[THIS TRANSCRIPT IS UNEDITED]

As all of you know, Don is off at a whirlwind tour of the Southeast. In Chicago, when we say the Southeast, we mean like Hegewich(?) and the steel mills. But I think he is out in Southeast Asia. It takes a little bit further to get to, but perhaps equally challenging health issues.

My name is John Lumpkin and I guess I will start off the introductions. We will go around and introduce the committee. I would like to remind everyone that we are going out on the Internet. So, please speak into the microphones because even though we may be able to hear you at the table, the people out there in CyberLand will not unless you are speaking directly into a microphone.

My name is John Lumpkin and I also director of the Illinois Department of Public Health and a member of the committee.

MR. SCANLON: I am Jim Scanlon. I am from the Office of the Assistant Secretary for Planning and Evaluation and I am the executive staff director for the committee.

DR. TAKEUCHI: I am David Takeuchi, Indiana University, Sociology, on the Populations and 21st Century Committees.

MS. WARD: Elizabeth Ward from the Washington State Department of Health.

MS. FYFFE: Kathleen Fyffe, Health Insurance Association of America.

MR. GELMAN: I am Bob Gelman. I am a privacy and information policy consultant in Washington.

DR. MOR: Vince Mor, Brown University.

DR. FITZMAURICE: Michael Fitzmaurice, Agency for Health Care Policy and Research and liaison to the committee.

DR. IEZZONI: Lisa Iezzoni, Harvard Medical School, member of the committee.

MR. FANNING: I am John Fanning. I work in the Office of the Assistant Secretary for Planning and Evaluation of HHS.

MR. BRAITHWAITE: Bill Braithwaite from HHS. I staff the committee.

MS. WAGENER: Diane Wagener, director of the Division of Health Promotion Statistics of NCHS.

DR. STARFIELD: Barbara Starfield, Johns Hopkins Medical Institution, member of the committee.

DR. COHN: Simon Cohn, Kaiser Permanente and a member of the committee.

MR. BLAIR: Jeff Blair, Medical Records Institute and member of the committee.

MS. COLTIN: Kathy Coltin, Harvard Pilgrim Health Care, member of the committee.

MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics, CDC, and executive secretary to the committee.

DR. FRIEDMAN: I am Dan Friedman from the Massachusetts Department of Public Health and I am really not late.

MR. MELAMED: Dennis Melamed(?), editor of Health Information Privacy Alert.

MS. CROM: Debbie Crom(?), National Center for Health Statistics.

MR. KOLODNER: Rob Kolodner, Department of Veterans Affairs.

PARTICIPANT: [Comment off microphone.]

MR. HITCHCOCK: Bill Hitchcock, HHS.

MR. HANDLER: Aaron Handler, Indian Health Service, staff on the Subcommittee on Populations.

[Further introductions off microphone.]

DR. NEWACHECK: Paul Newacheck, committee member, University of California at San Francisco.

DR. LUMPKIN: I guess, Bob, you had a --

MR. GELMAN: I just wanted to make a brief comment. I wasn't at the last meeting. I am sorry I wasn't here. As far as I could tell, the committee took a couple of actions and I just wanted to indicate my position on those.

The committee voted on several letters with comments on NPRMs. Had I been here, I would have voted in favor of those letters.

The committee voted on a concept paper on insuring a health dimension for the National Information Infrastructure. Had I been here, I would have voted "no" on that.

And the committee voted after a wonderful discussion that I thought was the funniest thing I have read in years, to rescind the guidelines for dealing with the media and other external organizations. Had I been here, I would have voted in favor of rescinding the guidelines.

And I want to make one other comment. I looked very diligently for a long time since the meeting for the minutes of the meeting on the Internet, which are not there.

MS. GREENBERG: The meeting of the September meeting?

MR. GELMAN: Yes.

MS. GREENBERG: They are not put on the Internet until they are in final form. They aren't put up on the Internet in draft. They are in your agenda book. It has just been sent out in draft --

MR. GELMAN: I understand. In order to find out what happened at the last meeting, last week, I had to read the entire transcript of both days of hearings. If the minutes are available, whether they are approved or not, they should be on the Internet. They are a document of this committee and, therefore, are public, whether they are approved or not. And it would have been great value to me and perhaps to others to see what happened at the meeting.

MS. GREENBERG: May I respond?

First of all, you can always call me if you are looking for something because I certainly would have been glad to send you the draft minutes. We have been informed by our general counsel that we are not obligated to put anything on the Internet, that this is completely at our own discretion, whether we use the Internet.

The draft minutes are available for inspection at NCHS if anyone wants to see them. But we don't put them up on the Internet until they are in final form.

MR. GELMAN: Well, let me tell you that the advice you have gotten from your general counsel is wrong. The Electronic FOIA Act requires you to put things up on the Internet and that under the law all working papers of this document are public and, therefore, are available. And if it becomes necessary to get documents on the Internet, I will begin making Freedom of Information Act requests for them, which will then oblige you specifically under the Electronic FOIA law to make the documents available.

DR. LUMPKIN: The next item on the agenda, Dr. Hamberg will be here slightly later. So, we are going to begin with the report -- the other part of the update from the Department.

John.

Agenda Item: Department Update

MR. FANNING: Just a brief report on what has happened in the area of privacy. The Congress ended its session without enacting health record confidentiality legislation. At the very end, Senator Bennett introduced his bill, S-2609, and we are looking forward to vigorous debate in the next Congress.

The other item that the committee might want to be aware of is that the European Union Data Protection Directive went into effect on October 25th. There have been discussions between the Department of Commerce and the European Union privacy authorities about the effect of that directive on data flow from Europe to the United States.

Under the directive, data may not flow to countries outside of the European Union, which do not have adequate protection for personal information. Now, that is something of an oversimplification. There were many ways in which data can be sent, but the Commerce Department is at some pains to ensure that the data flow isn't stopped.

They have published on their Web site a document they are circulating, which is called Safe Harbor Principles. The concept of this is that an American company could sign on to these principles, state that they were going to observe them and then would -- that would create a presumption of adequacy in terms of Europe to U.S. data flow. They are on the Commerce Department's Web site, although somewhat difficult to find.

We did provide copies to the committee, as well as the copy of the EU press release that came out at the time of the effective date. We have worked closely with the Commerce Department to ensure that nothing in the Safe Harbor Principles would interfere inappropriately with data flow necessary for public health and other health transactions.

That is all I have.

DR. LUMPKIN: Any questions?

[There was no response.]

Good job. Thank you.

Bill.

MR. BRAITHWAITE: In the area of standards, I haven't got a whole to report since last time. The security NPRM comment period has closed. We now have four NPRMs, whose comment period has closed. We have begun to organize, classify, summarize the various comments that we received and try and put them in shape so that we can understand what the issues are and bring them through the departmental process for debating those and for deciding on what the final decision should be issuing a final rule.

The number of comments we got was less than some expected and more than others expected, but I still don't have final numbers on all of them, but, for example, we have got over 17,000 comments on the transaction NPRM and over 5,200 comments on the NPI NPRM; fewer than those, as I understand, on the others, but I don't have numbers on that yet.

The comments that we got, by the way, we are putting up on the Administrative Simplification Web Site. Due to technical problems with our contractor, it has taken quite a bit of time to get them up there. I think we have all of the e-mail comments up now and are in the process of scanning the paper ones to put them up as well. That will take a number of weeks more.

The fifth NPRM that we expected to put out this year on the payer ID is in clearance. There are a few issues that need to be resolved. At this point, I am reluctant to say it will be out by the end of the year, although that is what we were hoping for; so, probably early next year.

Then the process of trying to decide whether or not we should have a unique identifier for health for individuals and, if so, what kind of identifier, that this committee held a hearing on, that issue was addressed by Congress in the Omnibus Budget Report that they used to provide funding for the next year for our department. And in there there is a statement that says that the Secretary cannot use any funding to issue a final rule on what the unique health identifier for individuals would be or to implement such a standard until Congress passes an act, which approves that standard.

The department has not yet made a decision about what that means in terms of our activities in that area and we will let you know when we know what is going to happen.

Thanks.

DR. LUMPKIN: Any questions?

[There was no response.]

The projected time frame for release of -- or the issuing the final rules?

MR. BRAITHWAITE: The time frame is very hard to predict at this point. It depends on when we get these comments finally summarized exactly how complex the issues are and who we have to consult to try and work those things out.

Someone who has had experience in this area before believes that we won't have final rules prepared for clearance until April of next year. I am hoping to get at least one of them done faster than that, but I just don't know.

DR. LUMPKIN: So, my understanding is that the two year time frame clicks in either at the date of inaction of these -- or when the rules become --

MR. BRAITHWAITE: When the final rules are --

DR. LUMPKIN: -- or when they are published. One of those two?

MR. BRAITHWAITE: Right.

DR. LUMPKIN: A number of the comments that I have seen summarized relate to conflict with timing on people who have to do Y2K conversions. So, essentially, that makes that issue moot because if the two years goes into effect, then the rules will not be in effect before early in 2001.

MR. BRAITHWAITE: That is correct.

MS. GREENBERG: Could I just clarify? When you were speaking in terms of the final rules being available by April, that was, it was my understanding, ready for clearance within the department and the government said that they wouldn't be published until sometime afterwards.

MR. BRAITHWAITE: That is correct.

DR. LUMPKIN: I just wanted to establish for those who may be concerned about trying to make that conversion that that, in fact, HHS is being very responsive to that request, whether intentionally or unintentionally.

Any other questions for Bill?

[There was no response.]

Okay. Thank you.

MR. SCANLON: While we are waiting for our presenters, let me just make a couple of comments about the agenda. The next few items relate to data needs towards some initiatives in race and ethnicity data. One of them involve an HHS initiative, part of a presidential initiative to eliminate racial and ethnic disparities in six specific areas.

And we will be hearing a report there from Tony D'Angelo and Jack Anderson. They have been looking at -- they chair a working group that is looking at data needs to support measuring progress on these disparities.

In addition and in a related area we will hear discussions on some of the data issues that have been raised in a set of regional hearings around the country, looking at planning for Healthy People 2010.

Lisa?

DR. IEZZONI: Would now be a good time for Barbara maybe to talk about the letter that we would like the committee to send?

MR. SCANLON: Let me see. We don't have it set aside for any other time yet. Maybe at least a --

DR. STARFIELD: I think that is the action item that is referred to under the --

DR. IEZZONI: Yes.

DR. STARFIELD: I think it is, isn't it?

MR. SCANLON: It would normally follow the item itself, but --

DR. IEZZONI: Right. I am just wondering -- to fill some time.

MR. SCANLON: Yes.

MS. WAGENER: Or perhaps I could start just talking about the -- I don't exactly know what Olivia had planned in these sessions because she was in charge of it, but perhaps I could start because I think I have some points that are relevant to what I think is your letter, which I received by a fax this morning. So, maybe I can help inform that.

MR. SCANLON: That might be a good idea.

Agenda Item: Healthy People 2010

MS. WAGENER: Okay. Let me explain that the Division of Health Promotion Statistics is responsible for organizing the monitoring data for Healthy People both 2000 and 2010 and also providing statistical advice to the Office of Disease Prevention and Health Promotion.

So, hence, we are heavily involved with Healthy People and because the eliminate health disparities, which is the other session this morning, derives some of its major goals from Healthy People, we are also involved with that particular activity. So, perhaps I can help you inform it.

The Healthy People activity has had a series of regional meetings -- and, again, I apologize for not being familiar with what you already know about this activity, but we have had three of five regional meetings where we have gone to Philadelphia, to New Orleans, in Chicago and plan to go to Seattle and Sacramento to ask for advice and input into the proposed Healthy People 2010 process.

Now, there were three activities in these meetings where we get advice with regards to data issues; one having to do with the eliminate health disparities activity; one having to do with public health infrastructure and another session on data and surveillance.

Unfortunately, they were scheduled concurrently. Now, I have run the data and surveillance and I can tell you the feedback we have been getting in those sessions, but Mark and Olivia were going to tell you about the other two sessions.

We have had quite a bit of statement and local interest. Healthy People has really expanded over the past 20 to 30 years and looking at it from a federal perspective, I think that there may be more interest outside the Federal Government than there is within the Federal Government in this activity.

So, the state and locals have in the last decade come to us and said, listen, we want to be in the middle of this. We want to understand how you are doing these calculations. We want to be able to produce our own data.

They use the objectives for planning. They use the objectives for motivation in their local areas to get them to initiate activities. They use what we call the health status indicators, which is the set of measures that was developed by a special committee for Healthy People 2010 that identified measures that gave a good reflection of the health status of a group, a community, be it a local community, a state community and the like, and were obtainable in all these different levels because the majority of the objectives are not obtainable at a local area.

So, this kind of discussion is something that we have had quite a bit in our sessions and that is what kind of data are needed at the local level, given that the Healthy People 2010, which has been proposed, has over 530 objectives, of which very few can actually be obtained at the local level.

The discussions, I would say that there is no consensus on what kind of data are needed at the local level and I will give you a flavor for the conversations. We have talked about health status and there has been -- there have been arguments on both sides, saying that many of the health status objectives that we have in 2010 are for very infrequent events.

An infrequent event is not very informative if you are talking about a clinic population or a local population. So, yes, some health status variables are important and others are not.

Risk behavior is another set of information that many locals would like to have. On the other hand, there had been extensive discussions in our meetings about the fact that risk behaviors without set in the context of the social environment are not particularly meaningful; that is, you may have -- let's use an extreme example -- you may be interested in getting the frequency or the percent of the population that have gotten mammograms, but if it is a very young population, it is not particularly meaningful.

Another relevance would be if it is an area that doesn't provide those services. It wouldn't be particularly meaningful. So, we have had arguments on both sides of risk behaviors and whether or not that is the focus for data collection.

And then evaluation statistics in programs are another major area and there have been a lot of expressed need for that kind of data and then the flip side, that people are very concerned that whatever they have a statistic, it be able to be linked scientifically to the impact it has on the particular health status variable.

So, we have had, as I said, a series of discussions with local and state and there hasn't been a real consensus on the kinds of objectives they would like to see. I think that speaks to the fact that they would like to see -- providing them with a panoply, like we have in 2010 is probably the best way to go.

We have spoken to them quite a bit about the issues that are going to impact them in the near future, such as the change in age-adjustment standards, the change in ICD coding and the change in the race ethnicity classifications. The main concern that has been expressed almost universally in these meetings has been a concern on communication. The local and state representatives are very concerned that when we change in any of these three areas, the public have a clear message on why we are changing and why there is going to be an impact on the statistics and the fact that we can still follow trends over time, even though there has been this change in statistics.

So, they are very concerned about the communication that go out prior to these three major changes; that is, age adjustment, ICD and the racial ethnicity change in the categories.

With regards to Healthy People 2010, we sent you copies of our yellow book. Barbara, do we have enough copies to at least hand out at the table? Okay. I apologize. I got a request to bring these this morning and our xerox broke down, which is inevitable, but I think I have enough copies for the table here and we will get the remainder of the copies available.

What I am having made available to you -- and I want to focus on the large table first -- is a table that describes the 2010 objectives as they are right now in this draft document, so you can get a feel for the document. Now, I think it is probably best if you put at the top of this table that it is 11-12-98, because guaranteed these will change before the final document comes out.

What you will see in these tables is that the current volume has 531 objectives, but we have very smart people in the Federal Government and they have snuck in two measures on the average per objective. So, in fact, we have over 1,031 separate statistical measures that are going to be followed as part of Healthy People 2010.

As I said, this is what my division is particularly responsible for. So, it is going to keep us busy for awhile.

Of the 531 objectives, approximately 40 percent have no data currently. This is an issue that we will want to be bringing to the committee and working with the committee on an active basis over the next couple of years to try to identify how we can fill these data gaps.

In October, the steering committee met and we have begun a process with their advice of talking with the major data systems and trying to organize some work groups that can identify a pattern to the types of data that are missing. We would like to be able to work with both the Data Council and the National Committee on Vital and Health Statistics to identify ways to feel these data gaps. This is a cross-departmental initiative. We need cross-departmental help on this and we need outside help.

Only about 40 percent of the objectives are carried over from Healthy People 2000. What you see on the first page there of this large table is by each of the chapters, the number of objectives, number of measures and a variety of other characteristics about them and actually in -- this is labeled Table A. In Table C, it gives you a description of what those acronyms stand for.

But one of particular interest is in the middle of the page labeled "Race," and that is the number of proposed 21st objectives that have currently identified race as one of their demographic breakouts. And you will see that about 132 out of the 531 objectives -- that is about 25 percent -- actually have identified race as a breakout.

I will caution you on this particular statistic. You really should look at the next page and the reason is that the first page includes all the objectives and the editor made a decision in the document that whenever there was a developmental objective, they would not list any of the breakouts that they were particularly identified in this process.

As you recall, 40 percent of the objectives are developmental. Hence, this is a little bit of a misnomer to look at this page. I would say look at Table B, which is the count, again, only for the measurable objectives. Now, why they made this decision, I don't know, and they may want to revisit that decision. But I think -- I was not involved with it. My impression is they made the decision because it had a not of blank pages.

But if you look at the number -- so, the first page is a little meaningless. If you look at the second page, again, there are 128 with race of the 301 measurable objectives, which is about 40 percent. So, 40 percent of the objectives have identified race as a breakout; 17 percent have identified socioeconomic status, which is either income or education determined and about 5 percent have identified people with disabilities as a particularly important breakout group.

The next table, as I said, gives you a description of the variables and Table D is a comparison chapter by chapter of the sociodemographic breakouts. This particular table, I believe, is done with the total counts and not with the measurable. So, I would caution you on that table

The final table, which I wanted to bring to the committee's attention, Table E, is a list of all the data sources that we will be using in Healthy People 2010. There are approximately 190. This is a massive effort to combine data from many departmental sources, from many non-departmental but federal sources and from many non-federal sources. As I said, my division is responsible for working with all of these data systems to identify the data that are needed to follow the Healthy People 2010.

Approximately 94 objectives, which are in the last page did not have any baseline data or data system identified. These are some of the data gaps that we need to fill, in addition to the data gaps for identified systems where we would rely on putting together supplements to collect the pertinent data.

The breadth of this data collection activity is also its problem, of course. For any one data system we don't obtain a vast amount of data. I think about the most heavily used system is the National Health Interview Survey, where we have 68 objectives that pull from that particular survey. The vital statistics system has 54. NHANES has 34. So that for any one particular system we don't have a vast amount of data and yet we need to impact those systems to assure that the data are collected on an ongoing basis so we can monitor Healthy People 2010 over the decade.

Many of these data items are supplemental. They are not necessarily from the core. So, we need to bring together resources from across the department in order to sponsor these supplements because there is no one federal agency that reflects all the data needs here.

Finally, we need to be able to assure the generation of special data collections, like the primary care physicians survey or the work site survey, that are done on a special ad hoc basis. So, I think that this will be a continuing message that we will come to the committee and we welcome your active involvement.

Just briefly, the other thing that I handed out was an outline for what is going to be Volume 3. The yellow book that we sent you is being called Volume 2 of Healthy People 2010. Volume 1 will be a very short policy document. Volume 3 is intended to be a document on operational definitions and information about the major data systems so that people who are statistical staff can reproduce the objectives and the statistics so they can monitor their own particular populations.

I would encourage you to look at what we have outlined here and would very much like to have your input on the kinds of information we should include in this volume.

DR. LUMPKIN: Okay. We are going to -- because we are kind of playing with the agenda, we are going to combine the two panels on the data needs to eliminate race and ethnicity in 2010. We will have a discussion of Diane's comments at that time.

Right now we are going to have a presentation from Dr. Hamburg. Welcome.

Agenda Item: Data Council Report

DR. HAMBURG: Thank you.

And, you know, I think that probably we can blend my comments into that discussion, too, if you would like.

I am delighted once again to be with you and the continuing collaboration that exists. As co-chair of the Data Council, I think that the ongoing activities between this committee and the Data Council is the focal point or data related in terms of the department is a very critical I guess I might say vital one. And I really am deeply appreciative of it.

As you probably know, the Data Council has celebrated now its third anniversary within the department and it really, I think, is serving the mission that the Secretary had originally defined for it, which was to really serve within the department as an important coordinating and advisory body to her on issues of data and privacy issues, as well.

She not only wanted to strengthen the internal capacity of the department to look at data issues that are maximized, existing resources and plans for the future, but she also wanted us to strengthen our ability to work collaboratively across the administration and also with medical and public health groups and industry. And I think we are making strides forward in that arena as well.

In accordance with the Secretary's charge to the Data Council and the subsequent requirement put upon the department by HIPPA, the council has been focusing on six priority areas: data standards, health information, privacy issues, HHS data collection strategy, the ongoing relationship with this committee, national health information infrastructure issues and internal and external HHS-wide policy coordination in data policy, importantly, including international issues, which I think we are getting into in a more significant way as we recognize the international nature of data exchange and some of the emerging issues around differing policies with respect to data and data privacy.

I think that you just had an update on HIPPA. So, I won't say much about that, but it obviously -- implementing the requirements of HIPPA is among our highest priorities in taking up quite a considerable amount of the Data Council's time and attention and also within the department, more broadly, with a lot of back and forth between our data and privacy experts and our Office of General Council.

But there are other important areas of activity as well that I want to talk about. One thing that I think is going to be very, very valuable is that the Data Council has just about completed an update of the HHS Directory of Data Systems. This, you are probably familiar with. It was last updated in 1995 and really focused on race and ethnicity data. This time, while still continuing that focus, it really is broader and contains -- the new directory will include virtually all HHS-sponsored data collection systems and analytic databases of general interest.

We hope to be able to really get this directory out there through all the various dissemination mechanisms that we have access to, including the Web site, Internet, et cetera, integration with the communication technologies of other related organizations and activities, including, hopefully, your Web site.

I think it will be very valuable and will not only just include an inventory about, you know, what kinds of data sources are available, but more detailed information about how that information is available and information about how to really access it and utilize it. So, that will be an important resource.

Also, since the beginning of the Data Council, there has been a major focus on survey integration. I think, as I understand the history of the Data Council, that was actually one of the motivating forces of the recognition that we were doing a lot of activities that involved major surveys that were pretty labor and cost intensive.

They were sometimes difficult from a communications, public relations point of view for the Secretary's office to handle in that they would present data on similar issues differently and the press and the public, you know, would ask, well, you know, you just told us that youth trends and drug use are this and such and now you are telling us that. You know, what is the answer? She wanted us to look in a comprehensive way at how we were collecting data, through what mechanisms and how could we, you know, really integrate and consolidate in an effective way or what new areas really did need to be developed.

I think a lot of good work has gone on and is continuing and we, at our -- I think, the next, the November meeting of the Data Council, we are going to look at the status of progress in survey integration and the plan. This really focuses on looking at the status of survey coordination and integration with the population-based surveys, things like the National Health Interview Survey and the Medical Expenditure Panel Survey and then employer-based surveys.

Similarly, we have really been concerned about the issue of the department's efforts around race and ethnicity-related data collection and as part of the President's race initiative, part of our ongoing activities as well, have heard a lot of comments and recognized that we need to have ongoing modifications in this arena and there has been an ongoing effort, which I believe is just about in its final stages to really take a comprehensive look at this at this, you know, fairly complex set of issues and that report from the working group on race and ethnicity data will be presented to the Data Council this fall -- we are getting towards the -- sometime in the very near future.

And I think importantly what this will do is not only will it provide us some sense of where we are, it will look at how have we responded to past recommendations, because this isn't a new issue and there has been a lot of work done and I think there has been a lot of frustration, frankly, that recommendations get made, not just in this arena, but in many others and it is, you know, a book on a shelf, and we really want to look at, you know, the work that has been done and how have we acted on it, but also to identify new and emerging data needs.

In many ways, the couple that I have mentioned -- and more broadly, we really are trying to take a strategic planning approach to data issues and have developed a five year plan to address data needs, coordinate activities, identify resource needs and minimize duplication. As part of that, we have been working with the budget office to try to take a cross-cutting look at data needs and opportunities across the department, help to prioritize what from the budget office looks like a sort of confusing array of needs and projects so that we can continue to have this more integrated approach that not only recognizes the cross-cutting nature of many of the issues, but also reflects the fact that to do many of these things right, it does come with a fairly high price tag.

In that we have limited resources, we need to use those resources in the wisest and most way possible And as part of this long term agenda setting, Dr. John Eisenberg, who co-chairs the Data Council with me, and myself will be sitting down with the Secretary in the next few weeks to really later out for her this longer term agenda, get her input in terms of our priority setting process and really chart the course, hopefully, not just for the years that remain in the current administration, but as part of our effort to really make sure that the Data Council is an institutionalized part of the department, not just part of this Secretary's vision and appreciation of data needs, but really integrated into the structure of the department, how it functions, how resource decisions get made and how we think about critical problems.

So, I think a lot has been going. Certainly, the staff that supports the Data Council has plenty to do and it does seem that events keep serving us up an neverending set of issues and concerns. Many of those overlap with the work that all of you on the committee doing.

I welcome the chance to answer some questions and have a discussion. If you want to integrate in some of what was just being discussed by Diane, you know, I would be happy to engage in that as well.

DR. TAKEUCHI: I am new to the committee so I apologize if my question is naive. But I was wondering about the issue of survey integration. Is that more internal within surveys that would be done by the Federal Government or would it include studies that are conducted external to the Federal Government, say, by, you know, universities, who may be doing special surveys of their own?

DR. HAMBURG: Well, I think the particular survey integration plan that I was referring to is internal to the department and it really is to try to look at the fairly broad set of activities going on in the different operating divisions of the department.

The point about integration with other externally driven surveys, I think, is an important one and there is some of that that goes on in the sense that some of those surveys done by outside entities are funded through our department. So, we are aware of them and can see the opportunity for that. Some of them were not. There clearly is a need to make sure that we know what is going on in the given arena and that we utilize those investments wisely.

I think the welfare reform area is one example that mostly, to be frank, what the Data Council focuses on are health-related issues, both because the department is increasingly focused on health, the transfer of social security, et cetera, but also because recent years have put a new set of issues and mandates on the table for us through HIPPA, et cetera.

But, welfare reform, obviously, has created an important set of data collection issues for us and we feel very strongly that monitoring what is going on with respect to the impact of welfare reform, beyond just counting, you know, the number of cases that come off the case load is critically important and because of the way that welfare reform has happened with the state-driven programs and the way that evaluation dollars, et cetera, have been made available, it is very important -- they are one example that we integrate surveys that are being done a state by state basis, with broader data collectivities, data that is being collected for administrative purposes by states, et cetera.

And it creates some real challenges because not all of the data is collected in the same way with terms that mean the same thing to different groups, et cetera. But I think we are going to be doing more and more of that as we proceed further along the path of devolution, for one thing, but also as foundations, private sector and state and local governments become more engaged in running and evaluating programs.

DR. TAKEUCHI: Thank you.

DR. LUMPKIN: Vince.

DR. MOR: I want to know whether the Data Council is the right place to address a question like this. The National Center for Health Statistics periodically does analytic work regarding the quality and reliability of its various data collection strategies, survey, record abstracting, et cetera, and those are very valuable to the field of people who worry about those kinds of things.

Many of the federal agencies and many areas within the Health and Human Services collect administrative data for any number of different purposes and they are imbedded in many of these objectives, these administrative data sources and, yet, I haven't seen -- and I may be wrong -- any commensurate level of interdepartmental or even intradepartmental efforts to understand issues around data reliability, validity of these administrative data.

If we are going to be increasingly relying on those, that might be an interesting agenda, which is probably appropriate in terms of lessons learned across the various agencies. Is that the kind of issue that might arise as something within the Data Council?

DR. HAMBURG: Well, certainly, it is, I think, an important issue and I think you are right that it is a gap that raises some concerns and concerns that will grow. You are talking about in some instances data that is collected outside of the department, though.

DR. MOR: Or even within, but you are right -- for instance, the Bureau of Labor Statistics for certain kinds of data. The Health and Human Services, CDC, has other kinds of data. Health Care Finance Administration has masses amounts of data. All of those -- some agencies, some components -- while I know certain lessons that will be learned on vital statistics that might be applicable to other kinds of things.

DR. HAMBURG: Well, I do think the Data Council, you know, is an appropriate place, can begin those kind of discussions and some have, I think, gone on already, but I think, you know, in a more targeted way, you know, particularly the sort of lessons learned and the warning signs that can be communicated is important and will -- and working outside of the department with data collection systems is very important. I mean, a lot of what we do depends on data that is collected outside of the department.

DR. LUMPKIN: Paul.

DR. NEWACHECK: Dr. Hamburg, I wonder if we could talk for a minute about the data needs for evaluating and monitoring the state Children's Health Insurance Program.

DR. HAMBURG: Yes.

DR. NEWACHECK: There are a number of us on the committee that are concerned about the absence of data at the -- at least population-based data at the state level that would be needed to monitor these programs and to distinguish those that are successes from those that are not and learn from those successes.

I am just wondering if you could tell us a little bit about what the Data Council is thinking on this and whether there is a strategy emerging as to how data might be collected.

DR. HAMBURG: Well, I think, you know, the point you raise with respect to CHIPs is, you know, really the same concern that I was describing concerning welfare reform also; that is, you have programs that are increasingly state run, designed and monitored. We have this problem in terms of our ability to really make certain requirements for data collection and reporting to ensure consistency of data and, you know, to try to develop the kind of cross-cutting analyses that are really important.

The Data Council has certainly been involved in looking at this question and we don't actually have a committee, a subcommittee.

MR. SCANLON: Well, there is another -- at the December meeting of the Data Council, there is an HHS plan for monitoring and evaluation for CHIPs. It actually is five different elements. That is going to be discussed at the December meeting of the Data Council, but it includes surveys. It includes administrative data. It includes performance measures, but it is quite difficult, as you say, when the program is implemented in each state somewhat differently.

DR. NEWACHECK: One of the things that we recommended on this committee in a letter that was sent to the Secretary in September is use of the state and local area integrated telephone survey, the SLAITs data collection tool for monitoring these programs at the state level because it seems like it could be rapidly implemented and relatively inexpensive compared to in person interviewing.

Is that something that might happen, do you think?

DR. HAMBURG: Well, that has been discussed. I didn't get the sense that it was relatively inexpensive, but I guess compared to --

DR. NEWACHECK: Right.

DR. HAMBURG: I mean, I think that, you know, many share your sense that that may be, you know, the best mechanism to try to collect some of this data. I don't think that there have been, you know, final decisions made, but it certainly has been discussed and I think we will come back to that when we have our discussion of the issue at the December Data Council meeting.

It is really important that put it together and try to figure this one out because this is the way of the future or, at least, you know, for the immediate future, whether it is children's health insurance or welfare reform or probably a set of other public health programs and other human service programs. So, we have to figure it out.

I mean, even within the department it is very interesting because there is a real tension between those that are more in the business of evaluation and analysis and those that are in the business of running programs, those that run the programs are very loathe to put more demands on states than to ask for more in the way of data collection and those that are interested in the kinds of questions that you were raising really feel that we could go a little farther in terms of pushing states.

But the whole question of how to use administrative data, which is going to be collected anyway, but is collected for very different purposes, and, you know, sometimes I think with different meaning, is very important and I think, you know, we have real concerns that there is a sense that certain data is being collected when really the information doesn't bear on the real question that we want to ask.

So, I think that is another evaluation strategy we are going to have to think through, going back to the question that was asked.

DR. NEWACHECK: Just a final point on the SLAITs from our committee perspective, one of the things that we were thinking about is that the SLAITs data collection system would allow for collecting uniform data across the states for federal evaluation and for learning about which programs were in doubt, but also by having a component that would be state specific questions, that would be designed by the states. It would be able to generate information that would be useful for the states as well in monitoring their programs and fine tuning them.

So, anyway, thank you.

DR. HAMBURG: Well, I appreciate -- I am sure that that letter did cross my desk, but maybe stimulated by your question, I will get it pulled out again and review it in time for our discussion, as well.

DR. NEWACHECK: Thanks.

DR. STARFIELD: I would just like to amplify that conversation on the SLAITs. I still think our letter addressed the sampling strategy, but in terms of making this affordable and having comparability it is really going to be critical, how you select the sample. It seems to me that what you need to do is target the areas where these kids are going to live, not the whole state.

So, I think that is probably the biggest challenge from the federal point of view.

MR. SCANLON: The concept of a CHIPs-like survey is included in the HHS -- the monitoring and evaluation plan, but there are some complexities in terms of what you can get from a CHIPs specifically to measure sort of the marginal increase in the number of children, who were attributable to that program. So, I think there may be a look at a more robust -- potentially more robust way of approaching surveys, but the CHIPs is clearly being considered for the population-based plan.

DR. STARFIELD: I think there are probably only about three or four million children who are potentially eligible for --

MR. SCANLON: But these are relatively small samples and both the measurable issue and a statistical power issue, I think, come up with SLAITs.

DR. LUMPKIN: Okay. Thank you very much.

DR. HAMBURG: I am going to depart but I really do value the opportunity to work with you and have the chance on individual matters to connect with members of the committee. And, again, I am sure you know it is true, but if there are issues that you feel need to be brought to the attention of the department and the Secretary, the Data Council or that you simply, you know, want to ventilate on, you know, do feel free to contact me.

Thank you.

DR. LUMPKIN: Thank you very much.

Next we are going to move to -- again, we are going to sort of combine the two panels that we have and sort of jointly discuss these issues. So, Tony and Jack, did you guys have a preference on who went first?

MR. ANDERSON: I will go first. We will both be fairly brief and then take comments.

Agenda Item: Data Needs for Initiative to Eliminate Racial and Ethnic Disparities

We have been asked to talk about the data needs for the Initiative to Eliminate Racial and Ethnic Disparities. There are actually two activities. Tony and I co-chair the subgroup within the department for the data needs for the disparities. Tony and Olivia co-chair a work group that reports to the Data Council that is looking more broadly at the needs for minority health data and a strategy for meeting those needs within the department.

Our two groups have sort of been meeting together since we are very related. Briefly, let me go back to the initiative just to refresh, in case you are not clear. The department has really selected six areas out of the 2000 and the 2010 for a particular focus over the next ten years.

Those are infant mortality, cancer screening, cardiovascular disease, complications due to diabetes, HIV, AIDS and immunizations. There are work groups within the department for each of those six. The effort is co-chaired by Peggy Hamburg and David Satcher. And there is a seventh group, which is the data group that has been set up to look at how do we monitor the success of these efforts.

By definition sort of, the data group is lagging behind the other six because we are waiting to look at what the programmatic initiatives are before we can determine what the actual data needs are. I heard Peggy's admonition that we are late. We will say that is because the others are late and we are following behind them.

We have structurally taken from our work group and put a liaison on each of the six substantive groups. So, in effect, we are working integrally with the substantive groups as they develop their plans to look at what kind of data -- first of all, to define the problem and then, two, as they come up with their interventions, to look at what kinds of data we will need to measure success or lack thereof.

A second or -- as part of the initiative and as part of the 1999 budget, there is $10 million that will be available to actually do some demonstrations in communities. Thirty communities was the number we have been using over the last year and originally the budget request was for $30 million, $30 million in 30 communities. Ten million made it through the budget process, but the department is in a process also of seeing whether or not we can't find supplemental funds to bring that 10 million back up the to the original 30 million.

So, part of our charge will also be when the grant announcements are made for these 30 communities to look at sort of the best science of how you would eliminate disparities within these communities is specifically what the -- what will be the measures to look at baseline and then to look at progress in those 30 communities.

Another idea that has been put forward that we haven't addressed directly yet is whether we should do assessment ahead of time to look across the country and identify areas that have particularly bad records, if you will, in disparities and target the grants toward those areas or whether it be a totally open solicitation.

I will turn to Tony now because as I indicated he is co-chair with me on the disparities, but he is also co-chair with Olivia on the Data Council that is looking more broadly at the department's initiatives for the need for minority health data and we intend actually to have a joint report of the two groups, rather than two individual reports.

So, Tony, if you could talk.

MR. D'ANGELO: Yes, the other group is the Working Group on Racial and Ethnic Data and it is a subgroup reporting to the Data Council. Actually, there is overlapping membership between the Data Council working group, as well as the -- and the disparities data group. This is because a lot of the same people are interested in the two topics. And actually there is only so many people that get involved in these things anyway. So, they tend to be on the same committees.

The council has charged our particular working group with developing a long term plan for the department and improving the collection and the use of racial and ethnic data. So, therefore, you know, our charge on our working group sort of overlaps with the disparities charge, even though it is more short term, the disparities charge.

So, as Jack said, we do plan to produce a joint report. What we have been doing on our working group, though, is we have had each agency address their own specific racial and ethnic data needs before our working group. We have done that by giving them a set of questions that they have been responding to to kind of guide the discussion.

If you would like, I could, you know, give you a copy of those questions and, in fact, the response that Indian Health Service had to those questions as an example. We are also, of course, looking at what has been identified in each of the six focus area reports because in each of those reports, as Jack has said, each of those groups have identified various data gaps and data needs and we have from our disparities data group a person working with each of those areas to help them to come up with those parts.

But we are extracting that information as well. We are going to look at, you know, what is coming out of the disparities initiative, what we are coming up with with the testimony before our own working group and using all that information to come up with an overall plan for improving the racial and ethnic data in the department.

Actually, the plan, I am sure, will go to a certain extent beyond the department because of the disparities part. Probably our first step will be to come up with the outline of the report and that is not as easy as it sounds. There are various ways of cutting this and there has been some discussion on how to address it, but that will be our first step and then, of course, we will write the report, which will probably take a number of months.

I am not expecting a report until probably the spring of 1999 at the earliest. Jack --

MR. ANDERSON: Peggy said we will have it in the fall of this year.

MR. D'ANGELO: I guess no Christmas holidays. Well, we may have to speed it up a bit, I guess.

That is all I have to say, but we would be glad to answer any questions.

DR. LUMPKIN: We will take a couple of questions and then sort of begin to move into the -- blend it into the next panel.

Are there questions?

DR. TAKEUCHI: I was wondering, when you talk of the data needs for looking at race and ethnicity, does that include trying to tease out some of the underlying differences within those categories? For example, if you take Asian Americans, trying to understand, just collecting how immigrant status and nativity might understand better that Asian American category better or is it strictly focusing more on how to -- what is the best way to categorize racial and ethnic groups?

MR. D'ANGELO: It goes to the point that you raised. It goes way beyond the categorization and so on. It is looking at all aspects.

DR. TAKEUCHI: And it is also looking at different ways of measuring SES, maybe beyond income and education?

MR. D'ANGELO: Right. We are looking at the socioeconomic part, acculturation, looking at other possible reasons for the disparities, perhaps, you know, racial prejudice, things like that.

DR. TAKEUCHI: Thank you.

DR. LUMPKIN: I have a question about the initiative to reduce disparities in communities. How are you defining communities?

MR. ANDERSON: Olivia, you probably are -- I will if you don't, but I think you are closer to that than I. We have talked about the $10 million that we have at the moment for the grants for communities. I don't know that we have actually defined what we are going to mean by that.

DR. CARTER-POKRAS: Well, actually, tomorrow on the agenda, I understand that Dr. Nicole Lurie is going to have an opportunity to meet you. So, she may have some late breaking information about how the grants are going to be given out, but at the earlier discussions, we had discussed maybe metropolitan statistical areas, different ways of discussing it and it hadn't been defined.

The Centers for Disease Control and Prevention is the one that is actually going to be administering those grants. So, hopefully, Dr. Lurie can provide more information or once it gets developed, we can provide that for you.

MR. ANDERSON: I think the answer is we haven't decided. So, if you people have advice on how you think that might best be done, we are certainly still open to that.

DR. CARTER-POKRAS: There is also the recognition that rural areas be considered in some way or forms and when we are defining communities, we need to think broader than just metropolitan statistical areas, for instance.

DR. LUMPKIN: I asked that because I am trying to in my mind equate 30 or even 10 million with 30 communities and comparing it to a program like Healthy Start, which really looked at a smaller number of communities and one health disparity issue and certainly consumed a much larger amount of money over a period of time and then trying to also figure out because if you talk about eliminating disparities, it depends upon what your goal is, whether you are going towards the national average for that health statistic, the community-wide average, which is why defining community becomes important because if you take Grant Boulevard in the south side of Chicago, you could eliminate the disparities within Grant Boulevard, but they would still be in big trouble.

Of if you take it across the SMSA, again, depending upon which one you pick, how much progress you may be able to achieve, again, may be variable. So, you know, I think it is a very interesting process, one which I would suggest that if we are going to spend money on doing that and we want to see some ongoing change, given the volume of money, our efforts or the federal efforts might be expended at measuring any initiatives in some sort of real way because you can do that within 10 or 30 million to actually say we are going to go into these communities and have selected this as a priority and then expend various amounts of currently existing federal funds to eliminate it. But we are going to try to actually measure what they are doing and why what they do works or fails.

DR. CARTER-POKRAS: I did notice Allison Orris is in the room. I don't know if she has any additional information to add. Allison Orris is from the Office of the Assistant Secretary for Planning and Evaluation. She has worked very closely with Dr. Margaret Hamburg, Peggy Hamburg, on this initiative.

MS. ORRIS: I unfortunately don't have much new information to add because we just received word that we do have the $10 million. I know CDC is in the midst now of developing their RFPs and I just got an e-mail saying by the close of business tomorrow, they want to make some decisions about numbers of communities, because I think we may choose to focus on fewer than 30 communities because we don't have the money and we don't know where else those funds could come from.

And in response to what you just raised about what sort of is a broad effect of this initiative, I think we have been looking at this as a multi-year effort and maybe using the community demonstration programs in a limited number of areas, spread out -- spread among the several different disease areas, to at the front end come up with a few interventions and then decide how they could be more broadly applied.

So, I think this is -- we are viewing this as a multi-year effort looking towards 2010, but the first few years, I think, are focused maybe narrower than what you were recommending. But, again, this is all sort of up in the air and no decisions have been made.

DR. CARTER-POKRAS: The initial budget request from the President actually was for 30 million. We received one-third of what was requested. So, that is why Allison was mentioning possibly cutting down on the number of communities that are going to be targeted.

This also was thought of as a demonstration grants program, so, not to answer all the questions. And I know in earlier discussions, there was a possibility that come communities may decide instead of focusing on all six health focused areas, to focus on, say, maybe one or two, ones that they really were the most concerned about, the highest priority for them.

MR. ANDERSON: I actually see it very much like you said, John. It is a demonstration and it is trying to show what actually works in an area so that others could pick that up if they are successful. I really doubt that anybody would -- any group would probably try and do all six. It would probably be much more focused.

DR. LUMPKIN: That is good to hear.

Agenda Item: Healthy People 2010

DR. CARTER-POKRAS: Mark Smolinski, Dr. Mark Smolinski is from the Office of Disease Prevention and Health Promotion and we all work very closely together, NCHS, ODPHP, OMH and the other entities within the department to help track what is happening with Healthy People 2000, as well as what we hope to do with Healthy People 2010.

There are various aspects that we wanted to present to you today. One is the general effort within the department to improve our data systems to provide both baseline and monitoring data for Healthy People 2010. And Diane, I am sure, talked to you about some of the internal discussions that have gone on about that.

Also, we wanted to touch on the presentations that have been made during the data challenges sessions at the five regional meetings, three of which have already been held; Philadelphia, New Orleans and Chicago. Also, the comments raised during the data challenges sessions, not only the normal presentations but the comments raised, as well as the comments raised during the eliminating disparities sessions and the public comment periods.

Then a fifth component are the written comments that have already been submitted. The written comments are actually not due until the middle of December, but some people have already been sending them in and they are being posted right now in the Office of Disease Prevention and Health Promotion on the Healthy People Web Site for review. And it has a very nice search capability so you can go in there and I did search for data statistics research, just to see what came out of that.

Presentations made during the data challenges sessions at the regional meetings covered the following issues. The changes in the International Classification of Disease for mortality, as well as morbidity or ICD-10, which I understand you have been briefed on. And compared to ICD-9, there are a couple of changes that we are sharing with folks. One is that we will see an increase in certain causes of death, in the numbers of deaths because we are broadening the definition, so to speak.

Cardiovascular disease and cancer mortality and a decrease in a couple of causes of death, chronic liver disease and cirrhosis and pneumonia and influenza. One of the questions that was asked during a presentations, it was asked whether we were going to recalculate mortality rates and numbers of deaths by cause because of these changes. And the answer from NCHS was "yes," because we are going to need to be able to cross walk and see if, indeed, there has been a change over time.

The next issue that was discussed during the presentation were the changes in classification of race, ethnicity by federal agencies, the October 1997 revision of OMB Directive 15, which I understand you have been briefed on. In addition to changes in terminology, the orders of the questions on race and ethnicity encouraging a two question format versus a single question format, creating a new Native Hawaiian and Pacific Islander category.

The new revision also allows persons to identify more than one race. This is the issue that people seem to be raising the most questions and concerns about because they are waiting to hear what the Federal Government is planning to do as far as presentation of the data.

As far as collection of the data, we already have those guidelines and that was the revision of OMB Directive 15. Now we are waiting to hear from the Office of Management and Budget as to what the guidelines are for the federal agencies for presentation of the data.

So, that was a typical question. Also a question that was asked was what happened to non-Hispanic whites and blacks. People were a little confused by the fact that the two question format is actually being encouraged and instead of having a single question format, which is non-Hispanic white, non-Hispanic black, Asian Pacific Islander, American Indian, Alaska Native. So, they were a little confused and they were reassured that we can still get that information, as well as information on Hispanics, whites and blacks.

NCHS also raised the issue that they have been asked to take a look at how we tabulate natality data and they were asked to consider tabulating natality data by race of the child. This may also be an issue of concern for you. Another question that was asked is about what about Middle Easterners. They wanted to know how Middle Easterners were handled with this new revision.

A third issue that was brought up during the presentations was the change in the standard for age adjustment going from the 1940 and 1970 standards that are currently often used by the Federal Government to change to the year 2000 estimate. So, that was fairly well discussed. People felt fairly comfortable with the decision that was made, but they did ask questions about why we don't compare age specific rates by race ethnicity if, indeed, the ratio of the age-adjusted rates are going to be decreased substantially.

They also suggested that in the Healthy People 2010 documents that we note very specifically when age-adjusted rates and numbers are used and what the standard is so that is clear to people who are reading the documents.

The fourth issue that was brought up during the presentations was the funding for data and the link to resources, both resources for the data and resources for programs. One issue that was brought up is that it is hard to garner funds for data collection at the local level, considering the fact that program administration has now sort of moved down to the state and local level, that we need to ask decision-makers what they want to know, the questions that they want to answer and we need to get buy-in from policy people.

The fifth issue that was brought up during the presentations was the need for national and state monitoring and surveillance and one question that was asked is how does the Federal Government plan to provide for state and local estimates, what changes are we seeing in the near future for our national data system. And they emphasized the need to have information on a community basis, particularly when we are talking about the eliminating disparities initiative, that is almost recognized as a real need.

The sixth issue that was brought up during the presentations was the need for more detailed studies of the relative importance of underlying factors, as you were discussing, Dr. Lumpkin. How do we examine all these underlying factors together. In particular, with the national data systems, often times we run into a small number issue. So, there was the general discussion that we probably need to have three -- at least three, possibly four, approaches for the eliminating disparities initiative and also for healthy people.

One is the need for descriptive information that is used for monitoring and surveillance, both at the national level and also at lower levels, state and local levels. We also need more detailed studies that look at the relative importance of these underlying factors.

Then, thirdly, we need also a better assessment of what works and what doesn't work. We are talking evaluation of the programs, what we are putting dollars into.

Then, fourth, which may tie into any of those three is the need to have very quick turnaround estimates, information on emerging issues, health issues that are out there.

So, one of the other issues that was brought up during the meetings is a need discussion of the underlying factors in the Healthy People 2010 documents. They felt that that was not sufficiently developed in many of the introductory sections and that we need to address underlying factors in the Healthy People 2010 objectives, actually create new objectives, whether they are developmental or ones that we already have baseline data, that the high school education are completion that we have and there seems to be the only SES one that is in there. We may want to consider including Healthy People 2010 objectives that target, for instance, poverty or unemployment or something else like that.

Other issues, which were raised that were outside of the issues raised during the presentations were how are targets set. In general, we heard a very clear support for single targets but there was at least one question raised about using the better than the best approach because they felt that this was unachievable. They pointed out the need for patient satisfaction data and noted that there aren't many tracking systems available to states and perhaps the national level for health services use and access. This seemed to be a general concern because we heard it more than one place.

There was also a concern about GPRA, the Government Performance and Results Act measures, because they felt that the GPRA measures tend to be outcome measures while the process is easier to measure at the local level. So, there seems to be a disconnect there. And it was very problematic for folks to address the Healthy People 2000 objectives in grant applications because of this, that socioeconomic status and disability status should be used to break out the baseline and monitoring data for the Healthy People 2010 objectives.

And you will note that Diane had done a very nice job of giving you a view as to how many objectives are actually broken out by particular groups. They also raised the need to have role health data and they raised a very important question, how do we train the states and others on these upcoming data issues. How do we get the word out? How do we plan to let people know of these changes?

They were very concerned about the time that it takes to make the changes; in some cases as long as eight years. One state person said that it took them eight years to make a certain change. For one thing is they had difficulty in even getting people to code causes of death and another issue that they brought out, which isn't quite privacy or confidentiality, and that is that certain agencies or organizations may feel uncomfortable using data collected by another agency.

There was the general issue about the need for aggregate, as well as detailed data, and almost the competition there for that. And another, which I thought was a very interesting issue is that it is important to find victories in our data because if we want people to continue to fund our data systems, we need to show not only how things are getting worse, but also how things are getting better.

We need to be able to say that something good has happened. So, that sort of gives you a flavor of what happened in the written comments, as well as the data challenges session. And Mark is going to talk to you about data issues brought up during the eliminating disparities sessions, as well as the public comment periods to date.

There is another one this -- tomorrow? At the public health consortium.

MR. SMOLINSKI: Correct.

We had a session at each of the regional meetings. It was up to each of the regions to decide what critical focus issues they wanted to talk about and eliminating disparities was chosen by all of the regions. As Olivia said, we had three of the meetings so far. The purpose of our session was not to focus on data because we have a critical issue session that did focus specifically on data, but I would like to highlight those data issues that came up nonetheless in the session on eliminating disparities.

First of all there is universal support in Healthy People 2010 to have a goal of eliminating disparities rather than reducing disparities as we had in Healthy People 2000. There was also support that it was necessary to put no data in every objective where racial and ethnic data was missing because this was the only way constituencies were going to get support for trying to collect that data, rather than just leave it blank.

In general, there was support for a single target using the better than the best approach. However, there were questions about when choosing better than the best, were there any international comparisons to say that these targets are even achievable in countries that have universal access to health care or are these targets just simply chosen arbitrarily.

There were also comments that the targets need to be realistic. What can really be done in ten years? It was not felt to be too realistic to choose that across all of Healthy People and consequently set up the vulnerable populations for failure right from the start.

There were concerns that we need to do a better job of explaining what we really mean by eliminating disparities. For instance, the constituency for the Native Americans were very strong in saying that they would be delighted just to get to the national average and to not make the targets even harder for them to achieve by setting better than the best.

Comments were made that in any society we will always have disparities between groups and how realistic is it for us to say we are going to eliminate all disparities in all health areas.

There are pitfalls to setting targets without any plans for getting there. Targets do not take into account the interventions that are necessary and what we can realistically expect to do within each goal. Comments were made that resources need to be allocated to the identified populations with the greatest disparities if we are serious about reaching our goal of elimination.

There was a great deal of confusion over the numerous data gathering systems in Healthy People 2000, especially when there were several for each objective. And comments were made that a single data source should be chosen for each objectives, so people would understand at the local level what they should be supporting.

In light of the proposed changes in the census, where individuals may choose more than one race or ethnic category, there were several questions on how this was going to impact on calculating denominator data. There were concerns about what do we do about undocumented immigrants. There were concerns about what does it mean for multi-ethnicity children, such as someone born from a black and Native American couple and how do the targets we are setting for Healthy People affect them?

When reporting out data on objectives that were not reached for the year 2000, it is important for us to take a look at where we started from for each objective. Since most of the targets are set arbitrarily by different work groups, it is not really fair to say a certain objective was not met when they were trying to have a 50 percent reduction as opposed to a target that only set a 25 percent reduction.

As Olivia stated, there was a lot of concerns with eliminating disparities, that we have no objectives on things, since as poverty and employment and also there was a lot of comments about elderly population being identified as the select population since this is an important focus for many areas.

So, that was in general -- most of the comments received had to do specifically with data. Of course, we have several volumes of comments that have to do with the goal of eliminating disparities and you can look at those on our Web site if you are interested in them as well.

DR. LUMPKIN: Thank you.

Questions, comments?

DR. CARTER-POKRAS: I did have a quote, in case you are interested, because I know that this was one of the things that was mentioned that you may want to quote one person or some other individuals that may actually emphasize one of the things you are interested in and this was about pushing -- well, pushing for having more information by education, socioeconomic status and that sort of thing, the underlying factors, that they really felt that we need to address. So, I will be happy to give that, I guess, to Barbara or anybody else, who wants to consider it as quoting it for your letter.

MR. SCANLON: Can I ask a question, too?

In the hearings, did the issue come up of perhaps following the examples set in the Healthy People 2000 framework for actually having a set of objectives dealing with separate objective area dealing with data, data capacity --

DR. CARTER-POKRAS: Well, actually that was something I should have mentioned. That was something that was brought up in Chicago. The public health infrastructure section has just two objectives on data and that is it. So, they were saying that they really thought that there should be a separate section on data, rather than having it buried within public health infrastructure because they felt that way it wouldn't get the attention that is needed and that some of the issues that they were discussing here, if we don't have the data, then that should be come a developmental objective.

DR. LUMPKIN: Vince.

DR. MOR: I have a question for Diane. I have gone through the listing of various data sources here and I may be wrong. It is difficult to know, but I have a sense that some data sources -- the format for the naming of the source and agency, et cetera, is not consistent and it makes it really difficult for a reader to know where it comes from. So, I am hoping that ultimately there will be a much clearer road map.

MS. WAGENER: Yes. The staff person who put this list together didn't feel like she could make the corrections and this just came off the press, so to speak, so I haven't gone through it. But that is clearly something that needs to be in the final document standardized.

So, for instance, there is the National Immunization Survey and separately they list it as a provider check for the National Immunization Survey. So, this is very clearly a draft. I did want to share it with you so you had a sense of the magnitude of the effort and how important it is going to be to try to get your assistance in making these issues global and having everybody understand that we need to work together to get these data.

DR. MOR: Now, the other question I had was sometimes these measures have -- appear to -- might have two different data sources in constructing a single measure, one for a numerator and one for a denominator.

Is that correct?

MS. WAGENER: That is right.

DR. MOR: So, some of the data sources provide -- all right. Never mind.

MS. WAGENER: And there is certainly a reluctance from the statistical point of view to do that. However, I might point out that even the mortality rates come from that kind of a ratio. So, there is precedent, universal precedence, in a sense, to provide ratios. There are some ratios in here that I feel a little uncomfortable with, but the programs that proposed the objectives felt that they were very important to have that kind of data, even though the data sources were different and there may be some uneasiness about the eventual ratio that you come out with.

DR. MOR: And a number of the data sources listed are -- don't, at least to my knowledge, appear to be necessarily repeated. So, the whole issue of how frequently they will be done and whether they will be done and whether data elements that may have been on a certain survey that occurred three or four years ago would ever be repeated in another one that might occur on an intercurrent basis.

MS. WAGENER: There are examples in the objectives where the groups didn't quite adhere to that, but by and large, this is a much improved document over Healthy People 2000. We have much more current data that have been proposed.

Obviously, all the numbers you saw in your yellow book, virtually all of them, will be updated before the final book comes out because that -- we, of course, still 14 months in advance of that. So, we will have even more recent data. By and large, they did not cite occasional articles, which was done in 2000. So, it is a much improved document for data.

DR. MOR: And the Volume 3, when do you expect to have some sort of draft of this Volume 3, which sounds like a very interesting resource?

MS. WAGENER: Well, one of the objectives in 2000 was to provide information so that they could be comparably produced at federal, state and local levels. We began late in this decade to do a series of statistical notes where we were providing that information.

Volume 3, which will come out at the same time as the other volumes, i.e., January of 2000, is meant to try to address these issues up front and if you look at the outline that I have given you in the handout and some examples of how we would propose to provide these data, you will notice that for the developmental objectives where I can't get anything out of the work groups, I have given the work group leads phone numbers and e-mail addresses. I am serious about getting some sort of specifications on what they mean by these particular statistics.

So, what you have in this packet is an outline for the table of contents, some -- and table of contents includes three major parts, one a general discussion of -- or a discussion of general issues, like age adjustment, ICD changes and the like. I don't have an outline or I don't have specifics of those and I will hope that we could share something with you, let's say, within six months on those sections.

The second section goes over each objective and operational definition. We provided you with examples that we provided the work groups. We have requested the work groups to give us operational definitions for all the measurable objectives by December 1st of this year. So, I am hoping we get some systematic response on that and can share something with you again in the spring.

Then the last section of the book would have discussions of major data systems and you will see a kind of table format that we are proposing for that. I would very much like to have you look at this on a considered basis and get back to us on suggestions that ought to be considered for inclusion here.

MR. MAYES: Bob Mayes, Health Care Financing.

Vincent, I just -- and actually also, Diane, I wanted to mention some other work that is going on that I think might be particularly relevant. The Census Bureau has been building a Web-based registry of all their surveys. That should hopefully be made available publicly soon and as an extension of that, there has been a collaboration formed by quite a number of federal agencies now, looking at this issue of data management across enterprises, based on like data registry -- meta-data registries and such. In fact, the third meeting of that is next week.

So, we are encouraging other agencies to build upon the same infrastructure and the purpose of this collaboration, in fact, is to increase the number of agencies using the common infrastructure and to go forward in a collaborated, coordinated way with this infrastructure. So, there is a lot of other interest, even outside this collaboration, in DOD, VA, but also includes Department of Energy, Department of Transportation, Department of Justice. So, it is not even within the health care alone that there is a big interest in this, in trying to develop some infrastructure to make this a lot more accessible than just a published document, because one of the problems is, as soon as you publish it, it is out of date and you really need to have.

MS. WAGENER: All of those departments you mentioned, we use their data in this --

MR. MAYES: Exactly. So, this is being recognized and we are trying to move forward with some actual implementable solutions to this.

DR. LUMPKIN: Is there an assessment that -- I was just looking at the sources of data for the 531 measures and about a --

MS. WAGENER: No, 1,000 measures, 531 objectives.

DR. LUMPKIN: I am sorry. The 531 objectives. About a third of them come from four surveys, the behavioral risk factor survey, the vital records, NHANES and HIS. But only the vital records and the behavioral risk factor give you state level data and only the vital records give you community level data.

So, it seems to me that we have a long way to go to be able to get the data that is specific to a community if, in fact, we are addressing the community-related issues.

To what extent do we have ability or a way to monitor because it seems that this issue came up for Healthy People 2000 and we really haven't made a whole lot of progress, but maybe this is something we ought to be monitoring between now and 2010, as a more concrete objective of the number of -- even just looking at the number of measures in the Healthy People that can be assessed at the local level.

MS. WAGENER: I think that would be an interesting suggestion to add an objective like that. Of course, one of the reasons that there isn't a systematic effort at the federal level to make some of these surveys expanded to the level where we could get all the local information is that not all the local areas focus on objectives. What has been done under 2000 is that when a local area focuses on an objective, they can develop a mechanism to collect the data.

Maybe it is expanding the BRFS in the state to get extra enrichment in a particular local area to get information on those particular objectives, but not any one local area adopts all 531 objectives.

DR. LUMPKIN: And I think that there are a host of reasons. For instance, the BRFS, even expanding that only gives you state level data and maybe if you have a major metropolitan area, unless you expand the survey, and then it is -- the cost is very heavy unless you do it on some sort of rotational basis, which we have been doing in Illinois.

But the other thing I think that sort of addresses the whole issue is that to what extent are we almost oversurveying the public because now with HEATUS and various other measurement systems, there are also surveys that are being performed and to what extent can we conceptualize some sort of unified process.

There are a lot of folks who are looking very seriously at implementing caps and if you have a 50 percent market penetration in a community of managed care, you could almost do the other 50 percent and have a population-based assessment to those particular measures.

So, there are some pieces here that require a larger perspective in the data collection. I think that would speak to a more focused and certainly a more unified focus within the 2010 objectives on data issues.

MS. WAGENER: Could I just point out that under the assessment initiative, we developed a mechanism that was -- we tried to keep the cost quite low so local areas could do their own survey for smaller areas than Houston, Texas or a large metropolitan. Unfortunately, the direction of the assessment issues is moved away from that kind of activity, but we did try in the last decade to develop mechanisms for locals.

DR. LUMPKIN: Okay. Other questions? We do have an action item under tab -- related to this under Tab F.

DR. STARFIELD: This discussion -- this proposed letter follows very naturally from the presentations that we just heard and I think there won't be a lot that is new in our letter because of the excellence of what you have done. Thank you very much.

I am going to in the interest of time not read the whole letter, but just go over the recommendations, but first to highlight the fact that we addressed it to Linda Myers and I guess there was some question about whether we should address it to the Secretary or to -- well, we perhaps need an executive decision on that.

We reviewed the whole document and in a sense provided a reliability check on the table you handed out this morning and are pleased to report that there is over 99 percent interrater reliability, which is amazing, given how you can count these objectives.

Okay. Recommendation 1, baseline and monitoring data should be provided by race, ethnicity and income or when not possible by some other proxy for SES for all objectives, including those that focus on program or governmental activities. If these data are not available for a particular objective, it should become a developmental objective. Each developmental objective should be accompanied by plans to obtain data by race, ethnicity and income at a minimum.

With regard to race, the racial, ethnicity category specified in OMB Directive No. 15 or its revision

should be used. Consistency across chapters in reporting of baseline data or plans to obtain it where not available should be attempted. And if you look at Diane's table, you can see very clearly that the consistency with which disparities are addressed in baseline data varies enormously, even when it is not obvious as to why that should be the case.

DR. LUMPKIN: Barbara, can we maybe take it objective --

DR. STARFIELD: Sure. Let me just point out one thing --

DR. LUMPKIN: -- recommendation by recommendation?

DR. STARFIELD: Sure. Let me just point out one thing is that this recommendation is based on a rationale and the rationale is based not only on the specific objectives in the different areas, but also on the introductions to those chapters, some of which address disparities and others don't or do so to a lesser extent.

DR. LUMPKIN: Any concerns on that? We are going to obviously move this letter as a whole, but I just want to make sure that we can focus the discussion on each recommendation. Okay?

No. 2.

DR. STARFIELD: Okay. A separate chapter -- this is on the third page of the letter -- a separate chapter on information systems and data needs should be added. We recommend that the HHS Data Council be directed to do this. The chapter should address obtaining data on disparities by race, ethnicity and income or proxy for SES where income is not possible.

This came up in the hearings, as well, I believe.

DR. LUMPKIN: Okay.

DR. MOR: Is that something you would recommend would go into the main volume or into a chapter -- Volume 3 or --

DR. STARFIELD: Yes. In fact, the Year 2000 Objectives do have a separate section on data and this one doesn't. It is buried in other chapters.

DR. LUMPKIN: Any concerns about that?

[There was no response.]

Okay. 3.

DR. STARFIELD: Okay. Three, the most likely source of primary -- the most likely primary source of data should be given for each objective because we noted, in fact, that for some objectives, secondary sources are given. Those clearly are not useful for monitoring, you know, compendia of data.

DR. LUMPKIN: Okay. Any problem?

[There was no response.]

DR. STARFIELD: Geocoding should be used when individual data on race, ethnicity and income are not possible to obtain, are not available.

DR. LUMPKIN: Okay.

DR. MC DONALD: The question is why wouldn't we always use it? Can't you always translate addresses in the geocodes now? I don't know this for sure, but --

DR. STARFIELD: We don't oppose that, but we notice that, you know -- we note that where there are no individual data, at least geocoding should be used.

DR. FRIEDMAN: Barbara, I would also suggest adding there a little bit more specificity about the type of geocoding; so, for example, perhaps block group where feasible. Because I am not sure that it is really a self-defining term. I think different levels of geocoding yield very different utility.

DR. MOR: That would be my concern is that the higher the level of geocoding, the less adequate that is as a measure and I know, as you do, that a number of these data sources have no geocoding history in them requiring probably fairly substantial revision for future collections, as well as addressing the important privacy and confidentiality issue.

DR. STARFIELD: Well, this is the problem with being awfully specific is you do get into these issues of privacy and confidentiality. Not only that, you get into which level is useful because there is a lot of literature now that suggests that the characteristics of the area itself influence the health characteristics of the area; the non-health characteristics influence the health characteristics. And the question is how large does that area have to be.

DR. MOR: But if you are making it as a -- if geocoding would be a substitute for racial, ethnic and income distributions, it probably has to be reasonably small.

DR. STARFIELD: It probably has to be reasonably small, but the data that you would collect would be, in fact, addressed, I suppose with appropriate privacy, confidentiality. And then you could aggregate it at any level.

DR. LUMPKIN: Kathleen.

MS. FYFFE: Geocoding, I assume, is some sort of geographic cut in order to code statistics. It sounds like a generic term. Are there any specific, you know, schemes that are more common than others and has any thought been given to using --

DR. MOR: That is what Dan suggested, block coding of the individual addresses.

DR. STARFIELD: And that, of course, could be aggregated to larger levels if you wanted to.

MS. WAGENER: Just a quick comment, which may be obvious and that is that not all of the objectives are population-based. We have, for instance, a number of school systems who do this and a number of states who do that. So, you may want to address whether or not you want to know, for instance, the race, ethnicity mix of a school system or some of these other non-population-based issues.

DR. STARFIELD: Thank you. We do think that geocoding is useful for all the data, the program data, as well as the individual data.

DR. LUMPKIN: Paul.

DR. NEWACHECK: Barbara, you might want to just consider defining the term "geocoding." It is obviously not clear to everybody what that means and then go on to the issue that Dan is raising.

DR. LUMPKIN: Perhaps if I could suggest that we could leave the recommendation where it is and then under the rationale enhance, because it is a little bit harder to try to buff up the recommendation when you can spend a sentence or two describing the type of geocoding and what use the ability to aggregate would be.

MR. SCANLON: I would think that we may want to even in the recommendations itself, Barbara, include some sort of a term that says to be commensurate with confidentially, protection or disclosure, and it will vary, depending on whether it is block or county or whatever, depending on -- so, just the general caution about confidentiality and disclosure.

DR. MC DONALD: I still want to clarify. Are street addresses and are person's addresses in many of these databases to start with? I thought they were.

MR. SCANLON: In population surveys they would not be, like the Health Interview Survey --

DR. MC DONALD: Well, they pick a person.

DR. MOR: Well, the Health Interview Surveys, actually they have a block. They have a block and the block is the exact geocode and it is just not accessible and available for anyone doing work with it.

DR. MC DONALD: What I would suggest is -- I mean, the geocoding is just a matter of your coordinate on earth. I mean, it is not a -- and you can walk around with one of these little machines and you can see where the hell you are and geocode it if you want by -- within 10 feet, I think, if you are near an airport.

But the key thing is, though, this is not new data collection if you have an address. There are programs now that will convert. I just worry still about if you don't have the data, you do this. Well, that usually makes a messy database. I mean, that would suggest we use geocoding. We have addresses so that we can verify all the other sorts of things, all within those same constraints.

DR. STARFIELD: In all data sets by the way they go back, they obviously have addresses.

DR. MC DONALD: They can go back and get it if they have addresses.

DR. STARFIELD: Okay. We will work with -- put something in the rationale.

Can we go on?

DR. LUMPKIN: Okay. No. 5.

DR. STARFIELD: Recommendation 5, maybe we don't have to make anymore because we know that Volume 3 is going to do this, a separate section for each major data source, including information on periodicity, inclusion of data on race, ethnicity, socioeconomic status and gender and, where relevant, disability and the quality of data with regard to representativeness, reliability and validity.

DR. LUMPKIN: I think we can say it perhaps and in the rationale that we understand that this is being done and we applaud that.

DR. CARTER-POKRAS: Actually, one of the things I would encourage you to do, because when we went to the regional meetings, this was an issue that came up. People thought that they only needed to provide comments on the things that they disagreed with. If you agree with something, also provide that comment to make sure it stays. So, if you like the idea of Volume 3, say you like the idea of Volume 3.

DR. STARFIELD: Okay. We will do that.

And the last recommendation is an observation on primary care, data for primary care, which is becoming increasingly recognized as important. The reason we make this recommendation is because we thought that was in the document is not state of the art with regard to primary care and, moreover, data are available. So, you don't have to develop new data. There don't have to be new developmental objectives.

The objectives relating to primary care should be revised to reflect current knowledge. These include objectives for access, continuity, comprehensiveness and coordination of services.

DR. LUMPKIN: Mark.

MR. SMOLINSKI: I serve as liaison for the office for that specific focus area and I think if you are going to make a recommendation like this, you need a little more detail on what you are trying to say. They work very hard in trying to find data sources for what they currently put into that chapter and I think if you don't think it reflects what is currently happening, then it would be much more helpful to include what specific things about primary care you would like to see in that chapter.

I think if you don't think it reflects what is currently happening, then it would be much more helpful to include what specific things about primary care you would like to see in that chapter because otherwise it is not going to be a very helpful recommendation to that group.

DR. STARFIELD: Well, that is the suggestion. I am not sure that is appropriate to do in this document. I think it is more appropriate to do as working with the individuals who are concerned about that. I have personally talked to David Atkins at AHCPR and AHCPR has got the data source to do this.

I think it would unnecessarily clutter up the document to put that in here, but I recognize you want us to do that.

DR. LUMPKIN: Okay. We have gone through the six objectives. Can we perhaps -- Dan?

DR. FRIEDMAN: I would just either under the rationale for recommendation 2 or perhaps as a separate recommendation -- and this isn't said to shamelessly pander to the acting chair, but adding specifically something about an objective going to tracking the number of objectives that can be measured at the state and local level because I think that is an extremely important thing to earmark.

DR. LUMPKIN: So, that would be a recommendation --

DR. FRIEDMAN: Or under the rationale for recommendation 2.

DR. LUMPKIN: So, we can include that under 2, including measuring -- we could do that, this chapter should address obtaining data on disparities race and ethnicity and should include a measure of those measures that are measurable. We should have a measured response in that.

How do we want to handle this as a committee. Would you like the subcommittee to bring back a revised copy or are we comfortable enough to work on the modifications and then just have the chairman review it?

MS. WAGENER: Delegate it.

DR. LUMPKIN: Okay. So, in that case, we have the really fun task of entertaining a motion for Don to do something and to sign a letter for something when he wasn't here, which is what he gets for being in Southeast Asia.

I thought I heard Simon make the motion. Is there a second?

MS. FYFFE: Second.

MS. GREENBERG: You could give that the Executive Subcommittee, which might be helpful to him, particularly since at least one member represents a state and that was some of the issues.

DR. LUMPKIN: Okay. So, the motion would be is that we would adopt a letter kind of the way it is. The revision would then go to the Executive Subcommittee and then signed and sent.

All those in favor say "aye."

[There was a chorus of "ayes."]

Opposed?

[There was no response.]

Abstentions?

[There was no response.]

DR. STARFIELD: Question: To whom are we addressing the letter?

MR. SCANLON: I think maybe we should send it to Dr. Satcher, Assistant Secretary for Health, and we will copy Linda. The idea is to get it into the works of the revision and I think that will work.

MS. GREENBERG: And continue to copy the --

DR. LUMPKIN: Okay.

I would like to thank our panel. It has been a very informative discussion and certainly we appreciate the effort that is being made on behalf of Healthy People 2010.

Thank you.

At this point we are going to make a break, just -- for those of you who have a copy of the agenda, you will notice that half of the subcommittee reports and the work group -- I am sorry -- subcommittee meetings and work group meetings don't have a room number. That is because they are here. So, the meetings that are in an alternative location, particularly Room 325A, have that listed by them.

We will adjourn to the work groups and the subcommittees. We will reconvene tomorrow morning at 9:00 o'clock a.m.

[Whereupon, at ll:05 a.m., the meeting was recessed, to reconvene at 9:00 a.m., the following morning, Friday, November 13, 1998.]

[THIS TRANSCRIPT IS UNEDITED]

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS

November 12, 1998

Proceedings By: CASET Associates, Ltd. 10201 Lee Highway #160 Fairfax, Virginia 22030 (703) 352-0091

TABLE OF CONTENTS

PARTICIPANTS:

Committee Members:

Staff Members:

P R O C E E D I N G S [9:05 p.m.]

Agenda Item: Introductions

Agenda Item: Department Update

Agenda Item: Healthy People 2010

Agenda Item: Data Council Report

Agenda Item: Data Needs for Initiative to Eliminate Racial and Ethnic Disparities

Agenda Item: Healthy People 2010

Proceedings By:
CASET Associates, Ltd.
10201 Lee Highway #160
Fairfax, Virginia 22030
(703) 352-0091