[This Transcript is Unedited]
Hubert H. Humphrey Building
Room 505A
200 Independence Avenue, S.W.
Washington, DC 20201
DR. LUMPKIN: Okay, we are starting a little bit late but we’re actually ahead of schedule. I believe that my guess is is that we can finish up most of our work this morning, we have a very crucial presentation at 1:00 on the National Child Health Longitudinal Study and depending upon how much work we complete we may or may not be done at the end of that presentation. So just to give you an idea, a heads up of what we’re looking at.
So the first item on the agenda is call to order, I forget now, do we do introductions again? Because we’re on the internet and they missed the introductions yesterday, so any of the folks who stayed over have just been waiting for this particular point in the meeting. My name is John Lumpkin, I’m senior vice president with the Robert Wood Johnson Foundation and I’m chair of the committee. Marjorie?
MS. GREENBERG: I’m Marjorie Greenberg from the National Center for Health Statistics, CDC, and executive secretary to the committee.
DR. MAYS: I’m Vickie Mays, University of California, Las Angeles, member of the committee.
MS. BEREK: I’m Judy Berek and I’m the liaison to the committee for the Centers for Medicare and Medicaid Services.
DR. STEINWACHS: I’m Don Steinwachs, Johns Hopkins University, member of the committee.
DR. HARDING: Richard Harding, University of South Carolina and a member of the committee.
MR. LENGERICH: Gene Lengerich, Penn State University, member of the committee.
DR. HUFF: Stan Huff, Intermountain Health Care and the University of Utah, member of the committee.
MR. LOCALIO: I’m Russell Localio, University of Pennsylvania, committee member.
MR. HUNGATE: Bob Hungate, Physician Patient Partnerships for Health, and member of the committee.
DR. STEINDEL: Steve Steindel, Centers for Disease Control and Prevention, liaison to the committee.
MR. BLAIR: Jeff Blair, Medical Records Institute, member of the committee.
MR. HOUSTON: John Houston, University of Pittsburgh Medical Center, member of the committee.
DR. ZUBELDIA: Kepa Zubeldia, Claredi Corporation, member of the committee.
DR. COHN: Simon Cohn, Kaiser Permanente, member of the committee.
MS. SQUIRE: Marietta Squire, CDC, NCHS, staff to the committee.
MS. SIGMAN-LUKE(?): Marilyn Sigman-Luke, AAHP/NHIAA.
MS. TRENTI(?): Nancy Trenti, American Psychiatric Association.
MS. CANAAN: Susan Canaan, writer to the committee.
MR. KILE(?): Frank Kile, American Dental Association.
MS. JONES: Katherine Jones, NCHS.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, committee staff.
MS. BROWN: Katherine Brown, Consumer Purchaser Disclosure Project.
MS. RAOUL(?): Patricia Raoul, American Nurses Association.
MS. PICKETT: Donna Pickett, National Center for Health Statistics and staff to the Subcommittee on Standards and Security.
MS. FRIEDMAN: Maria Friedman, CMS, and also staff to the Subcommittee on Standards and Security.
MS. PAISANO: Edna Paisano, Indian Health Service, staff to the Subcommittee on Populations.
DR. ORTIZ: Eduardo Ortiz from the Agency for Healthcare Research and Quality and staff to the Subcommittee on the NHII and the Quality Workgroup.
DR. EDINGER: Stan Edinger, Agency for Healthcare Research and Quality and staff to the Quality Workgroup.
MS. CLAUS(?): Linda Claus, American Health Information Management Association.
MR. RODEY(?): Dan Rodey, American Health Information Management Association.
DR. LUMPKIN: Okay, good. The first item on our agenda for today is the Quality Report.
Agenda Item: Quality Report – Mr. Hungate
MR. HUNGATE: Let me get this thing tuned up right and make sure I’ve got it, have I got it in the right place yet? Are we okay now? How’s that? There we go.
Okay, I’ve got good news and bad news, which do you want first? The good news is this presentation will be shorter then scheduled, and the bad news is that you have to listen to me personify things for a couple of minutes. I have found this report a very frustrating experience and I’m a fairly goal oriented accomplish oriented kind of an individual, so I have not enjoyed this. But I want to thank the committee for all the help that you’ve given, this is the third time this has come up for discussion, and the progress is probably not progress that you can really see but I think it’s significant. Because the thing that has happened is that I think our workgroup now really understands what it needs to do in terms of how it works within this full committee and what the function of our report is. So I’m first going to state that it is my full intention to take ownership of that report and have it ready for full approval by the March meeting. And I say that with some confidence because I think we’ve made real progress with your contribution and help.
Now I have to also make a personal confession that yesterday I wore my wreath fish tie so that I was of the swarm, now today I have my silver tip shark tie on, and so I’m serious today about getting this thing straightened out.
DR. LUMPKIN: I thought you wore that for the lawyers on the committee.
MR. HUNGATE: One of the things I do to get myself ready for presentations that I deem important, and this is one, is I try to get a good nights sleep and then I try to get a little exercise, so I went running this morning. We used to live in Washington for a while and we were over near the Potomac River, near the River Inn actually and so there’s a beautiful running route out along the Potomac River that takes you by the Kennedy Center, which was enshrouded in a little fog with dim light, it was beautiful, there were fish swimming out there in the river, rising to the surface, geese on a little farther, later on there was a cormorant eating the fish, and then I finally got around the Jefferson Memorial and around the Tidal Basin and then back to the hotel. And as I was doing this I was reminded of a quote that I saw on the ceiling of the Capitol building some years ago, a fellow named Henry Adams, and the quote was “Chaos is the law of nature, order the desire of man” and I thought about it and I said well that’s been our workgroup. And so I’m here to report on the apparent order that we’ve brought out of this chaos and we’ll go forward from there.
I want to thank all of you for working with our survey, we got a total of nine responses, including my own, Russ Localio, Peggy Handrich, Simon Cohn, Marjorie Greenberg, Steven Steindel, Mark Rothstein, who is not here, Steinwachs, and John Lumpkin all responded to the survey. The responses are interesting and helpful, perhaps in ways that you might think. There is some observer variability, it turned out that one of the questions appeared twice and the answers were not consistent from the same person.
PARTICIPANT: Bob, did you apply the right statistical test to this?
MR. HUNGATE: And so the objective nature which we attempted to achieve is somewhat subjective in its final analysis, but let me put it this way. Most people on most things said we ought to act sooner rather then later. Most people on most things said they were more rather then less important, the bias was all the things were important. And the bias generally was that they weren’t quite as hard as they were important. Now there were a few little exceptions where the hardness looked like it was bigger then the importance, and some of those are ones that have been on the agenda for quite a while and things that have been around.
It was interesting, too, the way in which the responses came to me was that I got most of the responses and worked through them and what I tried to do was put a boundary on the left side for the one that was clear over there with some initials and the boundary on the right side with some initials, so that if we got to the point of getting into a good discussion I could get a little debate going between those two points of view. Then I said well, let’s try to find a central point, what’s the center of this distribution, I put an X in that center and then that X I used to judge the relative group vote on the importance of that particular issue. And then I got John’s answers last and so I went back and copied the other information onto John’s, and I have to say John’s pretty consistent and his vote was more likely to be where the consensus was then off on the extreme and I thought well, that’s impressive. There were a few places where he put pole descend off at the edge but I won’t recount those, but they’re there.
Now the conclusion that really came out of all this survey and assessment fundamentally is that, two aspects to it and we talked about this at some length in the committee itself. As I’ve watched and learned how this committee operates I’ve seen the specificity and depth of the recommendations themselves and as we reviewed our recommendations when we compare those to what’s coming from the Privacy Subcommittee or the Populations Subcommittee or Standards and Security or the NHII, each of those has much greater depth then our recommendations had at this point. And so we really kind of faced two choices, one choice would be to say let’s get the recommendations fundamentally to come from the other committee rather then from us, that’s kind of punting in a way. Or we just only reflect those things that have already been decided to avoid the controversy. Because if there are basically 21 kinds of discussion items and the spread of opinion and I know the strengths with which opinions are held within the committee, the spread was such that I think there was not going to be committee time available to take it through a full vetting process of each of those recommendations.
So what we elected to do was move back, just step back a little bit from the recommendations themselves, move them up a level of specificity and I’ll go through that. Kathy Colton participated with us in the discussion, it was very helpful, and she had prepared a business case to take to the X-12 Subcommittee thanks to Kepa Zubeldia’s opening for us. We decided not to do that because there were personal conflicts with getting someone who could go and the specificity of the recommendations was beyond what we felt we really had vetted adequately in the discussion, so that we thought we’d be getting ahead of ourselves to do that.
So we’ve back up a little bit in a major part of the recommendations and said let’s take it to the level of what we found in what we heard and so let me run through the slides that Kathy put together for us here. Be prepared for some non-quality here because I have never used Power Point, I’m finding there are many things in my life that are different from the life in this committee, and that’s one of them.
I should recount a little tale about Power Point, I work with a Group Insurance Commission in Massachusetts and Delores Mitchell is the chairwoman, she’s a very effective speaker, absolutely refuses to use Power Point. I found a book that talks about how Power Point is moving this population to be unable to deal with complex problems, that in fact we’re so headline driven that we can’t deal with orders of complexity. And so if anybody’s interested in pursing that I can find you the name of the book.
DR. LUMPKIN: Can you give us a Power Point presentation on that?
MR. HUNGATE: So it’s a complex world. This is the charge that we started out with, identify issues and obstacles that hinder adoption of the President’s Advisory Commission on consumer protection, quality and health, you can read it. We held hearings over a long period of time, I believe that’s recounted on the next slide. Here’s where we took data from, quality oversight, federal state agencies, large employers, business coalitions, health plans, provider organizations, consumer advocacy, quality measurement experts, it’s a lot of opinion based on, and then we used the survey which we talked about to look at importance, urgency, and feasibility and I’ve kind of recounted the results of that.
What I take away from it is that the ratings of importance on all the recommendations were sufficiently strong that we shouldn’t back off any of them, all the things were important issues, and so we should simply find the way in which to articulate that the problem was brought to us and not necessarily be as definitive in all the specific answers. And you’ll see how we’ve dealt with that in the next.
It was I think John’s contribution that got us moving into a three category pattern, the first of which of those categories was to build on today’s administrative data and that’s what is shown here. We have an error on the third one under the data elements for risk adjustment and outcomes assessment in that that should be the secondary diagnosis on admission, so somehow its label got changed here and is different from what was in the questionnaire.
These are the things that we have identified as issues, gaps, and obstacles in the administrative information, the critical data gaps and administrative transactions.
Any questions there, you can read them, I won’t reread them for you, and let’s pause for a minute here, the four categories within this slide are data elements for risk adjustment and outcomes assessment, these are major categories, data elements for assessing timeliness of care, and data elements for attribution of accountability, the last one being data elements for assuring equity and assessing high risk populations for quality improvement.
For risk adjustment and outcomes assessment there are five, lab results for selected tests, an example might be A1C, hemoglobin A1C measurements for diabetics. Vital signs objective data for important health risks and outcomes. The third one is the secondary admission diagnosis, which is important risk adjusting for co-morbidity and also presence or absence of infection on admission, so there’s several things that relate to the quality of the management of the episode that should be indicated up front. Presence or absence of a do not resuscitate order is an important risk adjuster for mortality data in hospital admissions. Functional status is an issue the group has dealt with separately but is as yet unresolved, it’s an identified need and identified unanswered question. Data elements for assessing timeliness of care, episode start and end dates for global procedures codes, dates and times for admissions and procedures. There are a lot of things that are increasingly understood to be time relevant, the admission of antibiotics for a deep wound infection management is a well publicized example where the specific timing is critical to whether it’s right. And much of the pressure for these needs comes from the I’ll say broadly speaking the payer community which is part of where my experience comes from. Where there’s been a long standing wish to measure the quality of what’s done because there is real understanding that quality in the long run really does cost loess and if you get the right job done up front you’re going to be better off in the long run.
But unfortunately we’ve got a cost driven health care system where it’s so much easier to get the measure of cost then the measure of quality that we’re stuck and so there are an awful lot of decisions that are made on inadequate information. And the payer community has worked hard to try to generate the pieces of information that will enable that to change to a different pattern. Now you’re all familiar with the Leap Frog Group and groups like that. I think we may want to involve that community in a little more of our discussion to see what should go forward here but that’s part of what’s driving this.
The thing that interests me and why I’m part of the Quality Workgroup is that I think we need to change our health care system from a cost driven system to a quality driven system, that’s far easier said then done but you’ve got to start someplace, and if I take the optimistic chaos view theory that we’re the butterfly wing that might do it then why not start.
The next elements, data elements for attribution of accountability, operating physician for each procedure, for assuring equity and assessing high risk populations, race and ethnicity coded by OMB guidelines and primary language were all part.
Now the nature of our recommendations now will be to recommend fixing these, not how, and that’s the change that we’re making that is substantive in the recommendations. And I think it will make it easier, I don’t think that the ones we have identified are fundamentally controversial, it’s hard to fix them that’s controversial. Jeff?
MR. BLAIR: Should I save comments until you’re done or do you –-
MR. HUNGATE: I don’t mind, go ahead, go now, it’s better while it’s topical.
MR. BLAIR: The way you phrased it was that we should switch it from a cost driven to a quality driven and while I in large part agree with that because I think that there’s a lot of flaws, number one I agree with the fact that it has been cost driven, number two, or primarily cost driven, number two I also feel as if we have to put quality as a major area. I feel as if there’s an unnecessary conflict that we create if we say it’s one versus the other and I think it’s unnecessary from several standpoints. Number one is as we go forward moving health care into the information age, in many cases improving quality will reduce cost. And I think we ought to take advantage of that new paradigm that information technology offers us.
MR. HUNGATE: I don’t disagree with that observation. You’ve articulated part of my difference between Power Point and complexity. The label that I’m using of converting from cost to quality is a simplistic one, it’s a slogan, there are parts of it that I don’t like a lot because it tends to demean what’s been done because those that are working in the system believe that they’re providing quality. And so there are a lot of traps in that statement.
DR. COHN: I couldn’t decide whether to engage now or to let you go through some more –-
MR. HUNGATE: Now’s fine, I think now is fine, while it’s up let’s talk.
DR. COHN: Well, first of all, I was, as you know I had some real process issues related to coming to recommendations that had never been vetted in public or discussed in public, so I’m really delighted to have you try to raise it a notch.
MR. HUNGATE: We listened.
DR. COHN: I did appreciate that. I guess as I’m looking at these lists and I don’t know if we should be talking about the things on the list or not, and I guess I’m hearing that you’re going to get more public input on them, or are you?
MR. HUNGATE: It was not the intent to seek more public input before completing the report.
DR. COHN: Okay, well can we talk about that then for a minute?
MR. HUNGATE: Sure.
DR. COHN: I think on one hand, at least from my view, I enjoy that you raised it up a notch rather then getting into exactly how the things are going to be solved. However, and once again I’m not part of the Quality Workgroup, the only thing that I was party to was our larger conversations at the full committee –-
MR. HUNGATE: In effect the Quality Workgroup was the full committee.
DR. COHN: Well, in that case I didn’t miss anything. Only because in some of these cases I still feel that there needs to be some public input about whether or not we have the right things that we’re listing in the one through ten’s here. And for example I was actually pretty critical of item six, for example, only because every procedure has a global, every procedure, including I&D of cysts and abscesses, where I think the problem at least as I understood this one has been sort of a nagging issue by the quality folk related to post partum visits where there’s actually believe it or not a way to code that, that there was actually a visit that was actually, it’s part of the global but it’s a visit, you can actually do it. So I agree with the problem but I’m going geez, wouldn’t it be nice to hear from some people before we start going off and a solution that isn’t really even the right solution to the problem for example. And I also, even though I’m a strong believer in functional status I’m not sure what you would do with that up there, I thought it was sort of a research agenda but I’d be curious to hear what others think about that at this point. So I guess I was, if I had something like this I would be thinking that we should take it and have a hearing and ask people’s input on do we have the right top ten most wanted, if these aren’t right what are they, what are the issues that people are seeing.
MS. GREENBERG: Well I certainly can’t speak for Bob but as I recall our discussion yesterday, on the one hand all of these came out of a number of hearings and were distilled from a lot of public testimony going over a period of a number of years and I think there is a real desire to bring this to closure, this phase of the report. But one thing, and I think one thing that was discussed yesterday was A, bringing it up a notch so that it’s not as specific, and then using this report as a vehicle for identifying these areas, identifying the overall context and the importance of improving information for quality, but then having on hearings subsequently to address not only the overall list but specific parts of the list and maybe having that done by the full committee or individual subcommittees or the Quality Workgroup, it could be a variety of methods. But that I think the group didn’t feel that it necessarily needed more public input to say these are identified problems that we heard on a number of occasions but probably would need more, definitely need more input to come up with solutions and maybe teasing out some of the problems or even identifying additional problems.
DR. STEINDEL: I was going to comment very much similarly to what Marjorie just comment. I’m taking a look at this list and being one of the people that responded to the survey, I see some of the things on this list that I really didn’t feel should be on this list, probably the same reaction that you had, Simon. But my feeling is that it’s time to move this report forward and I like the way Bob has structured the way the report will move forward. And I think we’ve got enough public input. The committee has responded in at ad hoc fashion to prioritize what the list of what was it, 17 plus items, and what they would like the Workgroup for Quality to attack, and this is the response that has been distilled out of that, and I think they should move on to the next stage. And as Marjorie points out I see the stage after that is when public comment will be resolicited.
DR. COHN: I guess we’re certainly allowed to have differences of opinion in this committee, it would be very boring if we all agreed with things. I guess once again my recollection of the testimony over the many years was that we heard extensive from quality experts and research all of whom wanted data. I don’t ever remember an occasion, and you should never use the word never, in my organization I never use the word always or never and so I would certainly not use it here, but I don’t remember in my recollection hearing anybody with a different perspective, such as somebody who might for example have to supply data, one who might be actually held accountable or any other part of the health care organization to comment about whether, I mean just to even understand what their perspectives were on all this. And that’s really what I was thinking about, not going back to the quality experts –-
MR. HUNGATE: And I’m not uncomfortable with that observation, I think it’s valid with respect to what’s there. The point that I would make is that it represents some portion of the payer community, Artie Millstein from Pacific Business Group on Health, Leap Frog folk, there was input from that community, not from the community that would respond, that step has not been taken. And so these are the viewpoints of the, I like to call it the force vector, from the purchaser side that’s trying to change the way in which we spend our dollars in health care. Now that creates a lot of tensions, we haven’t dealt with those tensions. At least that’s my perception.
DR. ZUBELDIA: I’m having trouble with this whole concept and discussion because of the variable target. I’m not sure the target is quality of care, quality of data, or quality of payment. And it seems that the target may be quality or perhaps precision of payment. The last time I checked payers typically are not the ones that contribute the data to public data repositories for analysis, it’s providers. And tagging the claims and the transactions that go the payer with this additional data elements I don’t think would help in the collection of data by data repositories unless there is a mechanism set up first to harvest the data that goes to the payers, which is not in place today and I haven’t seen this part of the discussion. So I’m not surprised that the payers want to have more quality of the data so they can make more precise payments, but I’m not sure that that was the goal of this exercise. And so I think there’s a big piece missing here and I’m having trouble putting the pieces together.
DR. LUMPKIN: Let me perhaps respond and just sort of revisit the vision that we did yesterday. First of all just ten days ago as part of a project that we’re working on at the Foundation I spent the day at the small community hospital in Princeton, University Medical Center of Princeton, and they’ve actually had a rudimentary electronic health record in place for over five years, they’re now going into their second phase. Many of those items that are on there actually are currently captured as part of their process of taking care and I was watching patients in an intensive care unit following a nurse around on that episode. And so as we’re looking at this process that the quality team is focusing on, and I think they’re focusing in on quality of care, our recommendations shouldn’t be seen at what the current activities are in health care but in the future that we’re moving towards. So that as Kaiser is implementing their new information system what are the measures that are going to be, the quality important data, and what I would conceptualize the committee’s report is really talking about then three approaches to capturing that data and using it to measure and to influence quality as well as feedback for quality improvement.
And those approaches are adding on to the current data stream, which is administrative data. Second is creating a new data stream, which could be something along the lines of a transaction, a quality transaction, that would be a separate set of data interchange. And then the third is new technology, which is electronic health records. And rather then coming up with a prescriptive report which says this is the only way to fix this is to add on and change the conditional to required on the 837, we say you can do this by getting this data out of the report.
Now there’s some additional work we’re going to have to do afterwards but it sorts of sets a transition, a comma, for the work of the committee and this report becomes that comma and then afterwards our work will be trying to flesh out some of those pieces. So within that concept I think that it actually can be written in such a way that it focuses on quality, it looks at data, looks at alternative for collecting, ways of collecting data without being prescriptive, recognizing that if we were to do it today the burden of data collection would be huge, but in the future that we think we’re moving to and moving towards rapidly the cost of collecting some of that data is going to drop dramatically.
DR. ZUBELDIA: Can I respond to that? Yesterday you mentioned that when you have a hammer in your hand everything starts to look like a nail, I think –
DR. LUMPKIN: It’s not fair to use my analogies against me.
DR. ZUBELDIA: The fact that there is an 837 transaction used for payment doesn’t mean that everything has to be fit into the 837 transaction used for payment. There’s also an 837 transaction used for data reporting and if you’re looking at quality of data reporting that quality transaction that you have on your slide already exists. And there is an implementation guide for it and you’re going to get the data for that transaction a lot easier then you’re going to get the data for the payment claim modified, because the payment claim is coming from a mysterious system in the back of the office that doesn’t have access to the clinical information. But that mid term transaction that you have there, quality transaction, that already exists. It’s an 837 and there is an implementation guide for it that has come out of X-12 specifically for quality reporting using the 837. So I think that should be the target of your efforts and that’s a transaction that can be produced directly from the clinical systems without being involved in the payment transaction.
DR. HUFF: I liked your characterization, John, where you said we could use administrative transactions, create a new data stream, because I can one better you on the transaction, I mean this lab data that you want is already flowing in HL7 transactions, so it has a standard defined as well and so I think one of the important, that’s one of the things that I noted here, is there seems to be a presumption of the administrative transactions being used to do that. I would be more in favor of saying let’s create a new data stream and whether it’s X-12 or whether it’s HL7 or something else, but I mean I think that’s the detail of issue we need to talk about because I can tell you that from my institution it’s almost zero cost for me to split my lab data stream, which is already flowing, and send somebody an electronic copy of that, it would be much more costly to take that HL7 information and transform it into an X-12 message and then forward it to somebody. I mean that’s the kind of discussion, that’s my position, I’m not saying that’s necessarily true, but that’s the kind of discussion that needs to go on to say what’s the best way to do this.
DR. ZUBELDIA: Stan, I fully agree with you, I think this data is in the clinical systems today along with essentially all the demographic data that you need and it’s easier to convert it into HL7 out of a clinical system then to try to bridge this into the payment systems that don’t have any of this data and modify the payment systems to carry it into a payer that is not involved in the data collection process that you have today.
MS. BEREK: In CMS for home health agencies and nursing homes we actually have the data system that measures quality and is used for payment. And although it is incredibly controversial and people hate it, watch them shake their heads, they hate it, it does both of those things. And it has enabled to do all kinds of things that are incredibly valuable. So I think it is quite possible, and I will go back to my childhood working in the State Department and Social Services, where we tried to get information that we thought would be interesting on the welfare enrollment forms, and we gradually discovered that if it didn’t impact on payment it never went in correctly, and we could not get people to give us the kinds of information that we just wanted if it didn’t in fact impact on how much we were going to pay for somebody’s welfare grant. They’d tell us the number of kids they had because it impacted on the welfare grant but if we wanted to created race and ethnicity data, this was a long time ago, hold your breath until you got accurate data because people just didn’t care to fill it in.
So I think the linking sometimes, the things you need for quality to the way you pay, sometimes helps getting what you need but having said this I also thought I heard Bob say that there is no question we need to have a long involved very publicly inclusive discussion on this subject and all you’re asking us to do is say these are the topics that we will base this long involved discussion on.
MR. HUNGATE: That’s accurately stated. Does that explain, does it solve some of the discomfort?
DR. COHN: Are you looking at me?
MR. HUNGATE: Yes.
DR. LUMPKIN: Let me modify your question, which is, it’s hard to answer that question without seeing a report and how we handle it, I think we kind of given a conceptual framework for the report, which is up there, the different components of ways to address the issue, and then use that as a guide to the future work of the subcommittee. Getting the report out, expressing what we’ve heard, some of the potential solutions and then this really sets the agenda for the subcommittee. Jeff.
MR. BLAIR: Bob, forgive me if my comments wind up being things that you may have already included in the report, it’s just that I haven’t heard them so far. But my thought, I think that the work that the workgroup is doing, is it a workgroup or subcommittee, I really feel like it’s extremely important and over time it will become more important. The thoughts that I have are that the world is changing very quickly and there’s two aspects that I haven’t heard expressed or maybe I missed them if they were. One is the clear definition of quality and maybe for the first time in our history we could come up with a definition of quality of health at the level of public health and have that drive things. For example, disparities in health by race and ethnicity instead of, that could be considered one of public policy’s measures of quality of health. You drill that down and you wind up saying are we capturing appropriate data within clinical systems for patient care and then you wind up saying how do I create the infrastructure for that in terms of information technology to do that without disrupting patient care as the number one priority.
So my thinking is that number one I think that this reaches right up to basic public policy and that we define quality broadly so that it includes costs, it includes responsiveness, it includes access to care, it includes, as well as patient care that’s effective and efficient and that when we look at this we’re at a stage for the first time when we can say that information infrastructure, that National Health Information, somebody used that word, that National Health Information Infrastructure can do this in a manner that facilitates continuous quality improvements.
So I don’t think we should wind up setting a bar, I don’t think that should be our role where we wind up saying nosocomial infections should be at this particular rate. I think what we should be doing is including in that infrastructure the ability to measure nosocomial infections in an automated fashion where we can facilitate continuous quality improvement. So I think that you’ve taken some major steps forward in assimilating all these years of work, I think you’ve done great work in doing this and I just think that we have an opportunity that I don’t want to miss to start defining quality broadly and on a continuous basis and not narrowly. Thank you.
MR. SCANLON: Jeff, I think this basic process of defining quality and measuring steps, there are already a number of other activities underway and I think rather then have the committee start a new one I think the committee may want to advise on some of those others and let me give you two examples. Legislation about two years requires AHRQ to develop a National Health Care Quality Report and that’s coming, there’s a draft and it’s coming through the process now. They had to go and look at all of the data sources, all the conceptualization of how do you measure quality, how do you define it, and the IOM in fact, they enlisted the IOM and others to help as well, so I wouldn’t want us here to sort of start another process, I’d rather that we build upon and advise that process.
Similarly in the disparities area there are probably half a dozen major strategic types of plans and reports that try to measure disparities, again AHRQ has one of them, and again, I think this is probably a role where the committee here can advise on sort of the current state and where the improvements might be made but I don’t think we want, I just don’t see it being that efficient to sort of start off in a new direction here.
The other thing Bob is it looks like these are all hospital based, the focus on these measures is largely hospital based, was that for –-
MR. HUNGATE: I think that’s true, it was not a conscious choice, I think it’s what evolved from what was before the group. I’m a little hesitant to say too much because I’m kind of coming after the fact.
DR. COHN: First of all, Bob, I do appreciate your, I mean I think there’s certainly a desire to get this done with and get on to the next thing so I do appreciate sort of the general view of some way to turn this into a workable product. I’m still sort of trying to figure out what makes the most sense here, I like John’s, and in your later overheads which we of course have already looked at quickly because we’ve been talking about them, you have a high level conceptualizations I think are very comfortable and I certainly don’t think that there needs to be any additional hearings or comments about that. However, I do sort of come back to that slide about the ten most wanted critical data gaps and administrative transactions as even though they are intended to be high level they get relatively specific and obviously Jim you were sort of commented, they either were hospital based or not hospital based or whatever, and I guess I’m wondering if this, given the nature of the testimony, the long timeframe we had, whether even talking about the top ten as more then examples and I might make an exception to that –-
MR. HUNGATE: Rephrase it as examples –-
DR. COHN: Well, I’m just sort of wondering because I mean when you get to this level of specificity as well as defining administrative transactions and not intend to ask the industry are these the right ones it seems to me to be a little cavalier. On the other hand I think as examples of the larger bullets, I mean the larger bullets I think are certainly things that we all agree that we need and it seems to me that as you move further on you get into some of the longer term larger issues which the committee is very good at. It’s just a thought on my –-
MR. HUNGATE: I appreciate that –-
MS. GREENBERG: I was just going to follow up on that, I think if you look at the four black bullets, the bold bullets, I think the workgroup and I guess the question is also would it be the committee, but could say these are the areas that were identified as critical gaps, data elements for risk adjustment, data elements for assessing timeliness, attribution of accountability, assuring equity, and assessing high risk populations. These are examples, so that picks up on, but the bold ones aren’t so much examples, they are sort of the critical areas and then the ones underneath are examples.
DR. COHN: I would certainly feel much more comfortable with that sort of a perspective on this.
MR. SCANLON: And I think the other way to proceed along the same lines, this is almost in the nature of a straw man where we don’t have unanimity necessarily on the specifics but this is the way the thinking was emerging and it could actually use the report for further dialogue. But again, they’re quite specific, it would be hard, it may be a straw man kind of an approach rather then a specific set of recommendations.
MR. HUNGATE: Kind of a status report in a way.
MR. SCANLON: Well, it’s more then that, it documents everything that preceded it and it identifies the conceptual and the directions but it’s the specifics that probably need more reaction.
MR. HUNGATE: Well the committee is taking upon itself to, I think we’ve been depending heavily on this dialogue and the transition that we made yesterday was to say it’s now up to us to take all this input, sit down, do the work, listen and do the best job we can of reflecting, and we expect to do that in the January timeframe.
DR. ORTIZ: Eduardo Ortiz with the workgroup. I just want to make a comment, too, that at least from, and I was very involved with Bob and Debbie and everybody in doing the survey, when we got these results back I didn’t see this and I don’t think any of us thought of this in terms of in the report itself being kind of a prioritization of top ten, that these are more important then others. What we really wanted was as we were thinking about further steps and where do we go next with the committee, we were hoping that as you gave us these priorities, because we asked importance, we asked for timeliness, it is something that’s going to be done now or in the future, and feasibility, we were trying to use all those numbers, all that information to try and guide us in saying okay, now we’ve got all these things, where should we channel our next efforts.
So I think that might address a little bit of Simon’s concerns, that we weren’t trying to do this in terms of the report, if we put this report out that these are the most important things, it was more like saying to us hmm, if there’s two or three things we can do next where should we channel our efforts. But that’s something that now we’ve also even gone back to the table and looked at but it was never intended, at least from what I thought, to be something that we were prioritizing in the report, that this is more important then something else.
DR. LUMPKIN: Okay, Kepa?
DR. ZUBELDIA: I have a question. So is the report going to be redrafted from scratch? Because the draft of the report that I’ve seen has some very prescriptive changes to the transactions and it’s extremely prescriptive –-
MR. HUNGATE: That will change a whole bunch.
DR. LUMPKIN: Yeah, the report has really two basic parts, there is the description of the process, the background, and then there’s the recommendations, and the recommendations will be dramatically redone between now and the next iteration.
MS. GREENBERG: And actually, Kepa, I think just from what a conversation I had with you and Michelle Williamson on my staff who participates in X-12 and is currently on jury duty, a bug that’s hitting a number of us on my staff, was that you felt taking these as identified gaps, problems, issues, to X-12 would be much more effective then saying change this condition, or being very prescriptive, so I think that certainly contributed to these discussions.
DR. ZUBELDIA: I think that’s a very important part of the process, let X-12 determine how to do it, just tell them what you need done and let them determine how to do it. But also I think that the feedback that you’re going to get from X-12 is probably going to be along the lines that this is great but it’s not for the 837 for payment, it’s an 837 for data reporting.
MS. GREENBERG: Well, that’s the kind of dialogue that should take place.
MR. HUNGATE: Alright, have we finished the subject? Is that an adequate discussion?
DR. LUMPKIN: I think so, I think we’re great.
MS. GREENBERG: I would just suggest that, I think is a nice presentation even if maybe some of the headings need to be changed, but that people just be asked to look at this as the group is working concurrently on the report and give Bob and Eduardo any feedback on the things in this slide presentation.
MR. HUNGATE: That would be very much appreciated. Thank you. I didn’t save quite as much time as I wanted.
DR. LUMPKIN: Good discussion and we gave some of your time to the Populations Committee, so that’s okay.
Agenda Item: Executive Subcommittee Agenda – Issues to be Covered – Dr. Lumpkin
DR. LUMPKIN: Next item on the agenda is the Executive Subcommittee agenda, items to be covered in our retreat on the 21st, let me just briefly go over the issues that the committee will be addressing, and I don’t believe this is in your, no, it’s not. So the first issue is a review of the work of the committee, where are we going, how do we get there, and what happens after the recommendations, we’ll review that process. Then we’ll engage in a strategic planning process to address upcoming issues and it would be upcoming issues, budget considerations, what’s happening with Health Statistics for the 21st Century.
The next phase would then be doing current operations and future expectations. One of the issues, as we’ve done in each executive subcommittee retreat is we look at the function of the committee interplay of the various subcommittees. One of the areas just so everyone will know that will be coming up will be the issue of our we maximize our representation in both the councils that we’re invited to participate in, both the Data Council and the Council on Application Health Information Technology.
Then we’ll have a discussion on coordination with the National Center for Health Statistics Board of Scientific Counselors and then going through the plans of the subcommittees and workgroups, both the accomplishments and work plans followed by some discussion on following up on the discussion we just had now on the Workgroup on Quality, we’ll spend a little bit of time getting input from the various subcommittee chairs on whether or not there are pieces of the follow-up to that report, and then a review of the upcoming meetings. So that’s a short synopsis of what will be occurring at the retreat and we’ll have a report, if there are questions or other items that you believe ought to be on the agenda, open to suggestions.
MR. SCANLON: Possibilities of new members.
DR. LUMPKIN: Oh, yes.
MS. GREENBERG: Hope springs eternal.
DR. LUMPKIN: Well, I think that that’s an important point and under operations we should probably, there’s a potential to actually have a fairly large number of new members so probably we should have some discussion about getting them up to speed. Because there will be some major discussions at the March committee, the Quality Report and we again have the potential of having a committee that didn’t participate in the hearings trying to pass a report. Richard?
DR. HARDING: Jim took it right out of my mouth, I was thinking more the words succession, with a large number of people rotating off over the next year, too, and important positions on the committee and how they can be trained up to be ready for that, because all committees like this have to have a process of training.
DR. LUMPKIN: Training and succession. I like that. Okay, Jim?
MR. SCANLON: I would ask in terms of resources for the, this is fiscal year ’04 for the committee, we’re planning to have from the Department the same amount of resources for analytic work and other work of the committee as last year, but if any of the workgroups or subcommittees is envisioning a larger effort that would involve contractors or consultants we should talk about that at the meeting as well so we can plan, that requires separate planning.
MS. GREENBERG: We had several really substantial projects this year which kind of used up our reserves, so that’s important for our planning.
DR. LUMPKIN: Okay, then we’re going to move on, we’re going to move into some of the items from the reports from subcommittees and workgroups, I think I just gave you the one for the executive subcommittee.
Agenda Item: Reports from Subcommittees and Workgroups – NHII Workgroup – Dr. Lumpkin
DR. LUMPKIN: The NHII workgroup, we held our meeting yesterday, we have decided that we wanted to take a full day of meeting of just the workgroup to assess where we’re at and where we’re going. We’ve pretty much begun to work through an agenda for that day, we’re going to be looking at as one of, probably the day that popped up on most people’s calendars yesterday was the 19th of December, which is kind of a short turnaround time but January and February in most people’s calendars with all the various groups seemed to a little bit difficult to try and schedule.
The agenda would include an opening evaluation of the recommendations from our 2001 report to determine where they’re pertinent, some updates from staff from the Department of Health and Human Services, also from CHI or other related kind of input. And then an opening discussion about where it is that we believe as a workgroup and also as a committee our input will best leverage the resources that we have to continue to move the agenda forward. And so that should allow us to get moving again on NHII.
It’s been very exciting with the workgroup because on one hand every time we talk about an area that needs work somebody else takes it up and that’s great but we felt we needed to do a more strategic planning session. Any questions on NHII Workgroup?
Okay, Simon, action items, done.
DR. COHN: We’re retiring. Do you want me to talk about the upcoming hearings we have?
DR. LUMPKIN: Please.
DR. COHN: Okay, well thank God we don’t have any more action items. We have hearings scheduled for the subcommittee on the 9th and 10th, the main topic is going to be CHI and we’re obviously waiting to really see how many of these final recommendations they really come forward with but we will make whatever time necessary to deal with that. We’re also planning on having a session on claims attachment, the current status and preparation for an NPRM which we believe will be coming out sometime in 2004, or actually hopefully I guess in the spring of 2004 as we heard yesterday. So this is sort of a refresher and update as well as to see if the industry has any issues that we should be aware of.
Now in January we have a hearing, a session on the 27th and 28th, that will be devoted to hearing any other CHI items that may be coming forward such as needed. We’ll also be meeting with the DSMOs to sort of review the current status of the HIPAA standards and recommendations for changes to that. I suspect that that will also evolve into a beginning conversation about lessons learned around HIPAA.
I think we also, oh I’m sorry, thank you, we have a half day scheduled on the security final rule to sort of see what issues that the industry is beginning to identify or if they’re having any problems with the implementation.
MS. GREENBERG: That’s January?
DR. COHN: That’s January, and I think that that’s basically it for the January hearings. There may be a couple of hours on dental content issues and that’s being reserved also.
DR. LUMPKIN: Just to cap off the meeting.
DR. COHN: Yeah, we wanted to fill it up, John.
MS. GREENBERG: Wouldn’t want to brush those aside.
DR. ZUBELDIA: I would like to ask Marjorie or Jim that given the interest of the industry in the ICD-9 letter, which is one of the action items, that it posted be to the website as soon as possible, ICD-10 letter.
MS. GREENBERG: Actually we had some discussion about this at dinner last night and because may not have the final final RAND report, which is an attachment, immediately because of some issues about, we had held them off on finalizing because we thought we might need to incorporate information from the hearings and we talked about that, I would suggest that we, because I do think we should post the letter immediately, I mean first we have to send it to the Secretary, we never post anything until after it has been received by the Secretary, we don’t want someone calling the Secretary’s office and he hasn’t even received the letter yet, but that will happen very quickly, that perhaps we could attach the pre-final of the RAND report, Donna, does that seem reasonable? With a note that when the final is available, this is being finalized and it will be substituted, because I don’t want to hold up the letter for that.
MR. SCANLON: We would literally want to make sure it is in the Secretary’s office before we post it but otherwise we can do it quickly.
DR. LUMPKIN: So that will be sent out very soon.
MS. GREENBERG: Today’s Thursday, so we think it will get sent tomorrow.
MR. SCANLON: We could probably get it up tomorrow then.
DR. LUMPKIN: Okay, so tomorrow and posted as soon as we get the return receipt.
Anything else on Standards and Security? Privacy and Confidentiality.
DR. HARDING: I was asked to, this is Richard Harding, I was asked to stand in for Mark Rothstein who is absent today and sent his apologies to the committee. We met this morning from 8:30 to about 9:45 and had probably three issues to bring to the group, no action items to report.
One is that we have privacy hearings coming up November 19 and 20 in Silver Spring, the privacy hearings, as you remember the privacy rule was brought forward in April of ’03 and the committee came up with, the goal of the committee from this point forward to track and consider implementation issues of the privacy rule, so we’re going to have testimony around the areas of public health, health science research, and then an open session to bring up issues that we feel will help us distill the questions that we need, to ask the proper questions to monitor and check the implementation of the rule as it goes forward. And we’re looking forward to that and we have a list of people who are in the final stages of accepting their testimony and we will get that out to the whole committee.
We also have a provisional schedule of February 3 and 4, which seems to have some difficulties with some of the members, Mark set this up, where we will have additional testimony and the topics of that one will be schools, which involves minors and so forth and the privacy regulations, law enforcement and payment chain mechanisms. Those are the two that have been scheduled, we hope to have one more during the year, continue the implementation monitoring of HIPAA privacy standards.
Two other issues and I’ll stop, one is that we discussed a letter that was sent to the editor of the American Medical News from two physicians making the statement that they would suggest that individual physicians who are in small practices of under ten employees opt out of HIPAA, and that they suggested that there was an ability to do that. We had experts, including Stanley Knockensome(?), who was on the phone with us and who in effect said that once in at the present time there is no mechanism for opting out of HIPAA. If you have never been in, of course, and you’ve never in effect done anything that gets you into the HIPAA transaction channel, whatever the term is, that’s a different issue and people of course aren’t opting out because they never opted in. But once you are in the system at the present time there is no legislation or rules that allow for opting out.
Is that clear the way I said it or do people have questions about that?
DR. LUMPKIN: Actually I have a question, are we talking about the transaction piece or the privacy piece?
DR. HARDING: The privacy piece. That you get into the, the transactions are what gets you into the whole thing.
MR. HOUSTON: The substance of the letter was that, the thought was that if the physician practice had less then ten FTE’s that there was this ability in total that you could one, not have to electronically transact Medicare claims and therefore if you did not electronically transact claims that you did not have to, the privacy rule did not apply and that you could then opt out after the fact from having to comply with the privacy rule. And the response was is that once you have decided to comply with the privacy rule there was not the ability, or the thought was there was not the ability after the fact then to opt out. But that in fact if you did not electronically transact claims pursuant to the transaction standards that HIPAA simply did not apply so it was correct insofar as that was the case but otherwise there was not this provision to opt out after the fact. Am I characterizing that correctly?
DR. HARDING: I think that’s okay.
MR. SCANLON: This is a legal question and I would just assume, I don’t think we need hearings, it’s either legal or it’s not, we can have our lawyers look at it.
DR. HARDING: And Stephanie Kaminsky was here and heard -–
MR. SCANLON: We may need a question on the website or something.
DR. HARDING: Stephanie said this was on their radar screen and that they were looking into it but they did not have a FAQ on that specific issue.
MR. HOUSTON: No, there was a FAQ on that issue, I thought the point was is that there was an FAQ that indicated once you’d opted in, if I’m not mistaken, did the FAQ say that you could after the fact opt out?
DR. HARDING: She stated that, I don’t know if it’s on the website.
MR. HOUSTON: I think it was ready by –-
DR. HUFF: It implied it but it didn’t state it, it didn’t state it implicitly the way you’ve stated it.
DR. HARDING: You can see that it’s not something that would particularly be helpful to have that rumor going around that it’s an option and so forth, so we wanted clarification and to try to get that out as soon as possible so that it’s clear one way or the other.
That was one issue, the other issue was Kepa brought up and I will speak for him, the issue of Gramm-Leach-Bliley and how the potential for data mining in transactions that occur in the banking industry around medical information that comes through the 835 forms, which is a lot more then I know about the forms, but in that information is medical information as well as banking information and that there is the possibility, it’s been suggested that some of the medical information has been mined and used and we were talking again, this is a very complex and technical electronic transaction and he was bringing this up and Stephanie who was here from OCR and that they have looked at that some, too. Kepa, did you want to say anything more about that?
DR. ZUBELDIA: They’re still looking at it to identify to what extent Gramm-Leach-Bliley covers or doesn’t cover those transactions that are exemption by 1179 in HIPAA.
MR. SCANLON: The Department has received letters from the American Banking Association and others raising these very issues, and at the same time the Georgetown Health Policy Project has sent us letters reacting to those as well, so our crack legal team is looking at these things, it’s a very complicated issue.
DR. HARDING: Yes, with Dan’s help one thing that was clarified for me is within this 835 form is not diagnostic criteria but treatment criteria, which could be used of course, you’re being treated for certain things that would show up and it would be something where data could be mined.
With that, that concludes the report.
DR. LUMPKIN: Let me just sort of toss one other thing and see if there was a discussion, the issue of out of country medical records has reared its head again, particularly as it relates to transcription services that are done off shore, and I don’t know if the subcommittee has looked at that but it seems like its –-
DR. HARDING: We did not this morning, you’re bringing up the issue where a transcription service in the United States had a contract with a group of transcribers in Pakistan and that some of the information of that individual, an individual received a message from a Pakistani transcriber saying I have information on you and if the people who I transcribe for don’t pay their bills I’m going to release it onto the internet. And this occurred two weeks ago and it was reported in the San Francisco Chronicle in San Francisco.
MR. BLAIR: Could I suggest that the issue is actually a little bit broader then just whether it’s done overseas? When I read the article it appeared to me that a large part of the issue was that if a health care provider contracts with a business associate and does not know that that business associate, which they may be dealing with for many years and has trusted, is then subcontracting, in this case I think it was subcontracted four iterations in a row before it got overseas, and the provider wasn’t aware of the fact that it was subcontracted once let alone four times. So I think that the subcontracting issue has to be dealt with as well as whether it’s overseas or not.
DR. HARDING: Good point. We didn’t discuss that this morning.
DR. LUMPKIN: The subcommittee obviously has a full agenda but that seems like an issue that comes up periodically and probably is worthy of taking a look at. John?
MR. HOUSTON: It’s interesting, too, because I don’t think it’s just transcription services are going off shore, you hear about radiology services and other types of telemedicine services that end up in Australia and other countries and it does raise an interesting point and I know, I’ve gotten involved from a credentialing and the licensure issue but I think there are privacy implications that are equally important to consider overall, so it might be interesting to delve into this one.
MR. BLAIR: This isn’t meant to be funny but if I recall, seriously, apparently the health care provider in San Francisco subscribed to somebody they had been working with for a long time, that then did a subcontract to another medical transcription service in Florida that did another one to a medical transcription service in Texas, and it was from there that it went overseas. And the original provider had no idea that all of these subcontracting activities were happening, so that’s I think the focus.
MR. HOUSTON: I’d hate to think of the quality implications of that, too, you have no way to do quality control in those type of cases.
DR. HARDING: Which brings up a whole issue of who’s responsible for those subcontracts, and if the physician says that he does not want to be in HIPAA because they’re going to do just paper billing and so forth and gives it to someone else who electrifies it or sends it by electronic means, is it his obligations to know that? Is that a business, what if he doesn’t and that happens? It’s a conundrum as always with privacy and we’ll be working on that.
DR. LUMPKIN: Great, any other –-
MS. BEREK: I just have a question, have you ever looked at some of the things that have been going on on identity theft relating to health care? Because that, I mean they steal all kinds of stuff.
DR. HARDING: It has not be on our –-
MS. BEREK: It may at some point.
DR. HARDING: It should be.
MS. BEREK: One of my staff does the testifying on all the cases.
MR. HOUSTON: What might be interesting as an offshoot to that is the whole issue, and we talked about this last evening, use of having the common identifier, and I know that that’s sort of out of the purview of this committee right now but the whole issue using Soc numbers and the like is, there’s no preclusion from the hospital using a Soc number, it’s my understanding, as a medical records number and it’s a common practice and it’s a way to identify individuals uniquely but the whole issue of use of Soc numbers in health cares is a surrogate for having a national identifier is one that maybe we don’t want to delve into but it all goes back, that’s sort of the source of identity theft I think.
DR. ZUBELDIA: There is identity theft both for patients and providers –-
MS. BEREK: And providers, oh yeah, most of the big cases expose the doctor and the patient.
DR. ZUBELDIA: The news in the last few weeks reported a person that changed his legal name to be the same name as a provider and was caught and put in jail six times for impersonating a physician. You have to be very persistent.
DR. HARDING: Jim do you know anything, just one more question, about the hold on the development of the unique individual health identifiers that was put in by the Clinton Administration?
MR. SCANLON: But it was also seconded by Congress, the prohibition currently is on the budget and for the past three years in the HHS appropriations it basically prohibited us from spending any of the funds that were appropriated for the development or the promulgation of such an identifier.
DR. HARDING: And that’s still –-
MR. SCANLON: We don’t have ’04 yet, we don’t actually have an ’04 budget yet, but we’ll monitor that.
MR. HOUSTON: A follow-up question related, is there any prohibition against us trying to understand the issue of identity theft and the use of Social Security numbers and its impact on privacy?
DR. LUMPKIN: Well, it just all depends on who you ask and if you ask me I’d say no, so we don’t need to go any further.
MR. SCANLON: It sort of depends on how you present it.
DR. STEINDEL: Just very quickly, we actually did have a hearing in an NHII Workgroup session from the banking industry and that included issues on identity theft, so it has been brought up in front of the group.
MR. SCANLON: About the hearing in November, the focus is on public health, the impact of the reg on public health and health services research?
DR. HARDING: Yes, and we have an open session that’s the morning before the executive committee, so we’re going to meet in the morning and go up to Jersey.
DR. LUMPKIN: Okay, great. Populations?
Agenda Item: Reports from Subcommittees and Workgroups – Subcommittee on Populations – Dr. Mays
DR. MAYS: Populations also met this morning from 8:30 until a little bit after 10:00, thank you for that extra time, because our agenda was already full and then we had things added to our agenda, so we were appreciative of the extra minutes that we took.
One of the things Populations is trying to do is follow-up on the two letters that were passed here in September. One was the letter on the targeted surveys in which we were asking the Secretary to try and get an increase in the number of target surveys, particularly for racial and ethnic minority groups that we don’t have a lot of data on and to try and think about the issue of increasing access to the data that’s available. One of the things that we’ve done is as you know one of the liaisons to the committee is Virginia Cane with NIH, so NIH is going to pull together a planning group and think about a workshop to really deal with some of the methodological issues in terms of targeted surveys. That will be a broad meeting, the planning committee will get broad input from across not only NCHS but a variety of other groups and I think in terms of the workshop that’s being planned it will be a small workshop, so when I say small I think they’re thinking like under 100 as opposed to the thousands that when the Secretary calls a meeting there are thousands of people at some of these summits. So that’s just getting underway, it hasn’t been appointed, nothing at this point has been done. We talked first at our meeting in order to give her some input to think about what the scope of that particular planning group would be in terms of holding a meeting, so that now has moved back to NIH.
The second letter that we did, which was that encouraging the health plans to collect data on race and ethnicity, one of the things that would be useful is for us to send a copy of the letter to the various health plans, and when I brought that up I learned that that’s not a simple task, there’s hundreds of health plans, so I think the thinking is that we will send it to some of the major health plans as well as making sure that it goes to the trade associations. So we will actually work on that, which is a bigger task then one realizes.
DR. LUMPKIN: But just one that gets smaller and smaller every day, as they eat up each other.
DR. MAYS: Maybe if we wait a little bit it might be an easier task.
DR. LUMPKIN: Save on postage.
DR. MAYS: Exactly. So that’s what we’re going to do to try and follow-up on that as a way of encouraging the health plans to realize what it is that we’ve asked the Secretary to do.
We also are having a hearing November 13th and 14th in San Francisco, and the reason we’re having it in San Francisco is because, well, for two reasons, one it’s right before APHA and we will be drawing some of the people who are attending APHA, but also it’s a hearing in which we are doing a follow-up to the hearing we had in May in Las Angeles with the Asian, Native Hawaiian and other Pacific Islanders. Again, the questions, we have a set of questions and they should be up on the website probably soon, a set of questions in which we’re trying to determine again about issues in terms of the collection of data on race and ethnicity, specifically for the diversity that exists in this population, we’re trying to understand a little bit about how the OMB directive is working. And also some sense of what the variables are that are important relative to health disparities for these populations other then just the collection of race.
It was suggested and I thought it was, it was Don that suggested it and I thought it was a very good suggestion because it came up yesterday about Health, US, is that we also start talking to some of our hearing participants about what kind of data they would like to see. In a sense we’re being asked ways in which to enhance Health, US in terms of the way in which it’s presented and its usability, it might be interesting to ask these questions while we’re having the hearing. So I think we’ll put that also on our list of questions, that will start at this particular hearing.
We are, or have done I should say, our work in terms of pulling together the population health panel for the March meeting in terms of, at least Barbara Starfield and Dan Friedman stand ready, and I think we just have to work with Jim in terms of the private person, so I think that’s for the March one.
The other issue that is on our radar screen which I think will begin to occupy some significant consideration by the subcommittee is that of mental health statistics. I’ll try and briefly tell you how we got here, which was NHANES recently had a forum, and in the forum they were talking about NHANES that will begin in 2005, this is kind of their planning cycle. And one of the things that I noted is that NHANES was going to drop the mental health statistics collection that it’s had for the past six years. Part of that is the way that NHANES works as you heard yesterday is that many of the components are paid for by other agencies and so NIMH had decided that it was going to drop the statistics and it really raised an issue for us of two things.
One is in general our notion of the definition of health and health also being broad enough to recognize that it includes quality of life, mental health kind of more behavioral issues. And then the other is what’s unique about NHANES is that it is one of the few surveys that collects biologic data so that you have the opportunity not only to have self report data but also to combine that self report data with some of the biologic data, and there’s a lot of evidence that has been emerging, well, in the past and even more recently of the relationship between mental health and physical disorders, and also the contribution in physical disorders the role of mental health in terms of handling those physical disorders. So there really is, it’s really difficult these days I think to talk about health and not have any measurement or any information about mental health.
So the committee actually wants to start looking at the issue of mental health statistics and the surveillance in mental health at the departmental level of raising the question of how it is we think about health, the fact that we have an ongoing collection of physical health data but not mental health data and seeing if we can include mental health data on the radar as equally important and necessary to be collected on a regular basis. The short term goals is to deal with NHANES and to figure out whether or not maybe a discussion with NIMH might result in some of the mental health data being paid for in the upcoming NHANES 2005.
Questions, comments, anything else about the mental health data?
New items that are kind of coming on our agenda that we’re going to try and handle but I think some of this is kind of, we’re looking forward to the resource discussion at the November meeting, the retreat. Health, US, you heard the report yesterday about Health, US, and with that being kind of a population health issue we will try and work with Health, US, to kind of figure out how to make that report a report that maybe is more usable or distributed or whether to have it become more prominent in some ways, so we will take up that issue.
We also have been kind of on the back burner trying to deal with issues of privacy and confidentiality as they relate to data that’s collected for these small targeted populations. We’re really trying to look at what some of the policies and procedures are to figure out whether there is a way that we can help increase access but at the same time ensure that we are dealing with protection, so we had had Mark at our meeting in September so I’m assuming that as we continue that we will probably do that with Privacy and Confidentiality.
And then geocoding is something we’ve also been trying to get to because of its importance to these populations. So that’s our full agenda.
Oh, sorry, two reports. We will have a report that we should be able to get to the committee for the March meeting and that should be the report in which we’ve looked at population data, we had a hearing a year ago in February so we’re trying to bring a report together from that. The letters actually were based on some of the recommendations that came from the population hearing. And then at the hearing that we have in November we will talk about the possibility of having a report on Asian, Native Hawaiian and other Pacific Islanders.
DR. LUMPKIN: Any other questions? Great, Bob, you want to try it again?
MR. HUNGATE: No further report.
DR. LUMPKIN: I thought you had enough. Okay, I think we’ve gone through reports pretty much, do we have items, action items that will be coming up in the March meeting? We’ll cover those in the executive subcommittee. What do you have?
MS. GREENBERG: We have the Quality Report, we have a Populations report, we may have some additional CHI activities if we haven’t, remember, we are talking tentatively about a half day meeting on January 29th, but we will be polling everybody with that to discuss the CHI domains and the Sixth Annual HIPAA Report. And depending upon what other subcommittees want to do around that meeting, we already know the Subcommittee on Standards and Security is meeting on the 27th and 28th, members who aren’t involved in subcommittee meetings may elect to participate by phone, but I know several groups, Quality for one and possibly Populations, which has a overlap in membership in any event, are talking about coming in around that time then to also meeting, so you may end up having more people in person then I originally thought. But if it can be combined with subcommittee meetings that’s good. Any other action items from Standards and Security anticipated in March? Beyond CHI?
DR. COHN: I hope not.
MS. GREENBERG: Privacy I heard might have a letter or two, no, yes? Something might come out of the hearings, you would not have anything for this late January meeting. So those I think are the action items.
DR. MAYS: Depending upon the presentation we have on the National Childhood Survey we might have a letter.
MS. GREENBERG: And when would that be for consideration?
DR. MAYS: For March.
MS. GREENBERG: Any other possible action items? Okay, so Vickie already mentioned that we are organizing a follow-up presentation on the 21st Century Health Statistics Report, bringing back the stars from the report, and hopefully also we’re going to identify an additional person to participate and that would be at the March meeting. We also have agreed to invite Dr. Julie Gerberding, the director of CDC, to address the committee regarding the CDC Futures Initiatives and what role there might be for the committee. A question has been raised about the NAS report that everyone has just gotten, did we want to have a presentation on that or has that been referred to the Subcommittee on Populations?
DR. LUMPKIN: It’s just a workshop report, I’m not sure that there will be a final report, in the spring, so I think that would be the time to look at it.
MS. GREENBERG: That was the workshop report but there’s going to be a final report with recommendations?
DR. LUMPKIN: That’s correct.
MS. GREENBERG: That might be for presentation in June? Or maybe even September.
DR. LUMPKIN: NAS time, so that’s June or September.
DR. MAYS: Before we had talked about the release of the disparities report and a couple of other things, I don’t know where those were.
DR. LUMPKIN: They’re going through clearance.
DR. MAYS: Would we want a presentation about it?
DR. LUMPKIN: The Quality? Probably, yeah, March.
MR. SCANLON: Now remember the committee advised AHRQ [rest of comment off microphone.]
MS. GREENBERG: That would be March. Is that both reports on clearance? Okay, so we’ll work with Jim on that and see who would –-
DR. LUMPKIN: Depending on the clearance time.
MS. GREENBERG: Is there anything coming out of NAS on quality that we are supposed to be –-
DR. COHN: Well obviously the IOM Patient Safety Data Report should be out, actually later this month, so it may be something, Jan Coregate(?) or something should come in.
MS. GREENBERG: I mean depending upon timing and availability we can say March, we can say June, but these are reports you’d like to –-
DR. LUMPKIN: Because we’re filling up.
MS. GREENBERG: Okay. Geocoding you said you were handling in the subcommittee and might come back with a proposal for something at a future date.
Okay, are there other suggestions for not only March but June, because we’ll be talking about agendas at the executive subcommittee retreat as well, it sounds like we have more then enough for March.
DR. LUMPKIN: It will spill over into June.
MR. SCANLON: I think the other, this is probably, this could be the full committee or it could be the NHII, again, assuming the budget for ’04 proceeds as we think it will we do have a $50 million dollar demonstration program for hospital health care information technology to promote patient safety, that will be the first round of requests for awards and I think the committee might want to hear how that’s being structured and what we’re seeing and future directions, could be either NHII focus or the full committee.
MS. GREENBERG: Anything else? I mean obviously this is not your last chance but some of these topics require some advance planning and may even require some contractual work or whatever so the more advance notice we have the better.
Anything else Debbie you want me to --
DR. LUMPKIN: Okay, there was a recent article in Journal of Epidemiology about an upcoming study that the NIH is launching on, estimated somewhere in the range of a total cost over 25 years of $2.7 billion dollars, and that might be something we would be kind of interested in so we have a presentation at 1:00, so we will be taking a lunch break, reconvene at 1:00, and then after the presentation we will then adjourn. But we have completed all the rest of the business in a very expeditious fashion. So let’s have lunch and then we’ll be back for a great presentation.
[Whereupon at 12:07 p.m. the meeting was recessed, to reconvene at 1:05 p.m., the same afternoon, Thursday, November 6, 2003.]
DR. LUMPKIN: AS I mentioned before the lunch break we had a very important discussion this afternoon, the National Children’s Study, and we have Dr. Peter Scheidt from the National Institutes of Health, who’s going to talk about this very important development and we hope one that we can follow into our retirement. Hopefully it will start before our retirement, we’ll be able to follow the results. Peter, thank you for coming.
Agenda Item: National Child Health Longitudinal Study – Dr. Scheidt
DR. SCHEIDT: Thank you for having me and I apologize for not being able to switch the schedule to be able to get here before lunch, I tried very hard and unfortunately it didn’t work out. I was asked to come and focus specifically on issues related to the study design and sampling issues and concerns about health disparities and over sampling with regard to that, so this presentation will really focus on those aspects of it. But for those of you who may not be familiar with the National Children’s Study and for a quick reminder, the first three slides are just a quick update and then we’ll go into sort of how we got here and then we’ll go into the details of that.
Just very quickly the National Children’s Study began with a proposal from the President’s Task Force on Health Risks and Safety Risks to Children in late, the task force was formed in 1997 and was charged with developing strategies to reduce and control the risks of environmental exposure, various environmental exposures to children. That concern stemmed from understanding of the increased vulnerability of children to environmental exposures compared to adults, pointing to specific examples such as early life exposure to lead and fetal alcohol syndrome, and those kinds of experiences. It came from the convergence of understanding that there were known measured exposures of concern for which there was not information about possible effects, and there were known conditions, particularly neurocognitive development, autism, and other physical conditions for which there was concern about possible environment exposures.
And so the absence of studies to answer those questions and the convergence of those factors led the task force to realize that in order to provide these national strategies they had to consider the kinds of research that would provide these important answers and very quickly came to the proposal of a longitudinal type study in order to link exposures to outcome, and look at events in different life stages of the participants.
Following the proposal by the task force that lead agencies that were involved with the task force, and by the way let me mention this task force has been reconstituted twice by the Bush Administration, so the current Administration has embraced the work of the task force, including that of the National Children’s Study. And then in the year in the fall of 2000 the Children’s Health Act was passed that authorized NICHD to lead this effort along with the CDC and the U.S. Environmental Protection Agency and a consortium of all those federal agencies with an interest in this concern. And so at that point those lead agencies dedicated staff to begin planning in earnest with an interagency coordinating committee that has led to the development of the study as it’s currently being planned.
The concepts of the study as they as the interagency coordinating committee are planning it are that it be a longitudinal study of not just children but to include extensive information about their families and their environments in which they live and grow. That it be national in scope, that environment be defined very broadly, to include not only classic chemical exposures but physical exposures and behavioral, social and cultural exposures as well. That it be able to look at outcomes that are burdensome but in frequent, so that it be able to deal with outcomes on the order of two to three per thousand, which would allow it to examine potential effects on conditions such as autism spectrum disorder, on diabetes, on cerebral palsy, and those kinds of conditions of those frequency that are burdensome but not real common. That it include genetic data and be able to examine the ways in which environment and genetic expression interact. That it include and matter of fact is made possible by state of the art technology, the ability to handle massive data sets, extensive data management tracking and so on.
As I mentioned the charge is to include all of those agencies with an interest or stake in this study, as well as extensive public/private partnerships for sub-studies, sub-analyses, and adjunct studies. The study is planned as a hypothesis generated study, however, given that stipulation the planners also recognize that a study of this size and complexity will provide an extraordinarily valuable national resource and should be planned and carried out in a way to optimize the opportunity to look at future questions and hypotheses that aren’t currently before us.
Well given that then in the planning of it one of the first considerations is who are the participants and what kind of sample needs to be gathered in carrying out this study. And the immediate first concerns are the extent to which representativeness is important and that maybe seen on a spectrum from a pure probability sample with the entire U.S. population as the sampling frame through varying degrees of clustering all the way to a purely convenience based sample, which might be the easiest and least expensive but may not be what we’re after. And consideration of this sample emerged as a very important consideration about, shortly after we began planning the project, and we recognized that there are a number of important issues to take into consideration, the degree to which representativeness is essential, and there are good cases to be made that many of the hypotheses require an adequate degree of variability in the exposure and the ability to look at outcomes but for internal validity may not necessarily, may not require representativeness. And the degree to which that is required is an important consideration in designing the sample.
To what extent, since the proposal is to look at the effects of exposure on pregnancy the ability to look at impacts on the very early and vulnerable early weeks of pregnancy raises the importance of enrolling women who are not yet pregnant and the degree to which that is important is a big sampling issue, pre-pregnancy versus as early as possible in pregnancy. To what extent do we over sample for sub-populations, and it’s anticipated that, and we’ve set as a criteria that we’ll be able to include outcomes with regard to underserved and disadvantaged populations and minority populations. To what degree do we need to over sample and which populations in order to be able to analyze and interpret outcomes with regard to those populations. Other possible populations for over sampling include agriculturally exposed, highly industrial exposed, those kinds of over sampling.
To what extent do we need to describe the attributes of communities, and to that extent may drive higher degrees of clustering, a widely dispersed unclustered sample may make it impossible to look at effects of neighborhoods, communities, and so on. How stable are the exposures? We maybe concerned about an important exposure this year but due to circumstances unrelated to the study at all those exposures may no longer be relevant five years from now and those are important considerations when those exposures are important factors in determining how we conduct the sample. To what extent are the observations complex, can we gather the data by interview by phone versus computer based interview, or do we and when and to what extent do we need to have the sample participants actually visit a medical center for a facility for more complex measures. And then there’s all the issues of sampling units, segments, and design effect that this committee is far more familiar then I.
Well, to help us with this a year and a half ago as one of the very, one of the first pilot studies we thought we needed to identify what the lead sampling strategies should be, what lead sampling strategies should be considered in order to move forward on this important component of planning. And we through the National Center for Health Statistics implemented a project with a contractor for National Health Statistics, WESTAT, to undertake a systematic analysis of sampling options for us, and they were tasked with suggesting alternative ways for sampling the pregnant women to be enrolled in the study, to conduct a detailed analysis of, a comparative analysis of four lead sample designs, and to contrast the relative strengths and weaknesses of four lead sampling designs, and to do at least a limited cost analysis of these sampling designs.
They came up with, and this lasted about six months of regular weekly meetings and concerted work on the part of the lead sampling statisticians at WESTAT, and they proposed three major models, a household model that would be based on a household sampling frame. That would propose to screen about 500,000 households in order to identify enough women, and this would be about 350,000 women, that in order to be followed through a recruitment period of approximately three years to achieve a sample, excuse me, to achieve a study population of 100,000 live births. There are a number of assumptions based on that that we won’t have time to go into but that was one of the proposed sampling frames, I mean sampling models.
The second model was an office based model, which proposed to sample providers of obstetrical care by regions and/or nationally and enroll women at the earliest prenatal visit through providers of care and the providers of the care would be sampled on a probability basis.
And then the final model was a medical center model that proposed to recruit through major medical centers as more or less a convenience sample over a period of three years and this is more or less analogous to the recruitment and sampling of the Women’s Health Initiative or the Collaborative Perinatal Project as another example.
These proposals were then examined with regard to a number of attributes and just very quickly the perceived advantages of, relative advantages of these three models are for the household model, that is a probability sample and that offers certain advantages, that it allows for enrollment during the very early and conceivably the pre-pregnancy period which offers some advantages and especially enrollment during the early pregnancy even before a participant actually seeks out obstetrical care.
The office based model offered the advantage of being a probability sample also but with reasonably good power, the lowest overall cost, good geographic dispersion, and a fairly quick recruitment that would not have to be prolonged.
The medical center model offered relatively low costs after the initial set-up costs and a particular advantage, the greatest opportunity for biologic samples especially at delivery and especially of placenta tissue if that was a requirement of the study.
The most accessible and I’ve already mentioned that, it offered the best standardization of health outcome measures because of the fairly high degree of generalization and control over recruiting.
This set of considerations is much more detailed than I can present in just a few slides, we examined these in some detail with a selected panel of experts and brought us to the conclusions at this point with the understanding that these models were a starting place only, and it was very clear from this process that before we could go further we had to have much greater clarity about exactly what the measures are, what’s the study all about, and what the study is about has to define how you go about the sampling, and we had a number of givens but not enough clarity at this point to go beyond this point.
The advisory committee, the working groups and the interagency coordinating committee were in fact at this point very much engaged in refining and determining at least initial hypotheses that could be used for this purpose. We also came to the conclusion that whatever lead sampling strategies we ended up with would require field feasibility testing to determine feasibility.
Over the preceding year, and this is about a year ago when we reached this point following the WESTAT project, the working groups, the advisory committee, and the interagency coordinating committee had been working quite hard on the hypotheses that we felt should drive this study. Over 50 hypotheses had been reviewed by the advisory committee in September of 2002 and the interagency coordinating committee then took those and focused, applied criteria that were developed for core hypotheses that would determine the structure of the study. Those criteria for hypotheses are one, first, the recognition that no single holy grail if you will hypotheses was going to be the driving force of this study but it would accommodate and in fact best serve a number of hypotheses or groups of hypotheses.
We also had the stipulation that hypotheses should be required for costly elements of the study, that if it was proposed to collect a certain class of data or see participants at a certain interval at a cost of X millions of dollars behind that should be the necessity of a testable hypotheses.
The hypotheses should be important for health and development of general with regard to prevalence, severity, public health importance, etc. That there should be with the hypotheses a reasonable scientific rationale. Most importantly the hypotheses should require a study of this size, this is not a study of otitis media or bed wetting in children. And it should be answerable with a study of this size, there are conditions such as childhood cancer as an incident outcome that are even as big a concern as they are are too infrequent to be primary core hypotheses for this study. And it should require longitudinal follow-up.
At the conclusion of this point in the deliberation of hypotheses we recognized that a fairly disparate set of hypotheses that met these criteria could be organized by five major outcome areas and we’ve done that in order to simplify the ability to convey what this study is about to those who are interested. Those major outcome areas are listed here with an example of each one of the hypotheses, and in the handout I provided you, at the very end, after the conclusions, I included I think four or five slides that actually include all of this current set of initial hypotheses, but we won’t focus on those in detail.
But the major outcomes, priority outcome areas are undesirable outcomes of pregnancy and one example hypotheses is that infection and mediators of inflammation during pregnancy are major causal factors associated with pre-term birth. A second outcome area is that of neurobehavioral development and one of the initial hypotheses is that low level non-persistent pesticide exposure in utero is associated with impaired neurobehavioral and cognitive performance. And all of these are based upon a certain amount of empiric data as well.
A third major outcome area is that of injury, and that is, an example is repeated head trauma without anatomic damage, this is minor head trauma, is a causal factor of cumulative adverse effects on neurocognitive development. Another major outcome area is that of asthma because of the importance of asthma in the public health of our children and youth. And the hypothesis is that early bacterial and microbial exposures are related to the severity and outcome of asthma, that is the hygiene hypothesis if you’re familiar with that.
Obesity and physical development is also a major outcome area and one example hypothesis is that impaired glucose metabolism as effected by activity and diet is related to obesity in the mother is related to obesity and abnormal physical development in the children.
It’s important to understand and recognize that this hypothesis development process is a work in progress. There are doubtless more considerations and discussions. Not all of these hypotheses will necessarily stay as the study design evolves, and certainly others will be included. And the data that we anticipate collecting will allow many additional hypotheses to be tested, many of which may not even be enumerated in the study plan but will certainly be able to be addressed. We can only handle a manageable number of hypotheses. But nonetheless this initial set of core hypotheses is being used to start framing the study and to guide our thinking about such issues as sampling.
With regard to sampling, having focused and developed an initial set of core hypotheses we realized that we need a process by which we can come to clearer decisions about sampling, and we think it’s important to have an informed deliberation and decision making process. We have proposed the advisory committee and have embarked on a process that will culminate in a workshop to consider options and issues with regard to sampling and lead us to decisions about sampling. And this is, in our conceptualization of this conference we’re modeling this, at least at the moment, along the lines of an NIH consensus type conference, although it is not a formal NIH consensus conference. And it’s being planned by a planning committee with individuals from the advisory committee, the study design working group, and the interagency coordinating committee. This task of this planning committee are to develop the background materials, to lay out the background materials needed to inform these considerations and the deliberation, to determine the participants of the panel of this conference, and to lay out the format for the conference.
With regard to background materials, we have defined a set of white papers if you will that are needed in order to inform this process and I’ve listed them in this slide. One white paper is to lay out in a systematic and careful way the cases for a probability based sample and the cases for a purpose sample, specific to the NCM. The second paper is describe and lay out the impact and implications of recruitment, of anticipated recruitment and retention rates on sampling decisions. The degree to which we anticipate certain rates of recruitment and especially retention may effect what kinds of sampling decisions we need to make.
Thirdly, a paper that lays out the measures needed in support of the core hypotheses as currently defined, a fifth paper that defines sampling options for further considerations. And we’re coming to the realization that, and the end of the process with WESTAT introduced the concept that there may be a hybrid of different components that may not be a pure and simple one type of sample. For example, there may be a stratified or purposive selection of PSUs, but within PSUs a probability based sample for each PSU. Or the PSUs may be stratified and selected according to certain criteria. Or conversely we may have a household based sample for enough of the population to generalize findings to the U.S. but find that it’s way too expensive and logistically infeasible for 100,000 household based sample but a substantial portion of the sample then also be center based or even convenience based sample. But these kinds of options are being carefully examined.
And finally a paper that analyzes the strengths and does a comparative analysis of the leading sampling strategies at that point.
The timeline for this process is the papers are due in January, I’ve already seen preliminary drafts of three of them. The workshop is schedule for February of 2004 with a report to the interagency coordinating committee and the advisory committee in March of 2004. Then the report to the advisory committee and then onto the NIH director who ultimately will be involved in this decisions, with the second half of 2004 feasibility studies underway.
This is one of what are now counting to be about 30 pilot studies that are underway in the planning of the National Children’s Study and I have a series of slides that I won’t go into all the details, I’ll spare you that, but I wanted to provide them for you so if you had specific questions about the kinds of pilot studies that we’re conducting I’d certainly invite your questions. But just a few examples. We did undertake at the recommendation of the very first assembly meeting that we had through a contract with Lewin(?) Group a systematic review of candidate core hypotheses, which actually informed the hypothesis determination process. There’s a development of biomarkers, a very extensive catalogue of biomarkers that has been actually posted on our website that you may find useful for other purposes as well.
There are ongoing pilot studies, such as pesticide exposures in the health of farm worker children in California with respect to measures that may be needed. Demonstration of low cost low burden exposure monitoring strategies, etc. Some health related studies, non-invasive collection methods and storage of samples for genetic testing of using hair, nails, buckle swabs, etc., suitable for a large longitudinal study. The use of primary care practices in the NCS, we’re carrying out a feasibility study that asks the question is it feasible for practice office sites to be used as the primary data collection site for follow-up of the cohort. There are some advantages if it was feasible, there’s considerable question whether or not it’s really feasible using this strategy and through the HRQ practice based research networks we’re carrying out actually a field test of the feasibility of this kind of follow-up method. Other pilot studies, there’s a study on burden of data collection, utility of frozen breast milk, and how that can be handled, and a number of others.
Pilot studies currently being developed, measuring housing quality as a measure for physical environment. A pilot study that’s really a combination of literature review and white paper looking at the extensive range of information and literature available on developmental testing and what kinds of development testing would be most appropriate for this kind of follow-up given the hypotheses and the kinds of exposures expected to be measured.
There are a number of workshops, completed workshops include a workshop on community engagement, fetal and neonatal growth and development measures and out of that came a proposed pilot study of the applicability of 3-D ultrasound as a measure of fetal growth in this context. There’s a workshop on lessons learned regarding ethics of longitudinal study from other extent longitudinal studies. An excellent workshop on assessing the incidence and outcomes of traumatic brain injury that relates to the measurement difficulties with regard to that hypotheses that I described for you.
Scheduled workshops include upcoming, one that just happened this week, Monday and Tuesday on placental measurements, psychosocial stress in pregnancy and the infant since one of the hypotheses relates to psychosocial stress as an exposure, and so on. And then a number are being planned. Media as an exposure, the sampling design I’ve already talked about, probabilities of pregnancy with regarding to pre-pregnancy sampling issues, the rurality as a construct for measurement in this study. Possible roles with regard to childhood cancer. We recognize that as an incident childhood cancer is problematic but there are a number of ways that this large longitudinal study can contribute to childhood cancer, one of which might be as a control group for the National Childhood Cancer Registry and to collect biomarkers and genetic markers along with the registry, that could be enormously important and so we’re convening a workshop with staff from NCI and NHLBI and the Environmental Health Institute and others to examine what possible roles this study may have with regard to childhood and adult cancer.
Your concern has expressed some interesting in addressing health disparities which has been an ongoing part of this planning process, and it’s been a key feature and a consideration in the process to select sampling design, I’ve already referred to the importance of over sampling with regard to certain under served groups with respect to representativeness, stratification, sub-studies and so on. And it certainly was specified in the Children’s Health Act of 2000 and we’ve taken that seriously. And there are a number of other companion initiatives that integrate with this effort, certainly Healthy People 2010, there are other data sources, one of the papers that I’m not sure, I know I didn’t mentioned, it’s on that list, is the use of extent data sets to merge with the NCS from a whole variety of sources and certainly many of these also introduce the issues of over sampling and under served populations and we’re aware of that. And other cohort studies do the same.
One entire working group is dedicated to concerns about health disparities and that is the Environmental Justice, it’s called the Environmental Justice Working Group, and it has a proposed a specific hypothesis on racism and the impact of racism, it’s not race or racial differences but racism as a construct on, and as an exposure, on outcome and how it effects, that construct effects the outcome of children in this study. And that’s for consideration by the advisory committee and the interagency coordinating committee.
There’s a proposed method development project that aims to further develop the measures for the racism construct. There are a number of hypotheses that include concerns about disparities, particularly asthma because of the differential in under served groups with regard to asthma, and so on.
I thought I would just mention, we have begun the activities on actually compiling data collection protocol, we’ve started this effort with the themes and hypotheses that I’ve described to you, and then the program office and the interagency coordinating committee has begun work on developing hypothesis specific data collection schedules and the specific measures that would be required to answer these hypotheses. And from this we expect to be able to identify gaps and particularly overlaps and where we have to make some decisions with regard to subject burden and economics and feasibility.
We’ve developed a tracking database that actually takes each of the hypotheses and identifies the, compiles a database of those measures necessary to address those hypotheses with regard to all of the attributes of the particular measures. This is just a picture of that database with a list of the hypotheses by the major theme areas, and then we’re able to put into that database with regard to each measure when it would be taken, as you can see here at 14 to 26 weeks what the measure is, hemoglobin A1C, who the measure is taken from, the mother, the source blood by venipuncture, where, medical environment outpatient, etc., and a number of other backgrounds even including references in this database, and then can print out in tabular form the details about this measures, including who, where, the method, etc.
The projected timeline for the study currently is actually, has moved into the beginning of 2006 for the initial enrollment of the first subjects in the, what we envision to be the first wave or vanguard centers of the study, followed by enrollment of additional centers. We envision that somewhere in the range of 35 plus or minus centers using the Women’s Health Initiative as the most applicable size study ongoing that we can compare with. And which would result in the very first preliminary data being available from outcomes of pregnancy as early as 2009, and then we envision waves of analyses, first dedicated and funded analyses focusing on the hypotheses to be sure that they’re addressed, public use data sets with funded RFAs to ensure optimal and extensive use of the data, and to ensure that there’s creative investigator initiated analyses of these data.
And with that I invite you to check the website for the National Children’s Study that we update, not as often as we would like but fairly often, and by all means contact us at our email address, and I hope this leaves some time for questions.
DR. LUMPKIN: Great, thank you, it’s a fascinating project, huge undertaking. It’s kind of like going on a cross country trip, the more you’re prepared before you leave the better off you are when you’re on the road, so I can appreciate the effort that you’re putting into it. Do we have any questions?
MR. LENGERICH: Thank you. As with any longitudinal study part of the problem is in maintaining follow-up, so I saw you had a paper related to that. But you didn’t talk a whole lot about methods to ensure that sort of thing and then corollary to that are many of your outcomes or at least your possible outcomes that you’re looking at are actually much closer to pregnancy and birth rather then to the age of 21 where you’re planning to go out to. So I was wondering if you could talk a little bit about follow-up and attrition and where you really, where you might begin to experience some fall off that wouldn’t hurt the outcomes that you’re particularly interested in following-up on.
DR. SCHEIDT: We are very cognizant of the importance of follow-up in a longitudinal study like this and the farther you go in the study the more valuable the patients are and the more important it is to retain them in the cohort. We just spent last Thursday and Friday with some, as part of a mini workshop with our contractor Patel(?) and some sampling statisticians from Harvard and Newark(?) looking at these kinds of issues and one of the actually papers that we looked at were the kinds of follow-ups that we’re seeing in comparable studies and what we could hope to entertain as you’re well aware, it ranges from 30 percent to 90 percent depending on the study and comparability and what one does.
I can only say that we have a working group dedicated to this, we’re all enormously aware of its importance, we have pilot studies focusing on it, and we have several, and we have recognized that in priority it is of the highest priority and I can express with an answer to a question I often get, what will you do when the funding for this study is reduced by half or 30 percent, how will you contract and focus the study. And the answer that we feel is most appropriate is that first of all we make the argument that the real strength of this study is the ability to look at multiple exposures and multiple outcomes and for that reason being able to, because it’s the interactions and the ability to see those interactions and to see how an exposure might result in multiple outcomes or how they interact that is quite unique. Nonetheless, our feeling is that what we would compromise when pushed to it would first be the number of measures, secondly would be the frequency of visits, and last be attention to maintaining the cohort and follow-up.
Time just doesn’t allow to go into all of the kinds of mechanisms that we hope to use to retain the cohort but we do plan to have mechanisms that deal with the relatively high frequency of moving and high mobility in this country, and to be able to pass subjects from one region to another and retain follow-up through mechanisms like that. So I hope that helps.
DR. LUMPKIN: If I could just ask a question related to, it seemed to me that there were a couple of steps in the process of once you have the sample identified one is ascertainment of the data, collecting the data and coding it, and then the data analysis. And my concern has to do with the fact that this committee has spent a lot of time in looking at standardization of data and coding systems and allowing that one system can talk to another, and so to what extent is the data in the system going to be comparable because of the use of the same coding systems that we’ve been working on in relationship to the National Health Information Infrastructure, HIPAA, and so forth?
DR. SCHEIDT: There are a couple of process that are ongoing, one of which I’ve already mentioned, a pilot study that’s actually looking at existing data sources that could be merged or applied with this study and that would require some degree of standardization as well. The multi-agency nature of this planning process with virtually every cabinet level federal agency involved in some respect is another mechanism. We have three people in the program office actually, on the interagency coordinating committee or in the program office who are staff, full time staff at NCHS, one of the lead sampling statisticians at NCHS, Randy Curtain(?), is part of, is spending 20 percent of his time with us in the program office to help with statistics, at least with statistical issues like this. And those are I think processes by which these concerns can be addressed, that’s as far as I can go at the moment. It’s too early to be specific in terms of the actual steps taken to assure comparability.
DR. LUMPKIN: Well, I think let me just take that a little step further because I’m not really familiar with the people that you’ve named, it’s just significant effort through the Consolidated Health Informatics Initiative, which is a joint DOD, VA, and HHS initiative, and there’s a whole host of standards that are working their way through being adopted by the Secretary over the end of this year and early next year. Are you looking at those coding standards as being the tools by which clinical information, laboratory information and so forth would be encoded in the data set that would be created?
DR. SCHEIDT: To my knowledge we’ve not focused on those and it sounds like we should.
DR. LUMPKIN: Yes.
DR. SCHEIDT: Which is why I do talk with groups like this at every opportunity because we often learn things. So the question is how best to, who are the contacts and how do we –-
MR. SCANLON: Actually Peter NLM, you have representatives on this activity at NLM so somebody like Betsy Humphreys or Vivian Auld, not all of these are going to be, some of these are EDI standards but there are some code sets and medical terminology and to the extent you may want to look at when you get further along how well, if they’re incorporated it makes the study much more robust.
DR. SCHEIDT: I mean given the breadth of our considerations I can’t imagine how we would miss this, that we wouldn’t get to this at some point.
DR. STEINDEL: I think your best initial point of contact is Amy Patterson in the Office of the Director of NIH, who’s been charged with making sure NIH looks at and works with CHI standards. You asked my question, John.
DR. LUMPKIN: That’s even better. Vickie?
MR. SCANLON: Let me ask a co-question Peter, you talked about the multi-agency interest and the multi-agency planning, do you also have multi-agency funding?
DR. SCHEIDT: We do, we do for this planning phase. The funding of this planning phase comes about 65 percent is NICHD funded, and the rest is spread almost evenly between EPA, NIEHS, and CDC. And it’s been, actually as you see we’ve done a lot and we’ve accomplished a lot with remarkably little program office staff due to the efforts of the interagency coordinating committee and the various agencies sort of contributing in kind. But the real question, what you’re really asking is what about the study, what about the funds to do this study, and that has yet to come. And it has very much gotten the attention of Dr. Zerhouni, he’s been briefed a number of times, has taken some very strong steps within the Department to seek funding for this study, and I should mention something that will be public this week that’s underway that would be good to mention as well. The Human Genome Institute, Francis Collins and Allen Goodmocker(?) have come to NICHD and to Dr. Zerhouni with the proposal that to translate the human genome research and to apply what has been developed with the Human Genome Project to understand the polygenic, polymorphic kinds of diseases in adults. They need a very large adult cohort to do that kind of research with haplotype mapping, so they’re proposing a large longitudinal study on the range of 400,000 to 500,000 sample size. Discussions are underway with Dr. Zerhouni and with Dr. Alexander to consider basically including the parents and grandparents of the National Children’s Study as a multi-generational cohort study to look at large samples, and that my have some funding implications as well, both supportive and whatever, but those are just discussions underway, no decision has been made about that, but it’s worth noting.
MR. SCANLON: It would be a very, obviously this is a very expensive study and the money would be competing with other research and with other statistical activities as well. Is there some, in the planning, Peter, is there some sort of strategy for scalability so that as, obviously you need a certain threshold to even go ahead with the study, otherwise it’s like any other study, longitudinal study, but is there some sort of a scalable approach where you reach a certain threshold of funding and you can go ahead and to the extent funding increases you can increase or you can contract, because at this level for other then outside of DOD this is a lot of money and a lot of complexity.
DR. SCHEIDT: It’s a lot of money and it’s, the position that Dr. Alexander has taken throughout this that it was Congress’ intent that this study should be planned with the expectation that Congress would fund it. And he has sort of steadfastly said it is not the expectation that major other health research initiatives should be sacrificed to go forward with this study. Now there’s a certain amount of things that this study can address that may obviate the need to do other things. We’ve not analyzed that in great detail but it’s safe to say that without some additional appropriated funding it’s not feasible that this study go forward and we’re not expecting to have all the agencies sort of pony up and contribute to this.
DR. MAYS: I’m going to wear a couple of hats in terms of asking questions besides just populations because I’ve also served on the university’s IRB so I want to kind of put that all on the table. One of the things I’m trying to understand is regardless of which of the sampling designs you decide on clearly the study will have what we often talk about is vulnerable populations because my assumption is that some of the hypotheses are directly, will benefit those groups so in order for you to get an answer they would need to be in. As I think about how you’ve laid out your beginning work what I’m not seeing are actually, you have several concept papers and methodology papers that ask about the method but not about whether or not the people will fit the methods, and whether or not, for example, when the follow-up attrition question came whether or not for example as you set the study up whether or not these vulnerable populations are likely or less likely to participate depending upon which sample design that you take. For example, I’m going to take what you just said in terms of the Human Genome Project. As you begin to lay on these layers, so say it’s grandparents, parents, and then the child, there’s some background for three gen studies, the statistical issues in those are quite complex and probably haven’t been developed well enough, but beyond that what we find is that particularly in terms of racial and ethnic minorities, the questions that you’re going to ask about not only the health of the child but their environment. And when we begin to think about a lot of the complexities of what happens in terms of some of the environmental issues and if they get reported, can you talk a little bit about how you all will match up the issue of the design with ensuring that populations will want to participate, that the design doesn’t kind of scare them off. I don’t want to know if there’s an answer, but I’d like to hear the thinking, there has to have been thinking about this.
DR. SCHEIDT: You bet, you bet, we are very much thinking about how to not turn off populations and there are, the real answer to that is we’re not at that point yet. I can only say we’ve, in our discussions we’ve certainly acknowledged the importance of just what you mentioned. One strategy that we’re almost, we’re certain of including is understanding that this has to be a rigorous national protocol with a carefully laid out set of core measures. We’re also cognizant of the importance of leaving enough flexibility to accommodate the needs and issues of local and regional communities, both from the investigator’s standpoint for their input into the carrying out the study and from the community’s standpoint. So we envision that, for instance, if we anticipate a competitive procurement for data management centers we will, as part of those procurements will be the requirement that they have the concerns of their communities, including the communities that you mentioned, as part of their response and carrying out the study. And so we anticipate the necessity to include local and regional, concerns about local and regional sub-populations.
There are some populations that we anticipate we must be able to describe, the Native American population, which with a sample size of this will clearly require a significant amount of over sampling. And now other sub-populations, other then African American and Hispanic, we’ve not gone into at that level of detail exactly what those, how those will be defined and how those will be measured.
DR. MAYS: Can you just say whether or not you’re going to, what languages you’re thinking about, and also in terms of like some of the genetics whether or not you’re going to do any either concept papers or methodology papers to look at how certain racial and ethnic groups have very different perspectives about the ownership of biological samples and why some groups don’t participate in research because of the way in which we normally do data banks?
DR. SCHEIDT: We don’t have a concept paper on that topic specifically yet I can certainly envision some of the workgroup deliberations and some of the papers including that but not focused on ownership of biologic data, that’s a fairly narrow topical area. But we have a group, one working group, the Ethics Working Group is focusing on ownership of data and who owns it and how and how you deal with those. We have two ethicists that we’re bringing on board part time as IPAs to work on more details of those considerations, but I don’t have more detailed answers for those questions at the moment.
DR. MAYS: What about the languages that you’re translating?
DR. SCHEIDT: Unquestionably Spanish and we would like to go as far as we can. We recognized that there are logistical limitations with just how far you can go in terms of actual translation and back translation. The major languages are Chinese, Thai, Eastern European, Russian, those are large ones, and we’ve talked about how far we can go with that and we simply haven’t made that decision and that has to be looked at from a logistical and cost standpoint.
DR. LUMPKIN: Okay, we have time for one more question? Eduardo?
DR. ORTIZ: A lot of innovative and interesting and important ideas in research are often generated when you take this type of data set and you kind of allow creative and smart researchers, clinicians, etc., out there to work on this so I’m just wondering have there been any discussions or plans at this point for developing a public domain data set as data begins to accumulate from this project?
DR. SCHEIDT: As early as a year and a half ago I called the individual, and I don’t remember his name right now, at NCHS who is responsible for establishing the HIS public use data sets to begin the process of planning public use data sets for this study. He was dumbfounded that someone at this point in planning would think to call about establishing a public use data set and said it had never happened before. We do anticipate waves of public use data sets with associated funding for investigator initiated analyses and our feeling is it’s enormously important to make that available as quickly as possible. We absolutely do not want a situation where these data are basically owned and occupied by the principle investigators of the centers and used until they’re finished with all the things they’d like to do and then make it available to the rest of the world. We’re more concerned that these data be maximally and optimally used then we are that various groups get to publish off of them and a concern of ours is that large longitudinal studies do have a reputation of being under utilized in terms of analyses and we’re concerned about not letting that happen in this instance. I hope that answers your question.
DR. ORTIZ: Great, thank you very much.
MS. PAISANO: I’m Edna Paisano, I work with the Indian Health Service, you had mentioned about Native Americans, I just have a comment and then I guess a couple of questions. The term Native American gets to be very confusing and at this point with an OMB directive it would help to use American Indians and Alaska Native –-
DR. SCHEIDT: I’m sorry, it would be helpful to use –-
MS. PAISANO: To use American Indian and Alaska Native –-
DR. SCHEIDT: Well that’s where it was, I’m an alumni of the Indian Health Service and you never quite get over that and when I was in the IHS it was American Indian and Alaska Native but it seemed to me since then I’ve heard Native American more then that.
MS. PAISANO: I mean everybody in this room basically can say they’re Native American, so it does cause a lot of confusion. But I know you talked about over sampling with the American Indian Alaska Native population but there’s really different kinds of universes that you need to be accounted and there’s over 561 tribal governments and they have a sovereignty and you need to ,work with them on a government to government relationship, and involve them some way with all these studies. And the other kinds of universes of looking at on and off reservation, living in urban areas, plus Alaska, I mean Alaska is very unique.
DR. SCHEIDT: Yes, I know. There is a sampling issue, it is only 100,000 and so we will certainly need the help of sampling statisticians to guide us with this but we have felt that this study could not be done without being able to address both exposures and outcomes with American Indians and Alaska Natives if you will.
MS. PAISANO: The other comment I’d like to make is in terms of language and translations, all American Indian and Alaska Natives are oral languages, so it’s more oral translation into the questions and not written.
DR. SCHEIDT: Thank you.
DR. LUMPKIN: We’d really need to adjourn. Why don’t you grab him afterwards. We’d like to thank you very much for coming and finishing off this meeting with a very fascinating study. We’d like to have you come back periodically and update us on this because hopefully to the extent that this committee can play a role or provide input we’d be happy to and obviously we’d like to support this initiative.
DR. SCHEIDT: Thank you very much.
[Whereupon at 2:13 p.m. the meeting was adjourned.]