[This Transcript is Unedited]
Hubert H. Humphrey Bulding
Room 305A
330 Independence Avenue, N.W.
Washington, D.C.
DR. FRIEDMAN: Let's get started. I see a sea of name tags. Let me take a couple of minutes and recap where we are. Some or all of the report will be a joint NCHS/NCVHS Data Council product. The only part which may or may not be a joint product is the final chapter recommendations.
That could be a joint product or some of it could be a joint product and we'll just have to see as we go along. The report has been written by Ed Hunter and Gib Parish and myself with some terrific help from Rob Weinzimmer, and it's been edited by Judy Kaplan, the scientific editor for PUBLIC HEALTH REPORTS.
In early January, we circulated chapters 1-4 of the report to the workgroup for review and comment. We had a conference call on that and we have subsequently incorporated almost all of the line edits. There were 2-3 suggestions that Marjorie had that I thought were very good suggestions, but we have not had time to incorporate because they're helpful and cosmetic, but we have not had time to do it. I'm not sure when we will do it although I think they should be done.
There are 2-3 things that have not been finished yet. Those include one table that would summarize gaps in the health statistics enterprise that were mentioned during the testimony in the public hearings. That's going to be incorporated before we distribute the report for public comment.
There is also a table or a figure that's going to summarize an illustration of how data collection streams can overlap. As an example, we may use occupational health. We're not sure of that. In addition to that, we are also in the process of incorporating more references to the testimony we heard during the public hearings throughout the report.
Rob has done an exhaustive and comprehensive and exhaustive job of that. What we have is chapter 5 which is the recommendations chapter. It was distributed to the workgroup. We had a conference call. We received comments on it and I think that the comments were supportive overall.
I think that the chapter is in good shape. There was one particular point that generated some feedback from Barbara and Marjorie that we should discuss which is the notion of the Health Statistics Planning Board. The other thing that Barbara mentioned in passing which was something to the effect that she found it a little bit overwhelming.
Paul Newacheck also mentioned and it resonates with me and Gib and I wrote most of the chapter. If it resonates with me, something may be a little bit amiss. Basically, Paul was concerned that the critical recommendations are going to be submerged because we have so many recommendations in there.
He was putting on the table, is there some way we can not only separate out the critical recommendations which in some editing that I've already done by putting them all following each other upfront. In addition to that, whether we can clearly differentiate the critical recommendations from the other recommendations that may be important, but are not essential.
Any comments on chapter 5, particularly the Health Statistics Planning Board and is any suggestions for making the chapter more readable? Let's spend a few minutes on the remaining process which would be in mid-March, sending it out for public comment in draft form.
DR. STARFIELD: Where do we stand now with the Health Statistics Planning Board, what's your thinking about it? My thinking was that if there was no experience with this kind of thing. No precedent for such a board and I couldn't really see how it would work. Then we came up with a possible precedent and it's the quality thing that joins the art with the private sector. I guess that is a precedent. That's how you would visit this as a joined public/private, right?
DR. FRIEDMAN: We envision it as joint public/private, three governmental levels, and to include federal employees as well as state and local and so on. You have given it more thought after our conversation.
MR. HUNTER: Going back to what it is we want to do, to think of it as a truly advisory group where it is not just an advisory group that takes the individuals that might have expertise or experience and has them come up with ideas, but a group that is a stakeholder group that would have people from the public and private sector in federal and state and in associations that represent some of these institutions.
It would be a place where people could make joint decision making, not necessarily binding or enforceable. Something more than here's an idea. Let's recommend it to someone else. It is a gathering of people who might actually represent points of view that could be reconciled and worked through. That's partially what an advisory committee does, but it can have decision making powers because it can't organize and represent.
DR. LUMPKIN: You're actually describing an SDO for health statistic.
DR. STARFIELD: What did you say last?
DR. LUMPKIN: An SDO for health statistic.
DR. STEINDEL: I was thinking along a similar line. It would have more impact if it was external to the government.
DR. LUMPKIN: And then the analogy with the quality form of channeling recommendations that may need HHS action through the NCVHS and the FACA committee, just as the other SDOs looked for their recommendations to be anointed by the departments through the NCVHS.
DR. FRIEDMAN: You said it's going to be external to the federal government. We thought about it being autonomous but including federal government representation.
DR. STEINDEL: An SDO definitely includes the federal government representation. There are a lot of federal people involved in multiple SDOs. There has even been a good clarification recently about the roles federal employees can take with SDOs. Before, we were somewhat restricted in taking leadership positions and now we can. The influence of external SDOs we see at almost every NCVHS meeting. It can be vast. Likewise, we see the influence of internal government organizations on the NCVHS.
DR. FRIEDMAN: How are the SDOs funded?
DR. STEINDEL: Membership dues.
DR. LUMPKIN: The forum is funded about half from membership dues and half through grants and contracts with the federal government to develop various work product.
MR. HUNTER: You would picture this as being an organization that would be funded and have staff that would have membership or there are SDOs and there are different models of SDOs. This is where we agree. What we need to do for the next generation is be a little more specific about who will be on it and how it will be structured.
DR. LUMPKIN: There are SDOs. For instance, CSTE in some ways, functions as an SDO in working with CDC and developing standards for vital statistics and for reportable illnesses which, because they all agree that this is the way we're going to do it, then the individual states can implement it.
It's not really imposed by the federal government. There is a low-budget operation as opposed to the quality forum which is a $3 million operation. We can envision that we would look for certain membership, that there might be some dues or in-kind contribution. It could be an affiliate staffed by ASHTO. One of the other members could play a role in giving it a home.
DR. STEINDEL: The CSTE or ADHL model, CDC provides a lot of their staff money.
DR. FRIEDMAN: Do you think that it would be seen as competitive with NAFTA or CSTE?
DR. LUMPKIN: I thought you were going to say the Public Health Data Consortium. If it doesn't see itself doing that, I don't think it would be. There would have to be some careful evaluation of the Public Health Data Consortia because I think that's one of the areas where they may be some potential for overlap.
MR. HUNTER: Part of what we would hope would come through with all the next generations of chapters 1-4 and 5, is that we actually got a lot of different players that do individual pieces. The Standards Consortium does things in the standards and classification area.
But it doesn't do things in the planning of data systems or suggestion of content of data systems or the content of a survey or the interaction of the VRFS and national surveys. It has it's defined limits. NASIS does things in vital statistics and could do things elsewhere, but typically doesn't. CSTE does things in infectious diseases and doesn't deal with health insurance data.
DR. LUMPKIN: But the membership in CSTE is limited to state employees. If we envision this being a public/private partnership, then it's not competitive with those two organizations.
MR. HUNTER: They have to come to it. You could describe this as an umbrella to which the NASIS folks bring the vital statistics expertise and the standards people. There may well be some things that up the standards consortium and it might be a place to look for members for this. The membership of that is quite broad, but they all come with people who are there to talk about a specific subset of the general issue.
DR. LUMPKIN: The difference is that we would conceptualize this as an organization of organizations rather than a membership organization.
DR. FRIEDMAN: That's the point I was trying to make before. I don't think we're looking for the world of NCVHS or FACA advisory committee. We're not looking for 16 really good people who will come and give advice. We're looking for the organizations that will bring to the table the willingness to do something collective or joint, that actually have some collaborative planning exercise.
DR. STEINDEL: What I see as a basic difference between the organization that you envision and my picture of it and the Public Health Data Consortium is that they are dealing with day-to-day practical problems. You would be dealing with what is going on in the statistical world which is the very specific niche of that. It's part of their bigger niche, but just a very small part. Their biggest focus is HIPAA transactions.
DR. FRIEDMAN: I would hope in addition to being an organization of organizations, we could also have some individual representations. My concern is that if we limit it to an organization of organizations, it's going to end up being both the same five people. In addition to that, the nature of the business is going to be more restrictive rather than trying to think more broadly about what it is we should be focusing on.
DR. STEINDEL: Most of the SDOs have a structured set up like that where an organization can take a membership and there are also individual memberships. You get a fee structure for both. I can take an individual membership in HL7 and CDC has a membership in HL7. We both have an equal vote.
DR. FRIEDMAN: Who would appoint an individual member that would carve out the institutional interests of the bigger members?
MR. HUNTER: I don't know.
DR. STEINDEL: In this area, I can see adding to the individual membership.
DR. STARFIELD: How big are these SDOs that we're talking about?
DR. STEINDEL: Some of them are small. Hl7 and X12 are huge. A group like the Computerized Patient Record Institute is essentially nonexistent right now, maybe has 20 members. Some of the ANSI SDOs like the College of American Pathologists which took out ANSI accreditation as an SDO. It's probably as small as possible because they have a membership of one. It's just them.
MR. HUNTER: It makes collaborative planning much easier.
DR. STEINDEL: Actually, it makes it a lot harder.
MR. HUNTER: Do you think that this is a useful way to go? We're talking about some way and most of these standards are not strong arm enforceable except that once there's a collective, it's the magnetic center of gravity that pulls everybody towards it. That's what we're describing here. I'm moving more towards, do we use this is the thing that characterizes an SDO that we would use in describing this?
DR. STEINDEL: You have to be a little bit careful in describing what you're going to be doing as an SDO because they have a tremendous amount of power in terms of the federal government and laws. It's a procurement law that says we have to give consideration to nationally accredited standards in what we purchase.
For instance, if you found an SDO and got it ANSI accredited and you got ANSI accredited standards for the development of health data statistics and the federal government wanted to purchase health data statistics, as part of their procurement, they would have to say it must conform to it.
DR. LUMPKIN: The downside of that is?
DR. STEINDEL: I'm just pointing it out. I personally do not think there is a downside.
MR. HUNTER: I'm not assuming that we're talking about a standards development organization because we're not just talking about standards. We're talking about a lot of things in addition to standards, but the model is the organizational model based on the participatory.
DR. LUMPKIN: I'm not sure the document needs to go into that detail. It's important for us in our discussion to understand this new entity and how it relates to the NCVHS. By describing it at least when we go through discussions with the committee in a model that's well understood and accepted, it will avoid a lot of controversy that ought not be there.
DR. STARFIELD: There are a lot of things to be decided. Can anyone join?
DR. LUMPKIN: I'm not sure all those decisions need to be made in the document.
DR. STARFIELD: If we want to get this through the committee, we need to tell them what it is.
DR. STEINDEL: It depends on what you mean by open.
DR. STARFIELD: Can anyone join it?
DR. STEINDEL: Anyone who has acceptable professional qualifications. That's one form of openness. Anyone can join is another form of openness. You really want an open organization to people who have professional qualifications.
DR. STARFIELD: How do the SDOs enforce that?
DR. STEINDEL: In a lot of cases, they are not concerned with the professional level, but you take any professional society like the American Chemical Society, when you apply for membership, you have to show that you're a chemist.
MS. GREENBERG: I can't apply.
DR. STEINDEL: It's actually not a degree in chemistry, but a related profession or something like that and approval from two members. Marjorie, with the approval from two members, you probably couldn't join.
DR. STARFIELD: I don't know what the credentials would be to join this one.
DR. STEINDEL: What are the credentials to join NCVHS?
DR. STARFIELD: None.
MS. GREENBERG: I'm sorry to be late, but I was next door with the Privacy Committee.
DR. NEWACHECK: I had special dispensation.
DR. FRIEDMAN: I had dispensation for leaving yesterday at 4:30.
MR. HUNTER: You, however, chose another workgroup over us.
DR. FRIEDMAN: I'm not convinced that we're going to be completely nail it down. I do think that we've got a lot more detail now than we did before and we should really try to flesh it out in the chapter. There were a couple of other related recommendations in there that I think we should spend a minute on. One was 1.1 and the other is 1.2.
MS. GREENBERG: Are you done discussing the planning board here?
DR. FRIEDMAN: Yes.
MS. GREENBERG: You're still going with it?
DR. LUMPKIN: Looking at it like an SDO in its relationship. It's a public/private partnership. If there were recommendations like CSTE or standards related, it would go to an SDO. The conceptual model would be that rather than a new advisory committee.
MS. GREENBERG: Why do you feel it's needed? Is it so that the states will have more of a voice?
DR. FRIEDMAN: No. Right now, I don't think that there is any forum which fulfills the functions that we've envisioned for this. There is really no place where the different levels of government and the public/private sector get together to evaluate what should be done in health statistics.
Some of that may be a reflection of what's happening and not happening within the federal government. Having said that, even if the federal government were better coordinated, there still is lacking a convening body.
MS. GREENBERG: Would this board make recommendations to the department through the national committee?
MR. HUNTER: We have to figure out the channel. The department would ask the national committee to advise on it, but things come into the department. The more structural issue is there isn't a place the department goes when it has to sit down and agree with anybody else in order to find a point of view in this area.
It doesn't have a standing place to go for the department to deal with the states on how it might do certain things except in very specific and targeted areas. There isn't a place to go and actually jointly plan things. There are standards of things to go to.
There are plenty places to go to get advice, but there isn't a place to go when you work through joint things. That's what this is supposed to do. It's not any one of those individual things. It's the place where hopefully you start to pull some of these things together.
DR. FRIEDMAN: Related to that are recommendation 1.1 and 1.2 which are intended to try to remedy the lack of health statistics coordination with in HHS specifically and throughout the federal government more generally.
DR. LUMPKIN: I have a question. The first is, is there really a big difference between A and C in 1.1. There are three options that are being proposed. There are two options. One, a single entity within HHS should have this agencywide responsibility. Two, that the Data Council should be beefed up and do it.
If it's a single entity within HHS, there are two strong candidates. I would rank them in the order in which I would see it. One would be HHS, Center for Health Statistics with a broader responsibility and the other would be ASPE.
MS. GREENBERG: The difference between A and C as I read them is A is an agency responsible for planning, budgeting, and agenda setting. C is actually having a statistics candidate for health. It's actually having all the data collection analysis and all that in a single agency.
DR. FITZMAURICE: That wasn't intended to be the difference. There is a lead agency that wouldn't necessarily do it all itself, but would be charged with thinking more broadly than some systems. The other was, put it all in one place.
DR. STEINDEL: I agree with John. A and C should be the same. If you have one person just setting the agenda and telling everyone else to do it, you're not necessarily assured it's going to be done right. That can be just the Data Council like we have now. If you have a single agency responsible for planning and doing, it's a little difficult.
MR. HUNTER: Let me weigh in for a second on why the range and why we could argue which of these would be our favorites partly recognizing that this is an NCVHS Data Council product ideally.
That implies departmental clearance of whatever is said here and I think to end up with a single option on how the department would structure its things rather than to recommend that it be strengthened would make the process that much harder.
The idea was to signal here that something needs to be done for better coordination within the department. We would never get this report cleared if we try to sort through the politics of whether we're recommending one or the other.
DR. STEINDEL: But if you recommend one as being the department and the other being the Data Council, which has the responsibility?
DR. LUMPKIN: Listening to that discussion, that does differentiate the two. I'm not sure that's as clear as the differentiation between A and C. The third alternative is having them conducted and that's problematic because of the problems that NCHS is having in funding. If you have the data collection, the health statistics collection function within a problematic area, it's easier to fund them than if you have them as an isolated area where it's harder to justify funding.
DR. NEWACHECK: I see problems with having this within NCHS. When it was separate, it would have the capacity to do this much better. That could be a parallel recommendation somewhere down the line. Recognizing the point about being sensitive about making this broad and letting the department see that there are lots of options and we're not trying to narrow it down to one.
I wonder if the Data Council has the capacity to do this kind of thing? It seems that this is a huge enterprise that I can't imagine how they would be able to do it? Is this for political purposes that we use the Data Council as an option?
DR. LUMPKIN: Let me take that one step further. This may be the parting of the ways. Should we visit the issue of having a joint report on where the statistics are with no recommendations and then a report from NCVHS with recommendations? I think we can be more pointed and directed. At some point, the question becomes, what is the benefit for issuing joint recommendations if they're so non-specific that they're not helpful.
MR. HUNTER: I had in mind more trying to find as much as could be said jointly. If, in fact, there is a joint recommendation that says we can do more than there is now and it's signed onto by the department and committee, that has some value. In the transmittal to the department when the national committee is finished with this and passes it to the department along with it, it could go further to say in 1.1, we are making a stronger recommendation than was made by the collective party.
We had this discussion early, would we sever? There are a lot of things that can be agreed on. I wouldn't want to sever the recommendations from the report all together. This is probably the only one where there would be a substantial difference. I would think the committee would be better off with a form that they can say we all agree that this important and here is how we think you ought to do it.
DR. FRIEDMAN: That may be the way to address Stan's earlier comment. There is a transmittal from the NCVHS that highlights critical things that need to be done.
DR. STARFIELD: I think this much too vague the way it's stated. You say this is what we've got now except instead of "or" it's "and". You've got all three of them now.
DR. FRIEDMAN: I don't disagree with that. That argues for we need to put in more flesh around the Health Statistics Planning Board, 1.4. We clearly need to put some more flesh around 1.1 as well.
MR. HUNTER: I don't think we have any of these three.
DR. STARFIELD: We have all three of them together. A is ASPE, B is the Data Council, and C is NCHS.
MR. HUNTER: ASPE doesn't have responsibility for budgeting or agenda setting. They have a role in it. The Data Council is anything but strong and asserting leadership. There is no single agency that has responsibility for things beyond. NCHS has responsibility for what it does now, but not for 90 percent of what goes on in the department. I would say this needs to be strengthened to say we don't have an industry now.
DR. NEWACHECK: That would make it much more powerful to say that we don't have.
DR. FRIEDMAN: The way we tried to structure this was to leave the gaps and the issues in chapters 2-4. This was going to be purely recommendations. I'm not particularly tied to it.
MR. HUNTER: I think we need to revisit what we said about these things in chapter 3.
DR. NEWACHECK: It does seem like it makes sense to put in the recommendation to say, we don't have "x: and we need "x" and specify a little bit. I don't think there should be an option that we don't think is valid like the Data Council. If we really truly don't believe the Data Council can do this, why say it? It doesn't make sense. Combining A and C together and saying that either within an existing agency or creating a new agency to do this is what's needed. That's what we believe, isn't it?
DR. FRIEDMAN: I'm wondering if it's combing A and B?
MS. GREENBERG: There is now way C. We're going to take the statistics of NIH and out of ERSA and CMS.
DR. FRIEDMAN: I'm not suggesting that we eliminate C, but combine A and B.
MS. GREENBERG: C is so far from reality.
DR. NEWACHECK: C was more like old NCHS 25 years ago.
MR. HUNTER: Nobody else did much.
DR. NEWACHECK: NCHS was the leading agency then. It's gradually disappeared.
DR. FRIEDMAN: We may not have a full fledged C, but we might be able to move more towards and NCHS that is a national center of health statistics and not a center for national health statistics.
MS. GREENBERG: I'm all for it.
DR. FRIEDMAN: C is on a continuum as well.
MS. GREENBERG: That needs to be clear. The idea of a single agency within DHHS conducting everything just seems so far from reality and would be so opposed by AHRQ, CMS.
MR. HUNTER: It is the option that has been put on the table over and over. It's not a new thing to discuss this. It's been discussed many times and is moving on a continuum. No one is suggesting or advocating that it go all the way to the end of the continuum, but there are some things that could be done to streamline departmental data collection.
DR. NEWACHECK: We already have that. The Data Council is supposed to be coordinating AHRQ and NCHS on that. It doesn't happen. There is nothing there. There is no relationship. I don't see how creating a new agency and giving it a new name and coordinating council and say you coordinate NCHS and AHRQ together. It's not going to happen.
DR. LUMPKIN: One way is to make the recommendation that there needs to be stronger coordination, integration, and leadership within HHS and health statistics. That would be the recommendation. In the descriptor, testimony and hearings suggested three approaches to that: A, B, and C. The recommendation is that there needs to b a strong lead within HHS on the issue.
MS. GREENBERG: I haven't been sold on the planning board. You go with guiding principle one which is total support for content, and then say exactly how this can happen. That should be the first topic of this planning board. You get the stakeholders in and have a process in which this is openly debated and discussed. They come back with a proposal as to how it can be done.
DR. FITZMAURICE: I wouldn't want to make it completely upon the Health Statistics Planning Board because I do think that the holes in what's now going on at the federal level are such so that if we say we can't do anything until there is Health Statistics Planning Board, that may make it even less likely for any of those.
MS. GREENBERG: I was trying to think of something that would bring pressure upon the department to make this happen because it isn't going to be the Data Council.
MR. HUNTER: It seems even less likely to me that the department would wait for the planning board or private, state, and other partners in this to tell it how to run its own business than it would take it from the national committee.
We're just saying there is another year or so until a group of people come in and they're still not going to come in and tell the department how to reorganize or how to structure its own advisory process. That's an inherently departmental judgment.
You might well have people that can say it's crazy to have HIS so disconnected, but you can't say you have to reorganize the department in the following ways. Why would the department take that from an SDO-type body? The reason why your model makes some sense. In a transmittal, we can say this one doesn't work and this one does.
That makes more sense is that we will lose the jointress of this quickly. For the Data Council as an institution say the Data Council can't possibly work is a problem and I'm not sure we will do better than a strengthened Data Council.
I'm not sure we want to say don't have a Data Council if you can't go the rest of the way. I don't know what this new administration actually would want to do with some of this stuff. We either punt on the recommendation or we have to say something that says this is a problem and the Data Council doesn't really do all these things. If you want to do it, we can get some of that.
Even within NCHS, you would have very different answers for some of these questions, and certainly within the Data Council, and within the departments. This has been argued.
DR. STARFIELD: We all agree that C would be best?
MR. HUNTER: Yes.
DR. NEWACHECK: I do.
DR. STARFIELD: I do too, but not everybody does.
MS. GREENBERG: You would make the distinction between programmatic statistics which individual agencies woudl still have.
DR. NEWACHECK: And research studies that are focused at a narrow population. NIH would still have those. Things like MEPS and HIS should be together. This is an opportunity to get NCHS back in its own domain and to be a leader again as opposed to stepchild of CDC.
I don't necessarily agree with those recommendations, but that's a fairly broad range. To say that we should pull things in and make exceptions. I'm concerned about the funding politics. I used to be a big fan and experience has shown me that if you don't have a clear constituency or tied to a program, things don't get funded.
Look at the funding for NCHS. It's at 2 percent or whatever and that's 1 percent. There are some real problems with trying to set that up that's different than saying that there ought to be greater consolidation of certain datasets and health statistics endeavors within NCHS which I'm supportive of. It's not inconsistent with saying that somewhere in here that NCHS ought to be a separate agency within HHS.
MR. HUNTER: That's the kind of thing you have to deal with or we have to sever some number of things. NCHS can't institutionally suggest that it be moved out of its current organization.
DR. LUMPKIN: Let me ask this question. If I'm sitting in the Secretary's office and I get a transmittal that says the study was a joint project of NCHS, the Data Council and NCVHS. On the transmittal, NCVHS says NCHS ought to be stronger and have more stuff, I'm going to be very suspicious of those recommendations.
DR. STARFIELD: You already suggested severing the recommendation.
DR. LUMPKIN: I'm actually suggesting that we reconsider whether or not the entire report be a report of just NCVHS and whether the utility of a joint report and the ability of getting the recommendations followed is worth the appearance. That's not something that can be decided here at this point.
MR. HUNTER: It's worth considering. If we had made that choice at the beginning, it would be easier to say that there wasn't suspicion as to where these things came from. The fact that the process has been advertised as a joint one and the all the hearings and drafting and issuance and request for recommendations came jointly makes that.
DR. NEWACHECK: NCHS hasn't staffed to NCVHS in some sense. That's okay and a really good point. If what we really want to do is make a strong recommendation for NCHS to go with that and to empower it, it can't be part of the author of the report.
DR. FRIEDMAN: Having said that, it's going to be obvious to anybody who has been involved that it was one of the three major authors of the report.
MS. GREENBERG: That's okay, but how many of our reports by our committee are actually written by the members of the committee? They're all with substantial writing by either department staff or people paid by the department to write it.
DR. LUMPKIN: How many of them have things in them that directly bear on the institutional interest? We had a lot of push back because of our recommendation of their being in the central focus and it not be the CIO.
MR. HUNTER: What you're saying is that you would have used the CIO to draft your report and made it an NCVHS report contrary to what the CIO might have wanted.
MS. GREENBERG: This may not be possible, but I'm thinking of a paradigm. This has been a joint project in that NCHS and the Data Council because we do get funding and support and staff through them, have supported this project in a way the NCHS and the Data Council and others in the department might support a study by the Institute of Medicine by the National Academies of Science.
At the end of the day, the recommendations and everything come out of the NAS and they won't let you see them until they're published in the NEW YORK TIMES. In a sense, the power play would be that the committee could assert that we've done this together and couldn't have done it without all your support.
We want your comments and all that. We feel that we can't really go forward as a joint report and make recommendations that will be helpful to the department because we're too constrained by having to get everyone in the department to agree with them which is not the way an advisory committee works.
DR. FRIEDMAN: How does it strengthen the chances of getting it accepted if it's a joint recommendation? There are three recommendations in here that are really at issue, which is the Health Statistics Planning Board, the Health Statistics Council, the federal government and HHS coordination.
The rest of it probably is not an issue. My feeling of whether or not it should be a joint product or an NCVHS product, I don't really have a sense of what's best in terms of influence. I would also like to do is hear you discuss it.
DR. LUMPKIN: We should ask Jim. Some of the issues where there are those people who head up units of government who believe that stuff that comes in from the outside is better than what's generated inside.
MS. GREENBERG: Very little regard for things generated inside the current department.
DR. LUMPKIN: The NCVHS may be perceived and can be pitched at being a group of outsiders who are here to tell the department how to better run their business. In this environment, that may be a plus.
MR. HUNTER: I thought you were going to say that the one thing this department seems to want to do is to centralize things and manage them.
MS. GREENBERG: One department, one health statistic.
MR. HUNTER: It's more consistent with 1.1(c). I was browsing through the rest of the recommendations. The idea of having this be a completely NCVHS thing. There is a lot of smaller and practical things about the privacy stuff and others. They are consensus things.
The way out of this may be to have principle area one which is the only political one. We could say there is a clear conclusion that there has to be stronger central governance and coordination and that the department bears a responsibility for doing that.
Say we could not agree given the multiple partners and differences of views on how best to do that. The national committee could have recommendation that are just on these things. I'm trying to figure out how you sever the recommendations from the transmittal. Your point is a good one.
If we actually sit here, we can't agree because of the structure of the process. We didn't try to suggest the department's management solution although there is a range of options discussed. Then, make a much stronger discussion on not just the transmittal, but an additional set of recommendations that the committee advises the department.
It's then the department's role to advise the committee on how to do its business. That's the whole point of the department. It's not currently NCHS's role to try and tell the department how to organize and govern health statistics. It's not the Data Council's role necessarily. It hasn't been asked how to reform itself.
MS. GREENBERG: It's the committee's role.
MR. HUNTER: In this, we basically say, you have to do this and everybody including the Data Council said you have to do this because this is a political discussion. It's not appropriate to try to resolve it in that kind of forum.
DR. LUMPKIN: Let me see if I understand.
MR. HUNTER: We can be more explicit and say we could not agree because of the nature of the process.
DR. NEWACHECK: The Secretary and the experts couldn't agree on something.
MR. HUNTER: But we didn't attempt to agree on it.
MS. GREENBERG: That's the point. Recognizing that this did not seem appropriate to try to get agreement on.
DR. LUMPKIN: I'm sorry for tossing a stone in the works. When we issue a report to the Secretary, I don't delude myself that the letter goes to the Secretary's desk, ponders it, and tries to make a decision.
It's going to go to wherever the relevant organization. They're going to write a briefing paper or some sort of short synopsis about whether or not it makes sense and it will go along through the channels.
MS. GREENBERG: It will go to the Data Council.
DR. LUMPKIN: If the Data Council has already been involved in what's on there, it becomes a strange recommendation. I don't believe stuff that comes from the Data Council goes directly to the Secretary either. It goes through some sort of channels on it's way to a decision that there is a decision to be made.
The question is: What is our aim with this particular document? How can we best achieve that? If we make it a document of the committee, it would go to the Data Council. The Data Council would say, yes, we participated in this process.
There are certain recommendations that we agree with full out and others that bear evaluation, the 1.1. NCHS would do its review, we're onboard with the other recommendations and 1.1 is something that has to be decided. That's how I see it being managed during the process. I'm not sure the benefit of having the document start the process with endorsements by the Data Council and NCHS as opposed to getting the endorsement as its working its way through.
MR. HUNTER: That doesn't relieve the lead staff on the report of any responsibility or involvement.
DR. LUMPKIN: We have pulled in experts and people who have helped. The project has been a joint project. It still could result in a report solely by the committee.
DR. STEINDEL: Forgive my naivete on this. Do we know what the position of the Data Council will be on this?
MS. GREENBERG: We could just take your strongest recommendations to the Data Council and say can you support any of these?
DR. STEINDEL: My suspicion is that we'll see it split. I don't think the Data Council is too strong on its own survival.
MR. HUNTER: This is a tough time to predict the Data Council because it has such a transition in members and most of the people are not political appointees or representative of the policy position.
MS. GREENBERG: It doesn't have strong leadership.
MR. HUNTER: This recommendation is going to be difficult for the Data Council to deal with. It's going to go into the political chain in the department if it gets dealt with.
DR. STEINDEL: That's going to happen with every recommendation.
DR. FRIEDMAN: There are two options for proceeding. One is severing the recommendations chapter, having that be NCVHS. Another is severing the first set of recommendations and having that be NCVHS or having a general recommendation, and then having the more specific one come from NCVHS. The third is severing the entire report. I have difficulty evaluating it because I really don't know if you can spend half of your time on this report.
MR. HUNTER: That's probably secondary.
DR. FRIEDMAN: I think it's secondary, but we also need to get it on the table.
MR. HUNTER: We let Claire and Ed make the call. I personally don't feel that if they want it to be a joint report, that's fine. We want to strengthen health statistics. I would rather a strong recommendation about the role and the place of NCHS.
DR. FRIEDMAN: I think we need a strong recommendation here whether or not it's NCHS or a newly conceived.
MR. HUNTER: The committee will be able to make that recommendation. It's really going to be the call of whether we do it as an add-on, transmittal, or whether we take over the whole report.
MR. HUNTER: If you feel the committee has in mind to agree upon a clear, strong recommendation, that is the reason for severing. If you don't see a consensus among members of the committee on what that is in a way that would be strong enough or different enough than what we might do jointly, then it doesn't make sense.
DR. NEWACHECK: That's the only reason for doing it that way.
MS. GREENBERG: I always felt the involvement of the Data Council was a phantom involvement.
MR. HUNTER: I think he wants there to be a good report that has strong recommendations. I'm still wondering if there isn't a single thing you think the committee would recommend that is strong, that is worth recommending.
DR. NEWACHECK: I feel like this is a real opportunity for us at NCVHS to make a strong statement about NCHS. I would really like to see us do that. That's part of what our mission and obligation is. I would hope we could do that in a way that would make it seem to you that we'll be willing to give up authorship in a second to have report that was so strong in our favor as an agency. If we can't do that, I agree with you fully. There is no point in going further.
DR. LUMPKIN: If there is any disagreement on m part on the strength of that recommendation, it has to do with not overworking role that NCHS should play. It's whether or not they do all of that. I think they should do more and there should be more integration, but I'm not at the point of saying they should do everything. There is a fine point that we would work out in the committee in making that strong recommendation. I don't see that we're going to find too much problem in getting a strong recommendation like that through the full committee.
DR. STARFIELD: They don't have to do everything. They do have to have a role in coordinating. Other agencies can do their own data collection. It has to be in coordinating with the main data coordinating function.
DR. NEWACHECK: This is really what NCHS should be doing. It's a public service agency. AHRQ is a bunch of researchers at a university. That's what this is really about. I feel very strongly that this is our obligation. This is a real opportunity and I would like us to seriously consider using it.
DR. FRIEDMAN: I have no difficulty having a strong recommendation saying that there should be either a single health statistics agency or there needs to be a lead health statistics agency. We need to be careful of how we formulate it. I would not want to see it viewed as NCHS.
DR. STARFIELD: We don't have to even name NCHS.
DR. NEWACHECK: It's named whether you name it or not.
MS. GREENBERG: If you don't name NCHS, are you suggesting should NCHS be significantly strengthened? If it's not NCHS, then it should be weakened.
DR. STARFIELD: We're just saying it should be one agency.
DR. NEWACHECK: It would be like AHRQ's name every three years.
DR. STARFIELD: Like CMS, a new name.
MR. HUNTER: This is the next place for the discussion. I would suggest that we don't have a decision to sever until you finish that discussion. If there isn't an agreement on who you say what you might want to say, there is no point in trying to say it yourself.
We can agree that something needs to be done and somebody needs to think about it. The issues are going to come down to how strong, what does this agency functionally perform as opposed to what does it coordinate? Management versus matrix discussion.
How do you fund something like this in a sustainable way? Where is it organizationally located? What pieces do you take or leave from one place or the other? The leadership issue, the connection to governance boards like the Data Council and others. It's all of those things.
This department debated those in a very formal way in 1995. There is a history of paper that has laid out options that you could pick back up and look at. It does require that those things be sorted through. They become tricky. It becomes the mess is clearly one you look at being a lot like the surveys that are multi-purpose health statistics. What about grant support?
MS. GREENBERG: Current beneficiary survey.
MR. HUNTER: What about topic specific surveys that have the same characteristics and approach as a general purpose that can only focus on one population group? What about surveys that NCHS participates in that are funded by others, what do you do with the funding stream?
There are a lot of very bureaucratic issues that have caused this issue to get hung up and not be acted upon or changed. That's why the Data Council was created. That doesn't do it either. I don't want to minimize the complexity of some of these things.
DR. STARFIELD: Was there ever a time that NCHS functioned as the data collection agency? Were things better then?
MR. HUNTER: There are two different ways to measure that. One, NCHS's proportion of the data collection going on in the department used to be very large. It was 1960 when we started HIS. Most of what the department did was at NCHS.
Now, it's probably 10 percent of that's done in statistics. We still probably do most of the multi-peropus, large-scale surveys, but others have caught up. The 90 percent hasn't come at the expense of HIS except in the opportunity way. People have been able to get money for MEPS and so on.
DR. STARFIELD: Take a look at HEALTH US. Any table with multiple data sources, they don't agree. The statistics are completely different. There is no relationship. That can't be good.
MR. HUNTER: My point is it isn't that NCHS has gone down, it's that others have done more. It's things like having the discussion of the surveys that give drug abuse information. One is a grant, one is a federally administered scientifically grounded survey, the other is the school-based survey. There are very different reasons why we have them that way, but what do you do about that?
DR. LUMPKIN: It's like any sort of missionary thing. Those of us who think that data and health statistics are important and that was a positive development, but it has led to a dispersion of inconsistencies among data sets. This report is a call to action. What has happened is good, but it needs to be better coordinated so that the data actually matches up.
That gives a different role for what NCHS used to be. They used to be the doers because no one else was doing it. Now, they need to be the coordinator and the standard setter and other kinds of standardization roles. They ought to be the data quality gurus for HHS so that the studies and analyses are done in some consistent way and there is an external to the other agency's evaluations which talks about moving them up into a different role than where they are now, stuck down within CDC.
MS. GREENBERG: The question is: Do you want to pursue this in sending this out for broad review? Isn't that the next step? It's chapter 5 as well? The real question is whether you want to include anything in chapter 5 in this nature and get feedback or let it go in this homogenized.
DR. NEWACHECK: It seems as though there needs to be conversation between John and Dan and the two Eds.
MS. GREENBERG: It was my understanding that it was really CDC that made the case against NCHS being a separate agency.
MR. HUNTER: There is more to it than that. They had argued that, but that wasn't the only one. It was an important factor.
DR. FRIEDMAN: Let me get back to a couple of procedural points since we do need to adjourn. There needs to be another conversation. It would be possible to send out the entire report including chapter 5, but rewrite recommendation 1 by leaving it a general recommendation and saying here are options that we would like feedback on from the people to whom we're sending the report.
We've gotten feedback from the members of the workgroup and some others. I would 99 percent of the comment has focused on 1-2 recommendations. I would hate to lose the opportunity to move forward to the others.
Secondly, I agree with what Ed is intimating about the level of detail that would be necessary for a recommendation on structure to be helpful. This is something that clearly we heard from Phil Lee saying you need to be specific if it's going to be helpful.
Having said that, I really question whether or not we could possibly put something together that that's specific without spending a fair amount more time which may or may or not be something that we want or are able to do.
DR. STARFIELD: Because of more hearings or more discussions?
DR. FRIEDMAN: More discussions, more work, looking at the history and structure. I don't have that kind of knowledge. It's a lot of work.
MS. JACKSON: It's almost another report.
DR. NEWACHECK: That's operational.
MS. JACKSON: I was concerned about the report being veiled by this incredibly important issue. There is so much here. How much do you want to put this on the back burner as you thrust this banner of a process versus content?
DR. FRIEDMAN: The recommendation could be that we need to integrate and we need to develop mechanisms and that's the next step.
MS. GREENBERG: That's a variation.
MR. HUNTER: That is what 1.1 says now. I don't want to go back full circle, but I wasn't saying that we need to have named data sets that have to be done, but to say what is it you're bounding in general terms? John was reacting to some of what you said.
Is it based on an existing organization or are we talking about changing the way the department views its statistics, management structure, and organizational structure? If you don't go at least that far, all you're saying is the department ought to think about doing better than it's doing.
MS. GREENBERG: Stats Canada and what they do in Australia for some guidance. You agree to put something in about internationals.
DR. FRIEDMAN: I put something in, but I'm not sure it's enough. I've been doing some work on chapter 5 like putting in references and more stuff to one of the recommendations. I'm not sure that it would meet your criticism.
We agree on a couple of things. One, we need have a conversation around this issue. Second, I'm suggesting dumb number 1 down saying that here's the issue. There needs to be more coordination within HHS. There needs to be more coordination.
DR. STARFIELD: Why they don't work now the way they play out or why what we have now doesn't play out.
DR. FRIEDMAN: I'd like to have an editorial discussion of that. We dumb number one down, put the whole report out to comment, but in number one, say here are a variety of options, please give us your feedback.
DR. LUMPKIN: That conversation should have been with those parties. You need to make the decision about who's report it is before we send it out. I don't know what restrictions you have in sending out draft recommendations prior to them having gone through departmental clearance.
DR. STARFIELD: There is no point in going through that business about whether it's a joint report until we decide what the recommendation is. It may be a moot point.
DR. LUMPKIN: I'm judging from the four of us in this room. There may be some others who have some concerns about this, but the issue isn't whether there is a strong recommendation or whether we mention NCHS, it's just how strong that recommendation should be in regard to doing it. That's a lot stronger than anything that could go through with the partners.
DR. STARFIELD: You know what the recommendation is going to be?
DR. LUMPKIN: I don't know the final recommendation, but it's strong enough that there needs to be some clear consolidation of authority or responsibility into a single entity which ought to be NCHS. We're nibbling around the edges of that.
DR. FRIEDMAN: The problem I have with the concept is I don't want it to be seen as this is NCHS now and we want to make it like this. We have these functions within NCHS and we have additional functions that NCHS has not been doing which are standards and conceptuals. We need not only have a bigger pie, but a pie that has a very different taste and shape than what we have now.
DR. LUMPKIN: I'm there. We're trying to refine that. The big idea is not to give them more surveys. It's a much broader level of responsibility. That's what I need to take to Ed and Claire. This recommendation would sever that.
DR. FRIEDMAN: That conversation has to happen before we do anything.
DR. LUMPKIN: We can proceed along and it's just an issue of how it goes out for discussion.
MR. HUNTER: I agree. It would be awkward to send out the report with 1.1 dumbed down from the three parties.
DR. FRIEDMAN: We just need to set that up and I will talk to Ed either before or after he talks at our meeting and raise it with him and maybe we can set up something around the next Data Council meeting.
DR. LUMPKIN: I am out of commission for around 5 weeks starting in mid-April.
DR. FRIEDMAN: The next Data Council meeting is March 13th. If we can get that resolved then, then we should be able to proceed rapidly. We can do a conference call before a face-to-face right before the Data Council meeting.
DR. LUMPKIN: I'm wondering if we could schedule another conference call for you and Gib and the workgroup and Rob and anyone else who wants to join. There is an editorial issue around chapter 5. One that concerns me is that I'm hearing that we shouldn't just have the recommendations, we should have a problem statement of something. I'd like some suggestions and guidance and discussion of do we need a one sentence problem statement for each recommendation and so on.
(Whereupon, the subcommittee adjourned.)