you have some understanding about the issues. I am trying to practice what we preach. A big issue for them in terms of the data is the identification of a person if they are not in a tribe. So, part of what they want to talk about is even in terms of before you get to the collecting of the data, it is the difficulty if a person isn't in a tribe and identifying them as belonging to the population.
The issue of the determination of belonging, even if you are not in a tribe, but as I understand from being on the phone, the number of -- the criteria may differ, depending on what tribe you are in. So, even that is difficult. So, part of what we are struggling with is it is almost like a point that Dan was making is that we may not be capturing the full range of individuals from this population because of either they missed just a little bit from a tribal point of view or that they are in an urban area and once they were misclassified, they just never -- they themselves have just never corrected it.
So, one doesn't know if the increase that we have seen in terms of the American Indian in the census is because there was such outreach about the importance of people claiming their ethnicity or whether or not it was people who were really white and have a little bit of Native American in them that did it.
So, that is part of what the issues will be there. The health disparity issues from the tribal perspective, again, part of the struggle here is sometimes the tribes, particularly if they are being funded by Indian Health Services, what you may find is that a lot of money might get forwarded to some things like diabetes, like fetal alcohol syndrome and not into other things. So, from a tribal perspective, they may have issues about other health concerns.
Some of the collection of data when you are in a tribe has a lot to do with Indian Health Service and it has a lot to do with what is kind of funded there. So, there are some really interesting issues that come up in terms of if you are getting your health care through Indian Health Services. Also, the extent to which you have access to care once you are in a tribe what happens when you move to other tribes. So, there are some interesting issues there.
MR. HITCHCOCK: Lack of general health status data, too. They have some good data on health care. But they don't have a lot of data on the health status.
DR. MAYS: Well, their primary data collection is from Indian Health Service and that is because of a visit as opposed to being a part of regular surveys on health. So, those are some of those issues.
Urban rule, Indian issues were brought to our attention that it is important that we actually have some comparison because they can be quite different. In a rural area, how you get your health care, what you are actually -- how you are getting counted, which again is more likely to be in health care encounters than anything else. Then in urban, it is probably likely to be misclassification.
So, those are some of the major issues that were identified that have been brought to the table for us to look at. A set of questions, unfortunately, they are not in here. I will make sure that we have them by the time that
-- I will see if we can get them before we go or you definitely will have them at your place.
DR. NEWACHECK: How many people are going? Is it just the three of us?
DR. MAYS: I don't know.
You are not going. Barbara is not going. Gene is not going.
[Multiple discussions.]
We don't need to have a quorum. Yes, because we also thought that some of the other people -- but it is a long story about the NIH staff. I think Jackie also -- we had travel issues. I think that was part of --
[Multiple discussions.]
Okay. November 8th was what we could settle on as a date for the state hearing and we chose that because of APHA. APHA begins that Sunday. This is that Friday and we think several of the people will be attending APHA. MS. COLTIN: Where is this?
DR. MAYS: Philadelphia. I don't know who is committed for that.
[Multiple discussions.]
So, that is what I know for that. Maybe by then we will have it worked out in terms of these other issues about travel.
For the states, if my memory will serve me correctly -- this is where I have to rely on both Gracie and Audrey and Dale -- the first guy --
MR. HITCHCOCK: Delton from NCHS.
DR. MAYS: And his area of expertise?
MR. HITCHCOCK: He is a state and federal registrar and, what, past president of -- past director of NAPHSIS, I guess.
DR. MAYS: You could see from what Jim was saying that that is who is there. It is not that Jim -- the states that we are asking, that I remember correctly, are California, Massachusetts. Help me, Gracie or Dale.
[Multiple discussions.]
We have Allen Onaka, I think Peter Abbott. We have -- the states that were on the side of like needing kind of pull it together in this area, I think that we were trying to find out what the barriers are, how they can be helped. We were considering several other states. I think we are working on Tennessee. We dropped North Carolina because I think that came up. North Carolina actually is doing pretty well. So, it is Tennessee, Alabama, those were the two that I remember. I don't think we were doing any others.
Kathy, you suggested Florida for what reason?
MS. COLTIN: Well, simply because of the large immigrant populations in Florida.
DR. CAIN: Alaska is very interesting. I don't know what their systems are like. But they have a lot of issues there.
DR. MAYS: Okay. Let us consider that.
We also have --
DR. FRIEDMAN: Vickie, if -- I don't know about the propriety of these things but if I can go on November 8th and if Bruce Cohen, who is an epidemiologist, who I work with, can't go and if it is appropriate, I could do -- I would be able to do a presentation.
DR. MAYS: Okay. Good.
MR. HITCHCOCK: Is it Alvin Onaka?
DR. FRIEDMAN: Alvin.
MR. HITCHCOCK: He has got a national position now.
DR. FRIEDMAN: I think he is president of NAPHSIS and also Hawaii is really -- they have been collecting multiple race data for generations.
DR. MAYS: I think he is a "yes," if I remember correctly. The other pieces that we are trying to put in would be if we have time would be Suzanne Haynes in terms of the -- they actually from the Office of Women's Health have developed a system that they put into place for the collection of data on race and ethnicity by gender for all of the states. That is now on the NCHS web site. I guess she has a new one for -- that was for selected states and I guess now they are going to do it for all the states. So, I think it will be important to share it. Well, how did that happen.
They just did it. They got a contractor and they just did it. So, I think we thought that was kind of interesting.
The question of -- and I think since it is one day, I think the question of whether to do anything beyond states is that we won't have time. The issue of language, all this other stuff came up and it seems related but I think if we are doing one day, let's just keep it to the states.
So, that is where we are in terms of the planning of it. If Marjorie were here, she could tell you in more detail exactly who has committed for that. But again part of what we are trying to do is some of the state issues is we should also bring it back to some of the vital statistics issues, that sort a difference in terms of what we have heard so far have been about surveys and particularly population health. Vital statistics, they have issues and what is the other one that -- the reporting, the issue of your ability to collect data of racial/ethnic groups and the relationship of how it interfaces to the federal systems. I think that is the other sub issue.
DR. FRIEDMAN: Another related issue is the data collection schema for vitals as opposed to data collection schema for all the other data collection.
DR. MAYS: So, in terms of the others, we have to set a date and a time. I think that we have a report on the insular islands, which will help us a little bit in terms of the Asian, Native Hawaiian and other Pacific Islander hearing that I will send you by e-mail and ask you to look at because I think it will help us to formulate -- the language and translation issues hearing is still to come up. But I think before we schedule all this stuff, I really want us to get a sense of where we are going with the report and to pull the rest of the stuff together.
I really want to thank you for your hard work and trying to get it to us in time for these meetings because, you know, she is kind of starting from scratch in terms of learning it and identifying, you know, all the things that we have done and plowing through -- I think Debbie probably has, you know, dust on her clothes from pulling stuff out and finding things. Debbie has been very helpful, as well as Gracie and Audrey and Dale with helping us to find it.
So, some of these things, once we find them, I think we have quite a bit of work done. So, what I would like to do is to walk away from this with trying to see what we have, see what we can pull together and send this stuff to you and then in November to actually have an outline that we start thinking about what our report will look like so that we can move on from there, an outline and a time table.
Questions? Other suggestions, anything else you want to do?
MR. HITCHCOCK: The islands report dealt with a lot of infrastructure and training and distance related issues, which are not always the same as you will get from a hearing on health data needs for API for those who live in the continental U.S. As a matter of fact, the Department had a big API initiative not long ago. I don't think they even considered the people living on the islands.
DR. NEWACHECK: [Comment off microphone.]
DR. MAYS: As I read through it, I agree, that is what most of it was about. So, I think some of that we see where that has gone and then, you know, kind of balance it out.
Okay. I don't want to keep anybody. Thank you very much.
[Whereupon, at 4:40 p.m., the meeting was concluded.]