[This Transcript is Unedited]
Department of Health and Human Services
National Committee on Vital and Health Statistics
Subcommittee on Populations
February 23, 2006
Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington, D.C. 20201
CASET Associates, Ltd.
10201 Lee Highway, Suite 180
Fairfax, Virginia 22030
- Donald M. Steinwachs, Ph.D.
- Justine M. Carr, M.D.
- Robert H. Hungate
- A. Russell Localio, Esq.
- Carol J. McCall
- William J. Scanlon, Ph.D.
- C. Eugene Steuerle, Ph.D.
- Kevin C. Vigilante, M.D.
P R O C E E D I N G S [8:07 a.m.]
DR. STEINWACHS: Well, why don’t we get started. This is a meeting of the Population Subcommittee of the National Committee on Vital and Health Statistics. We are not on the Internet. We only are being recorded, but not broadcast, I understand.
I don’t know that everyone knows everyone. Maybe we can go around the room. I think it would be good. Edna, I understand you had a bad accident. I didn’t realize that you had been out. So, welcome back. It is wonderful to see you again.
I am Don Steinwachs, Johns Hopkins University, a member of the committee.
MS. GREENBERG: I am Marjorie Greenberg, the National Center for Health Statistics, CDC and executive secretary to the committee.
MR. HOUSTON: I am John Houston. I am a member of the committee and a visitor to this subcommittee today.
DR. STEINWACHS: We welcome you, John.
MR. REYNOLDS: Harry Reynolds, Blue Cross and Blue Shield, North Carolina. I am a member of the full committee and visiting today.
MR. HUNGATE: Bob Hungate, Physician Patient Partnerships for Health, member of the full committee and subcommittee.
MS. PAISANO: Edna Paisano, Indian Health Service, staff to the subcommittee.
DR. VIGILANTE: Kevin Vigilante, member of the committee, Booz, Allen, Hamilton.
DR. ELO: Irma T. Elo from the University of Pennsylvania. I am the liaison representative from the Board of Scientific Counselors of NCHS.
MS. GRANTHON: Good morning. My name is Miryam Granthon. I am the special assistant to Dr. Garth Graham in the Office of Minority Health and staff person to the Population Subcommittee.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, CDC, committee staff.
DR. SCANLON: I am Bill Scanlon from Health Policy R&D and member of the committee and member of the subcommittee.
DR. STEUERLE: I am Gene Steuerle from the Urban Institute and let me apologize in advance. I am about to attend a funeral. So I am going to leave a touch early. So, I apologize for my rudeness.
DR. STEINWACHS: Thank you for being here, Gene.
DR. STEINDEL: I am Steve Steindel, Centers for Disease Control and Prevention, liaison to the full committee and visiting this subcommittee and like Gene, I have to leave at 9:00.
MS. BURWELL: Audrey Burwell, lead staff to the subcommittee, Office of Minority Health.
DR. STEINWACHS: Thank you all.
MR. LOCALIO: I am Russell Localio, University of Pennsylvania, School of Medicine. I am a member of the committee and a member of the subcommittee.
DR. STEINWACHS: Thank you for joining us, Russ. I appreciate it.
Why don’t we look at the agenda for a moment, just to make sure that we are covering everything that needs to be covered today. There are three major items listed on the agenda. First is an update on the publication of the courts, a chance to talk a little bit about the distribution lists and other things we can do to try and maximize its impact and distribution to relevant parties.
Then we were going to turn to the two topics around which we had decided that this committee would hold hearings this year and look to make recommendations based on what we learn. The first deals with improving data linkages and there is an update on progress that particularly Gene and Nancy have made working on this included in your packet.
Then that will be followed by Bill Scanlon and Kevin Vigilante will be leading a discussion about our efforts to think through hearings on surge capacity or the capacity of the health care system to respond to disasters and bioterrorism. Then a wrapup.
Is there anything that should be added to this agenda that isn’t here?
Leslie, welcome. Good to see you.
Why don’t we kick off with a report of the — status report. Debbie knows all. I know that.
MS. JACKSON: There have been a number of questions that have been kind of floating along in e-mail. So, I thought I would just cover what I can so everyone will be on the same page as to where we are.
We have completed essentially the final galleys for the report. We are expecting a publication not the end of March. We have really got our fingers crossed on that. It really could and should have been, then at least the first week in April. We have a report to the Data Council that was originally scheduled in March and we thought it was a bit premature to have this great report to the Data Council without having something really to show them and give them the handout. So, we are really targeting April, early April so that by the Data Council presentation, we will be all set.
There were questions about the date of the report. There are two key dates to keep in mind. The date as it is currently posted on the web is August. That was the final date that it was approved by the Executive Subcommittee. I know that there have been questions. Could it be a later date to show more viability and be more current on the publication date. Well, the latest date we can use is November. We are publishing the date of the report as November because that is the date it was distributed and sent to the Secretary. That letter to the Secretary is posted on the web and you see it has a November date and that will be the date for the report. We can’t have a data report in March of 2006 if the report is distributed to the Secretary in November 2005. So, that will be the date of the report.
Again, the publication date, we are looking at the early part of April. There were also questions about the Executive Summary. Why couldn’t this have been completed on a parallel track, the report and Executive Summary at the same time. Well, the Executive Summary, I just abstracted portions of the report. So, it really is just a one track deal. We have to get the report gone through the editorial route functions and reviews and that is what has taken so much time. The Executive Summary materials will be extracted from that. I did talk to folks in publications and they are in good agreement that we can pull out essentially the same items that are on the Executive Summary posted on the web. They are going to be posting the background materials, the Executive Summary itself, the listings, the recommendations and that can be pulled together once the report is completed through its editing process. And that will be the Executive Summary.
The question now becomes what do you do in terms of trying to print an executive summary. Once this report is up on its feet and published, do we go back, extract everything out and create a printed copy? I am getting a little push back from publications in creating something like that right now. We can talk about this at a later time, but their preference is to go on and create this in electronic format and post it rather than have a hard copy.
We are focusing now on getting the actual report printed and I think that should really stay our target. It is going to be a very gorgeous report once you see it is all laid out. So, that is the status of the report itself and the Executive Summary.
The dates I mentioned. Were there any other basic questions? Leslie, you had some —
DR. COOPER: Can you tell me how many copies and what is the status of that distribution list?
MS. JACKSON: The distribution list is in your folder. We can talk about that in a second. The number of copies, the largest number we created for NCVHS publications, that is 2,500. Our usual number of publications is about a thousand, 1,500. The largest before had been 2,000. We are going to 2,500 on this one and that I think we can get our key people, the major groups from the minority conference. The main thing we need from that are labels. Sending us names is fine. We appreciate that, but if we can get actual labels, some kind of files that we create that we can pull together in our office because we do all the distribution ourselves. We don’t sent it out to be distributed. We all get together and just get the labels and the reports and the envelopes and just do a thing like that.
So, whatever you can get us label wise, it would be great.
DR. STEINWACHS: Miryam.
MS. GRANTHON: If it is specifically for the audience who attended the OMH summit, I can reach out to our contractor. I don’t know if we want to send to every single person. I don’t know if that was the idea because there were about 2,000 people and I don’t want to use all our copies — exactly. I can work with Audrey to be very selective about who actually attended, maybe the NCs, a breakout session of the presentation and who are the key organizations. So, I will make sure labels are sent to them.
DR. COOPER: There are definitely some key individuals and then even beyond the OMH conference is — you know, we have sent in names over the past whatever the period of time, but I don’t know the status. So, I don’t know if all the names that we submitted, I don’t know if they are on the list. I am going to have to double check.
Will there be an opportunity — will all of the committee members get x number of additional copies so that we can send them out, you know. I mean, I am just throwing out some questions. Will the documents go out cold to key people involved in HHS and, for example, like Offices of Special Populations or, you know, Centers for Health Affairs is will there be a cover letter that will go out with the document and what are we sending out to all those individuals that gave up their time in terms of coming to the hearing and contributing their information. Will we be sending them out a thank you letter and some type of a document saying like thank you for your services? We know that all of your concerns may not be in the document, but encouraging the partnership or whatever we want to do with it.
DR. STEINWACHS: Debbie, is it possible for the cover letter on —
MS. JACKSON: I think all of those are quite fine or suitable. The membership can get a certain number of copies, ten or something, to each member. We already have the mailing list of all participants. So, we have their names and addresses. We can always pull together a cover letter to them thanking them and this is a reflection of your hard work and, you know, however you want to phrase that, if you want to send some phrasing.
Our basic cover letter is kind of standard.
DR. STEINWACHS: I would think it comes from Simon.
MS. JACKSON: So, that is all part of the usual distribution.
DR. COOPER: — people there participating. I know one of the things is OMH conference. A number of people came up. Oh, is this the report from back when we had the hearings in —
MS. BURWELL: What we plan to do, Leslie, with the OMH summit participants is the contractor sends out periodic messages to everyone on the list and we are planning on announcing the availability of the report. They can get a hard copy or they can pull it down off the web, just like
— we only have 2,500. So, if there is a large demand maybe our office might be able to contribute some funding for another printing, but that is how we prepare to handle it.
DR. COOPER: You know, we had it on the web for the OMH conference. You know, we gave up a web site, but that was a draft. So it is kind of like how do you communicate — so, it is kind of like how do you let people know that this is the final copy. I don’t know if that is going to be —
MS. BURWELL: Like IOM does. IOM puts up a pre-publication copy and then, you know, many of us can’t wait for the book. So, we will print it page by page and then you still get the hard copy later. That is right. So, that is pretty much how we plan to handle it.
MS. GREENBERG: I don’t think we should consider the one that we submitted to the Secretary as a draft, but I would agree, a pre-publication copy is probably the way to term it. I probably should understand this, but I need a little clarification about the Executive Summary. Does the report that we are in the process of printing have an Executive Summary?
PARTICIPANT: Yes, it does.
MS. GREENBERG: Okay. So, what is the separate issue with an executive summary?
DR. STEINWACHS: I think the suggestion has been made that if we could print the Executive Summary separately, that you could distribute that much more widely. People are more likely to read sometimes an executive summary. Those who are really interested and involved will read the whole report. The other idea then on the web site to make it possible to download separately the executive summary. So, if you just wanted to see the executive summary, you didn’t need to download the whole report. So, those were suggestions and I think, you know, within resources and so on, I think Debbie was trying to respond to the difficulties —
MS. GREENBERG: It is not something different than what is already — a new document.
MS. JACKSON: And the report itself is so small. We really don’t see where we need a copy, a hard copy of the Executive Summary. The report itself is succinct and really very targeted.
DR. STEINWACHS: So, I guess if we could on the web, is it possible to have executive summary so you can download it separately? I think that would make it easier because otherwise are going to end up having to download the whole — Debbie, could we also — I think you have gotten a sense of all the enthusiasm, but it would probably help the committee if you sort of went through and made an allocation, sort of made estimates as to how many of the 2,500 and Miryam, I know, is going to work with Audrey about the — but it would be good to get out to the committee members once there is sort of an allocation of roughly how the 2,500 — and then I think I welcomed your idea of — and Leslie’s — that committee members could be asked if they would like additional copies to distribute themselves or not.
Then I would be happy to work with you on a cover letter for Simon and then get input from Leslie and others so that we are appropriately appreciative of people who came to the hearings and donated their time and energy. I remember the one hearing. It was from the islands way out in the West Pacific. It was like 2:00 a.m. in the morning. There were people on the telephone talking to us. Oh, my goodness gracious. There are only about 500 people on the island.
MS. GREENBERG: I think we need two cover letters, I guess. One that will go to those who came to the — who testified and then one to everyone else.
MR. HUNGATE: That island was Palou(?).
DR. STEINWACHS: Palou. Thank you, Bob.
MR. HUNGATE: If you are interested, there is an oceanic society that runs tourists to Palou so that you could onsite visit the folks on a more reasonable time of day.
MS. GREENBERG: Those of you who remember Lynette Araki was with the committee. I believe she lived in Palou for awhile, worked there.
DR. STEINWACHS: Well, just a side note, under Vicki’s leadership, which is really where the report came from and Vicki has continued to be very active in participation, but Vicki was trying to get us the opportunity to have the committee go visit these areas where there are, you know, small ethnic groups, who are part of the responsibility of this nation to deal with health and health status. But we never could quite get to Palou.
MR. HUNGATE: I will be happy to lead the trip there.
DR. STEINWACHS: You will lead the trip. Okay. I have been seen in skirts and coconuts, too.
Jim, welcome. We were just talking about the distribution of the report. I thought what you told us yesterday was you had a substantial allocation. If we spent it early in the budget year, we could go.
PARTICIPANT: Well, is it related to bioterrorism or —
DR. STEINWACHS: Of course. Palou is the western-most outpost of the protection against bioterrorism.
Are we ready to move on to the next topic? We will get information out — oh, I am sorry?
PARTICIPANT: The issue of the collection of race, ethnicity data came up in the review of the statistics that we did as part of the BSC for the —
DR. STEINWACHS: Would you talk into the microphone? Russell is on the phone.
DR. ELO: The issue of collection of race, ethnicity data came up in our review of the mortality vital statistics at the National Center for Health Statistics as part of the BSC review of their programs. One of the big concerns is the quality of the race ethnicity data that we are now getting on — particularly on death certificates because much of that reporting is done by the funeral director, without much consultation, really not with the individual but not even with the family.
So, I would suggest that — I looked at briefly at some of the recommendations you had here, that it would be very pertinent to have someone or maybe have some discussion of this report in the next BSC meeting so that it would reinforce some of the discussion that we had in January. Since this report just will have come out, I think it would be an opportune time to do that. We are not doing another review until September, I believe, so that there would be time to discuss this. It would be another way to reinforce some of the recommendations.
DR. STEINWACHS: We would really appreciate that invitation. Bill and I can talk about it maybe —
PARTICIPANT: — I think it would make sense.
DR. STEINWACHS: Then Jim invited a — we had already talked about it — a presentation at the Data Council, HHS Data Council, on the report. I think that is April —
PARTICIPANT: It would be the second Wednesday in April.
DR. STEINWACHS: I think the plan was for me to do it and have Vicki in by phone, I think, was what we were talking about.
Marjorie was explaining to me the cost of travel and the budget. But I wanted Vicki involved in the presentation.
MS. JACKSON: And the BSC has seen the basic recommendations from the report I think when Vicki was on before. I am not sure what date. So, we can get the background materials and make sure Irma and others know that this is not coming out cold. We had a copy that we — in pre-publication.
DR. SCANLON: Didn’t we circulate a copy to them in September or —
MS. BURWELL: They are also on the distribution list.
DR. STEINWACHS: Anything else on the report, Jim?
MR. J. SCANLON: Just to go back to the Pacific again, the HHS — this probably happened since the report, but HHS has been asked now to work with our colleagues at the Interior Department and we are actually going to sponsor kind of a health statistics analysis and profile. Actually we would like to get some continued building capacity in the Pacific jurisdictions as well. But we actually have a nice project underway. We ended up sending the money to NCHS, the vital statistics group, to kind of pull together from the various Pacific — this is American Samao, the Marianas, some of which have fairly well-developed or developing, you know, health data infrastructure, but others don’t really.
Many of them report data to NCHS and some of our other agencies, but, again, it is not clear, you know, what the quality is or it rarely gets pulled together, but at any rate, with Interior within the next month or two, it might be worth having those folks come in and brief this committee. That will be an attempt to kind of pull together the use of the data, not just collect the data, but use the data and build some capacity for applying the data in the health departments in those areas.
So, that should be — taken awhile to get going, but it is actually starting now. So, that is actually moving.
DR. STEINWACHS: That is exciting. Thank you.
Any other — why don’t we move on to the next topic? Gene and Nancy. Nancy is probably still fighting her way here. She told me 8 o’clock was a little early. I took the hint. So, she sent Gene. Nancy has just arrived.
I was telling them that you did point out to me that 8 o’clock seemed a little bit early. So, I took that as a hint.
DR. BREEN: Well, I had it in my mind that we were meeting at 8:30. So, my apologies. Have you started long ago?
DR. STEINWACHS: We were just talking about the report, about distribution and so on. Debbie is going to get out information in terms of what the allocation — there are 2,500 copies being printed.
Gene, you and Nancy, do you want to take over and begin to discuss — in here in the packet is an update report on conversations that Gene and Nancy have been having with people in agencies about some of the data linkage issues. The hope was to try and what would be, whether it is a workshop or a hearing — Gene and I were talking about, you know, how to label it, but an opportunity for the subcommittee to invite in people from agencies who are knowledgeable to try and understand what is being done and what progress is being made, where there are barriers, where we might contribute to trying to overcome barriers that limit the opportunity to do the kind of analysis that is needed to address our health problems.
DR. BREEN: Well, we sent out — did this get distributed or it didn’t get distributed? So, did people get a chance to read it or shall I kind of walk through it? Because we have sort of walked through this — and I don’t want to give a whole lot of detail if it is not —
DR. STEINWACHS: Why don’t you sort of walk through it just so that everyone is on the same page?
DR. BREEN: Well, Gene and I proposed the idea that we look at how we might use linkages better. It is something that we have been talking about in this committee for a long time and there are a lot of federal data sets out there and increasingly people are using geocoding methods or deidentified information with codes on them to link data.
It is such a cost effective and useful thing to do that we thought, you know, it would be a good idea to more systematically look at the federal data sets and see where we might be able to either encourage them, promote more linkage and also see what is already being done because actually there is quite a bit already being done, maybe not as systematically as we would like and maybe we could promote a more systematic linkage.
So, the hearings would be to bring together the different federal agencies that are collecting surveys or claims data or administrative data and see the extent to which these things can be linked. Is it just physically feasible is one question and then is it kind of I guess politically or culturally feasible is another.
Then there is the issue of confidentiality, which follows sort of somewhere in between, I guess, those things because we have strict regulations about confidentiality and they are important but they sometimes undermine the ability to co-mingle data on computers and that sort of thing. So, we had to take all these things into consideration and so we wanted to evaluate the feasibility and utility of that.
So, Gene and I came up with some I would say very basic questions that need to be elaborated because I would suggest the next steps would be to — we also came up with a series of contacts from different agencies that are collecting data, which would be linkable or are already linked, collecting and linking these data and would serve as models. I guess we should probably present the models first, the best models first and then sort of go from there and it would probably be good if we could get all of the people who are speakers at both days of the hearings, presuming there is a couple of hearings, to come to everything so that at the end we can have a round table where people can talk about what we have learned and where we might go from here.
Hopefully, a lot of barriers to moving this forward will come out in the discussion that we can talk about how we might overcome at the round table discussion. So, the questions that we have come up with that I think would need to be elaborated and probably put in to a letter and sent out to people as an invitation of we would like them to come. We would like them to address these issues for the data set and/or linkage that they are currently doing, are just these four.
I think either — we should decide how to proceed with this, but we need to come up with, as I said, a sort of more fleshed out version of these. What data do they have available? How easy is it for them to merge their data with other data sets? To what extent have they already done that? Then are there opportunities that may be coming up in the future that we should know about because a lot is going on. There is a lot of movement in the Federal Government. Wouldn’t you say, Marjorie, that there is really quite a lot going on.
But as I say, it is just not systematic and if we could help to move that forward, I think that would be the purpose of these hearings.
DR. STEINWACHS: Jim, I was just going to ask if you had comments because I remember yesterday you were talking about the Department is trying to do more. It seemed to me if you had some advice for us on these
MR. J. SCANLON: Well, actually there is a lot being done, a lot more — this is just kind of I think Nancy meant it to be a sampling and it sort of depends on who you talk to in each agency about what they know. So, I think you probably need part of the hearing just to be a more macro level view because there have been policy studies. There are strategies and there are plans besides specific projects and there are basic capabilities sort of built into most of our population-based surveys to — not all of them, but to be able to link to both claims data and then to be able to link to — ASPE, for example, sponsored a project that Gene may be familiar with, with NCHS, AHRQ, Social Security and Medicare where we basically, we link all of this information to a group to an HIS sample, but as Nancy indicated, this is actually quite expensive and quite hard to do. You have to kind of know what analytically you have in mind before you go down that path.
There has been another set of evaluations looking at — this was done by GAO and it wasn’t a particularly sophisticated approach I think but at any rate they looked at statistical matching, individual one to one matching and they looked at various — they looked at — they looked at statistical matching, individual one to one matching and they looked at various — and they looked at the concept of contextual data matching to our surveys and that would be — we can probably get our hands on that report because it would be helpful just conceptually about how — it wasn’t particularly nuanced or sophisticated. I think they didn’t quite understand all that is involved.
But there is a lot going on, a lot of proposals from the agencies. This is usually a resource issue. Agencies have to do their basic — the priority for most statistics agencies and research agencies is to keep the basic operations going and the linking is often as it is allowed or as resources permit, but it has always been, I think, as Nancy alluded, it has always been a part of the survey design, research design. Administrative to administrative is a little more complicated. There are confidentiality provisions and so on, but certainly that has been done as well. It is used for modeling. ASPE supports a big, you know, simulation model. Gene is familiar with it.
The TRIM(?) that takes census data and economic data and so on and then finally there are — I am trying to think at the National Academy — the National Academy of Sciences, you may want to include this, just completed a panel study on issues in privacy and confidentiality, including linkage and linking data, making data available for research. You might want to have them as part of the discussion as well.
This whole area is more than just a set of projects. It is actually a conscious strategy on the part of most agencies, resources permitting, but, again, it is important to know what analytically people have in mind. I mean, one of our basic linkages is the Health Interview Survey is — the whole MEPS(?) is now a part of the — it is a subset of the Health Interview Survey. So, any linkage to the Health Interview Survey helps us there. Any claims data of administrative data that MEPS gets, for example, is — again be related.
So, we built this capability and I don’t really think we have taken much advantage of it analytically yet. So, you have to — and the other thing is these sorts of things create new analytical problems. You have to know what you want to do and can it support it. But it is definitely an area of great interest.
In fact, I think we now have — this was a law passed about three years ago, but we now actually have at OMB — it was part of the confidentiality statute, but to some extent it makes it easier for the statistical agencies to share information for linking purposes, not entirely easy and it certainly helped the agencies that didn’t have confidentiality protections.
But, again, you have to — many of our surveys already have the capability for linkage and many of them have linked in the past. You have to sort of think of it as you are designing the survey. You can’t decide afterwards, which would be a good idea if we asked for a social security number, we could link now. You have to sort of build that in to the informed consent at the beginning.
CMS, I have to say, has been very helpful, Medicare and Social Security, in making the data available under the right protections or for research and statistics and policy uses. We actually work a lot with them.
But administrative data you all know. It is kind of the stepchild in statistics. The administrative data is used — it is a byproduct of the actual program. So, depending on what the research budget is or the statistical budget is, you have to really — at HHS we try to nurture that and protect that and in ASPE certainly we try to keep Medicaid data and we try to support making the Medicare data available. NIH does as well. But it is kind of in many ways a stepchild because these folks are running programs. They are not statistics agencies. So that keeping that in mind and helping to support those is always an important priority.
But I think it is a good idea. I think there is a lot of interesting work underway. I would step back on the hearing and have like NCHS and AHRQ and CMS talk a little bit more about sort of the history and — because there are ebbs and flows of resources in what can be done. But very interesting projects have been accomplished through — really projects that couldn’t have been done any other way. In the disability area, for example, the Health Interview Survey had a disability supplement. We have linked that. It has taken quite awhile. ASPE — we have linked that with social security information, as well as with Medicaid claims data. Even the nursing home, I think, Marjorie, if I am not mistaken, is trying to link some of the patient level data with the nursing minimum data set. So, there is a whole — it is more than a set of projects. It is actually a concept. It is another method, another tool.
I think you want to get that at the very beginning and then you can look at specific examples. But there is a lot going on.
DR. STEINWACHS: Jim, I think one of the things at least Gene and I talked about briefly was trying to identify the right person in some of these agencies.
MR. J. SCANLON: We can help you with —
DR. STEINWACHS: — the larger view and then people who are at the level that understand the mechanics.
DR. STEUERLE: Well, Nancy and I, we have just done some preliminary work, just to try to lay out — in fact, what we are hoping actually is that we would actually get some staff people, we probably would coordinate with people like you, who could really say, hey, you really need to contact these other people.
You know, this is one of these areas where it is very difficult. I think all of us who get involved in research or analysis on some level realize first you are dealing with an area where you don’t know what you don’t know. That is part of your research is to try to find out what you don’t know. Since you don’t know what you don’t know, you are really not quite sure how to get at it.
The other thing that happens — and I think Jim alluded to this very well is that a lot of these agencies a lot of people are doing things but under very tight resource constraints, are under very tight administrative requirements. So, for instance, there are linkages being done because something is required in the law to do X, Y or Z and the person sort of knows on the side, you know, this data set can probably be used for other things, but just doesn’t have the time or the resources. Then they run into the constraint often at the top of the agency quite honestly, statistical research is not — I mean, in ASPE it is, but in general statistical research is often not all that prized by administrators, especially if the results are going to be useful five years from now. It is like what do I get out of it.
So, part of the trick, I think, in not only contacting people but holding the hearings is we not only want to draw them out on what they are doing, but we really want them to be able to give us forth some real opportunities for what can be done. All of us have anecdotes. All of us having little pieces — Nancy and I went back and forth about little things she knew about or I knew about.
The one that I know a fair amount about is social security. As I say, I have actually done this study on the distribution of social security benefits and how it plays out. It turns out mortality offsets the — if you want a very quick summary — mortality effects offset the progressive aspects of the benefit formula and people get roughly similar rates of return on social security. So, the next question, well, what happens in Medicare. What happens if you had Medicare and asked what happens to the progressive data distribution of benefits?
At one point, I mean, literally the person I was talking to said, well, my boss really isn’t very interested in that. Then about three minutes later, he said, well, you know, actually we could do that. So, it is like sometimes we even — I am giving an example of this one because it turns out that Mark McClellan actually did a study before he became head of CMS on this very issue, was what is the distributional impact of Medicare benefits and came to some conclusion that it wasn’t very progressive, but in all honesty if you look at his study, he made about 47 assumptions because he didn’t have the actual data. He had to do a lot of — local area imputations or something like that.
So, they are interested in these areas and there are things that can be done. When I say we want to find out what can’t be done, I think we have to — the hearing very carefully because in some cases if you are asking people what could be done, they also face their internal constraints on what they can say. They don’t want to suggest — they don’t want us to suggest that we are going to do — they are going to do all sorts of health research when they know the Treasury Department is interested in capital gains research. So, we are going to have to figure out how to play with that.
Then some of the constraints are not really internal to the agency. There are almost these catch-22 type things, you know, like — it is the intersection of not knowing whether there are resources, not knowing how to deal with confidentiality, not knowing how to bring in outsiders where there is confidential — you know, for instance, some data sets are developed and then they literally just sit on the shelf because the agency doesn’t have any internal staff to deal with it. Outsiders, who are researchers, often would like to come in. Irma might want to speak to this — often might want to come in and use it, but in fact the system is not set up easily to let them in.
I can remember dealing with — this isn’t really a health issue but the Statistics Income Division, we have had this debate with them for 20 years and, you know, if you just figure out ways to get more people in. They said, well, if you bring people in then we have got to teach them on what is going on. In fact, our data sets really aren’t all that well documented. Then we would have to document our data sets a little better. We don’t have time to do that. Then all sorts of weird things, like, you know — I don’t know whether this is still true, but under IPAs, you couldn’t get — for awhile you couldn’t get former professors. You could only get current professors or you couldn’t get a research assistant and even if they worked with a professor, but you could get the professor.
There was all sorts of these bureaucratic hurdles and constraints and we sort of wanted to know the extent to which they operate to — so, I guess the next step for this committee is really to try to figure out — again, I am starting from the notion I don’t know what I don’t know, which is a lot, always has been. How can we then proceed to really forward in this area? What are the next best steps?
I think, Jim has made a very good suggestion about let’s find out what is going on, but I want to ask the question of these people in a way that doesn’t just stop there. It is like what possibilities do you see. As I say, Jim is correct. A lot is going on. A lot of people have thought about things, but only up to a certain level and then they just stop.
MR. J. SCANLON: It would actually be very helpful to, if your focus was on — you are right, Gene, you want to make it non-threatening. Otherwise, people don’t want to talk, but if you focused on the — in a sense the best practices, sort of the — and say that is what you are looking at, so you are not — in what cases are opportunities provided and we can find you folks from the agencies who are at the level where they can — you know, they can give you sort of the overall policy of the agency at NCHS or AHRQ or CMS, Social Security, even. Then maybe you want to look at examples of — there are different kinds of linkages and one is the contextual one where you are really getting small area data to correlate with the household data.
MR. LOCALIO: Jim, this is Russell. Please speak into the microphone. I want to hear what you say, but I am having the worst time trying to — it is breaking up.
MR. J. SCANLON: If you want to start with that overview of what do we mean by linkage or statistical linkage, there is a one to one. There is contextual. So, we have examples of those various things. If you are really interested only in the — literally the one to one, you know, household and a researcher survey study with claims data or with social security earnings, the history or tax data has been hard for us, but I guess there is confidentiality issues, but that is kind of another branch. I think that is often what is thought of as the classical linkage. But we have actually — other folks have done work in statistical matching, I guess it is called. I guess Irma would probably know, which is a little — it is used in other circumstances where we have actually done some work in ASPE with the Health Interview Survey and others, where you don’t actually have the same — it is not exact data, one to one on the same household or individual, but you match on statistical variables and, you know, so you are imputing to some extent.
You may just want to want to see the different examples and then you are right, best practices and what do they see as —
DR. STEUERLE: But I think — this gets to the discussion we had earlier in the larger committee is we want to enter this in a way where this subcommittee or the fuller committee has something useful to contribute. I think we don’t just want to have people come and report on what they are doing — we want to say, okay, is there a role we can play and it seems to me — I have this view about consulting and a lot of things — my view in general of consulting is the role of a consultant is often to go in an agency, find out the information that is already there and raise it to the right level. They weren’t necessarily worried about that we have invented a wheel nobody else has invented.
We think probably somebody down there has invented the wheel, but somehow or another, the wheel isn’t being used. So, we want to be — I think we just want to make a lot of effort to make sure that we are trying to find this frontier and trying to figure out how to be able to move it out in an efficient way.
DR. STEINWACHS: Irma and then Kevin.
DR. ELO: I really think this is a terrific idea to pursue it at the level of the national committee. Let me give you a couple of examples why and a couple of issues that I wish you would really seriously consider in the hearing.
One is issues of confidentiality. Once you link data, survey data or any other kind of data to administrative records, it provides a much richer data set for analytic purposes, but it also restricts access to those data. One of the issues that we are dealing with now in the National Center for Health Statistics is simply having the National Health Interview Survey linked to the national death records and that data cannot be released to the researcher without someone having to come to their data center to use it — the usefulness of that linkage, not that NCHS doesn’t have terrific researchers but they have many responsibilities that prevents them from devoting a lot of time to doing research.
So, we are trying to struggle with NCHS now how we can facilitate greater access to those data, how to make their research data centers more accessible. We are remote linkage and as we get into more and more Census Bureau or potentially IRS issues, this is going to be a really big concern.
So, I think the committee could address simultaneously the issue of access and how that could be then facilitated.
The other issue that we have — I was on a committee a few years ago that tried to understand how we could get better life tables and health statistics for — ethnic minority groups in the U.S. For example, at the moment, we cannot construct adequate life tables for Hispanics, Asian Americans, Native Americans. Pretty much, it is black and white.
We had discussions with NCHS of recommending linking the 2000 Census to the national death index so that we would get an ability to see how multiple race reporting in the census maps into national death index reporting on race or death certificate reporting on race. We ran into the issue that Jim alluded to, which is money. There was an interest in doing this — since 1960 — mortality by SES, not — this has never happened because they have no money to do it.
So, the national committee at this level could easily help push those types of efforts that would be very important and this is just a simple example of a linkage. Here the problem came with the states because the states — the national death index and they — and it was millions of dollars to do this linkage. So, it never happened. However, there might have been ways in which to — to at least do partial sampling. So, those are the kinds of things that I wish you would also consider what role the national committee can play both in the issues of confidentiality and resources for these types of linkages.
Particularly this relates back to your race/ethnicity report. It is becoming increasingly difficult to paint a picture of the health status of race/ethnic minorities with the increasing — population in the U.S. I think over the long run that is going to be one of the big issues.
So, I hope you would take some of those things —
DR. STEINWACHS: Thank you, Irma.
Kevin, then Bob.
MR. LOCALIO: Don, this is Russell. Can I sneak in at some time?
DR. STEINWACHS: I got you on the list, Russ.
MR. LOCALIO: Okay.
DR. VIGILANTE: The other thing is, you know, as Jim was saying earlier, you know, it sort of depends on what kind of analysis you want to do and there is this catch-22. On the one hand, you know, you want to link data sets and you may not have the questions formulated, but you have a sense that if you link these two data sets, it would be a great thing because you can imagine all sorts of questions being answered.
On the other hand, I think there is some benefit from thinking backwards from specific questions that may be frequently asked in the academic community about, gosh, if I could only link these two things, that would be great. So that, you know, not just having a government centric point of view from — government testimony from folks of, you know, what have we tried to do? What do we think we could do? What — by really going out into the academic community and saying — particularly people working in the seams, you know, people working in interdisciplinary environments, say, between health care and social and economic data or education data. So, if people are really interested in that environment and really asking them, you know, sort of what are the questions here that people are really trying to answer and what is your wish list of things you might want to be able to link to answer those questions.
Obviously, there is a risk of getting somewhat parochial and self-serving answers for folks doing their own research, but hopefully we can — if we can identify the right folks doing the right kind of work at a high enough level, that might be useful input as well.
MR. HUNGATE: My question is not so much I guess linkage as the observation that the question of available funding came up quite often in what I am hearing. When you start to try to get to a subpopulation within a population, the data is more specific and gets weaker in the bigger data set. So, my question is really trying to think about what about the avenue of foundation funding of specific supplemental work to fill in the holes where there is interest in a specific subpopulation and how easy is it to make that happen is another kind of a linkage that I just — I am ignorant on the issue.
DR. VIGILANTE: I was thinking of the same thing. I think it is good to investigate the potential for public/private partnerships here would be a very — I mean, there might be other applications sort of — but I think it is a very interesting thing to pursue.
MR. LOCALIO: A couple of responses to — and questions from what people have said. First of all, Jim, the NAS report that you were referring to is the one that came out in 2005. The first author was Singer. Is that correct?
MR. J. SCANLON: It is a very recent — yes, very recent. We just — I think it was, but let me — it was supported by the NIA.
MR. LOCALIO: I think I have it. Okay.
The next set of questions that you brought up and then Irma’s response, I want to first confirm that at our meeting in November, I believe Marjorie may have announced that somebody has retained an outside consultant to look into the issues of the current regulatory and statutory rules on confidentiality. Is that still going forward?
MS. GREENBERG: That was Jim —
MR. LOCALIO: I am sorry. Jim, is that still going forward?
MR. J. SCANLON: Yes, we have John Fanning, who used to work at HHS, work with the committee. He is doing a little — a monograph for us, where he is pulling together all of the background and the best practices for research and statistical confidentiality.
MR. LOCALIO: But I think if we have hearings we should have not only that report at our disposal and maybe John Fanning come, but also a representative of — from at least NCHS and maybe elsewhere, who would describe to us the process of going from the statute, which are not altogether clear, to the particular regulations and what are the possibilities for — if there are going to be changes, what has to be changed? Does the regulation have to be changed? Does a — do you have to go through a rulemaking? Is there a policy that has to be changed or is it a statute that has to be changed?
Another point has to do with Irma’s very good point about what happens when you ask for record linkage, what does that do to the overall set? Of course, what it means is that if you take two data sets that are collected under two different statutory regulatory policies, the resulting database is going to be controlled by the more restrictive of the two.
That presents a problem in that with limited resources it is impossible for those who have — those who are inside the — well, how shall I say, the NCHS umbrella, it would be impossible for them to do all of the research that is needed to be done or that could be done. Even if they could do it, if they had the time, they are not — how shall I say as unrestricted as outsiders are in terms of the types of research that they can do and the conclusions that they can draw from those from whatever research they have.
So, there are separate issues and overall policies that one has to think about in terms of getting access. Finally, I just have to say the issue came up — I think, Bob, you mentioned foundation money. I am skeptical that foundations are going to be open to funding this type of unfocused work. In other words, record linkage is usually unfocused because it doesn’t — it opens up a possibility for research in a variety of areas and I am not sure the specific foundations that have specific agendas are going to be open to that. I think the funding and the mandate is going to have to come within the various agencies.
DR. STEINWACHS: Thank you, Russ.
DR. BREEN: Yes. I wanted to mention something about resources because I think it is really important that we be mindful of resources and I think that the committee — I as a government person can’t do this, but some of the funding constraints really need to be brought to the attention of policy makers and legislators by people outside the government, who are very knowledgeable of what is going on, in order to certainly maintain funding for what we have and make people, you know, bring it to their attention of what we are getting with the data that we are currently collecting, but also to discuss ways in which we could enhance the data collection.
I think, for example, a survey that I use a lot is the National Health Interview Survey. Researchers would love to see that data set expanded so that the data could be reported at a state by state level. Sub-state level would be even better, but at least state level to start with. At this point, we have got the BIRFUS(?), which is, you know, is it redundant, is it not? It is a different survey. It is a telephone survey. It was designed to get estimates for the states but not for the nation.
So, we have got, you know, these sort of two surveys kicking around doing different things. They are both useful. BIRFUS comes out staffed. So, the states love it. NHIS came out slower. It is coming out faster. As Jim said, NHIS is also the base. It is the sample frame for the MEPS. A couple of — well, they have decided to rethink how they field and fund the National Health Interview Survey but about a month and a half ago, it was under threat. They were going to take it out of the field.
So, I want to bring this example to your attention because these things are not — the funding for these, even the most basic stuff that you would think — I mean, that survey has been in the field since 1957 and it provides the basis for the MEP and it is the premiere health survey in the country and they were going to take it out of the field because there wasn’t adequate funding.
So, I think that, you know, we need to bring these kinds of — we need to dig this stuff out. Medicaid is another example. They have no research budget. As Jim said, you know, and as Gene said, the research budget is kind of the frill. You know, we are lucky to have Edna with us because, frankly, the Indian Health Service doesn’t have a huge research budget either, though they do keep one statistician on and she does what she can and she does a hell of a lot given the resources.
But, you know, we don’t have routine resources and to get these administrative databases into a position where they can be used analytically requires analyses. So, you have got to have people working on the databases and that is where, you know, the constraint hits because if they can’t go outside the agency and it can’t be done inside the agency because there is no funding for a research budget, well, you can see there is just not going to be any research and the data is not going to be improved for research purposes.
So, this is kind of the underpinning of the thing, the seamy underpinning that we need to be mindful of as we invite people in and explore with them what some of the barriers are and also I think we need to be thinking about as a group bringing to the attention of legislators what an excellent system of surveillance they already have and how they could build on it to make it even better.
I just think that that is very important to do and something that probably — as far as I know, we really haven’t gotten too involved in. I mean, we often send letters to the Department, but not so much to, you know, people in Congress, who could actually fund these things and we do have some supporters in Congress.
The last thing I want to say is data is so cheap, relative to everything else, but nanotechnology gets money. I mean, we are not on people’s radar screens and we need to figure out ways to say data is cool, data helps you. Here is what data does. You know, we need to just bring this into people’s consciousness so that they understand what they are getting from data and what they would be losing if they didn’t have it.
DR. COOPER: And how to really use their data, I mean, at the state level because the data is sitting out there but it is — I mean, we don’t do advocacy but how do you work with the congressional individuals to let them know in terms of how this can help you, in terms of maximizing health efforts within your jurisdiction.
DR. STEINWACHS: Jim raised his hand. I go to Jim, Marjorie, then Gene.
MR. J. SCANLON: I would just be careful. The committee is not an advocate, federal employees are not advocates or lobbyists. You have to be careful about what you want to do with the Hill. This is the President’s budget that you are talking about and congressional budget. So, you know, you have to make the best use of the data you have. There are always more data needs than you have resources for, just like everything else. But I wouldn’t want to see the committee necessarily going off on the track of lobbying. That would be bad for everyone.
DR. BREEN: No, but I just thought we should be mindful of the fact that, you know, a lot of people here have kind of insider information.
MR. J. SCANLON: Well, all the more reason — education is one thing and we do need some supporters, like we have for all the other programs, but I think if the committee sticks to its expertise and what it does well at, I think — there clearly is an area here where I think the committee could help in terms of, you know, what is the philosophy, what are the plans, what are the barriers, what are the best practices and what is the utility to be quite honest. Then confidentiality issues, as Russell raised, and then how do you actually make — once you have invested all the money to get these data sets linked and how do you — does it die at that point? Are only federal employees allowed to use it? It is the dissemination and the utility.
But there is a whole area to be looked at here without necessarily getting into the lobbying issue. It would definitely be helpful, I think to everyone, including HHS and other agencies. Census has issues here. Administrative agencies have that as well.
MS. GREENBERG: My esteemed colleague said one of the things that I wanted to caution also, that isn’t the role of this advisory committee to lobby. But there certainly are groups, there is a group through the Academy, which has always been very supportive of —
DR. STEINWACHS: It is the Coalition for Health Services Research.
MS. GREENBERG: Yes, but there is also a group through that of friends of NCHS that is with the Academy, where they visit congressional offices and others to help them understand more about what some of these needs and resources are. What I really wanted to say was that — and I think a lot of good points have been brought out and, of course, Nancy, you were definitely preaching to the choir. I couldn’t agree with you more about the needs there and really that it is — in a sense it is a rounding error, but it still is funding that often is not available to do some of these things.
But I think this is, as Irma said, not only a very appropriate topic for the full committee but for the full committee to partner with the Board of Scientific Counselors and NCHS is the premiere health statistics agency. We have a report, as Carol McCall reminded the committee yesterday, on the 21st Century Vision for Health Statistics, which really speaks to many of these issues —
MR. LOCALIO: Marjorie, can you please speak into the microphone. I can’t hear you. You are breaking up. Thank you.
MS. GREENBERG: I am sorry. I think this thing that the committee’s report of 2001, hard to believe it has been almost five years, but on the 21st Century Vision for Health Statistics is still, I think, very relevant and there is much there that still is — you know, needs to be addressed. So, we have talked about a joint meeting now for several years with the board. We have an opportunity for that and I don’t want to steal Irma’s thunder from her report later to the full committee, but we have that opportunity in September because we have — we are both scheduled to meet I think it is on September 14th, something like that, and it is possible that a hearing on this topic could be the basis for that joint meeting. It could be a joint hearing.
Alternatively, if you had the hearing before then, certainly members of the board would be invited to participate and then you could have a discussion on September 14th about recommendations or next steps. But this seems to me to be a very good topic, appropriate topic for partnering with the board.
DR. STEINWACHS: Marjorie, thank you.
DR. STEUERLE: Just a quick addition. I think it important to realize that we are not just about saying more data are better, which is
DR. STEINWACHS: Even though some of us feel that way.
DR. STEUERLE: But there is no one we will — from this community who won’t say, gee, if you would only fund us a little more, if our sample size would be a little bigger, if you would only do this, I think one reason we decided to focus on linkage was because we viewed that as a real opportunity set. We are relatively speaking, we are not just talking about adding to a budget, but that linkage offers tremendous efficiency. There is an indirect hint — I don’t know how far we would go there, that, in fact, in a lot of cases if you would link data sets you could do a lot more cheaply than going out and trying to do a whole other survey because if you think about it, when you have administrative sets, assuming they are decent enough to be used for research purposes, you have already paid the cost of collection.
So, all the upfront huge cost has already been paid. So, part of the reason we are looking at linkages, we are thinking that they are actually some very efficient types of things that could be done there. I am sure if we went around the table, there are a few of us who might say, gee, if we spend more money on this and we didn’t decide that Census had upped its sample size because we are worried about congressional redistricting, which we are willing to throw aside. It is over a hundred million dollars more at one point because that became a big issue. It is not just a matter of more money. It is also a matter of really finding opportunities.
Since our time is running short, I wonder if we really ought to turn to sort of what — whether the staff themselves, people like Jim, could suggest to us what might be sort of a process to take us to the next step. We have talked about possibly having hearings and stuff like that. I think this document itself needs to be fleshed out. Members of the committee who have suggested different ideas need to add to it, even though this is just a laundry list of ideas. I think we need to white board, flesh out the laundry list, but then I think we need to have it go back to some staff people who are really onto this stuff. Jim, you mentioned people in HHS, who are already doing something similar. It would be nice — maybe they might find this a burden, too, but it would be nice if some of those people could come on board and help us think about how we could really organize something that they would find useful, something that is not repetitive of what they are already doing internally but where we can really make a value added.
DR. STEINWACHS: Jim, do you have suggestions for sort of next steps in working with —
MR. J. SCANLON: Well, I think we really should plan a hearing, I mean, at least a session where we bring in a panel of — the first thing would be to flesh out the questions a bit more because — and, again, you probably want to focus more on the policy and planning level, the macro level, and then look at specific applications and then what plans and utility are — I think it would be a good idea to plan — for the subcommittee, I think the subcommittee, as an exploratory, just to get into this, a panel and we could help you set it up, but I think it would be a good — you could get some of our agencies here in HHS, probably Nancy and NIA could help us get some of the extramural researchers, who used the linked data sets.
Mark McClellan actually was a big user and then some of the other federal agencies — and then with the right set of questions that come in, I think that would be a good place to start and then ask them for sort of what they think, what are the barriers, what are the opportunities and what would be helpful.
I think they would include the resource issue as part of the limitation. Confidentiality, you know, if done well, this can be done. It is not so much a big barrier. It just has to be done in a way that protects the individual identities. Then, of course, it affects the — but that would be part of the hearing. To what extent — what are the confidentiality issues. Again, I would focus particularly on — and then how is the data available. If it is only going to be available to a few people in the agency, that is one approach. But if it is actually meant to be made available to others through a kind of protected research data center or in some other way or to extramural researchers, like NIH often does, they, particularly NIA, I think they often issue grant announcements where you could use some of these data sets that are publicly available or under limited conditions are available.
DR. STEUERLE: Can you give Don or us a point person who is both interested enough but — knowledgeable enough and also perhaps well known enough that if he or she would be calling the agencies, they wouldn’t feel —
MR. J. SCANLON: We will get you some names. But I think that would be a good idea.
DR. BREEN: I think it would be really helpful, Jim, if maybe we could meet with you. I don’t know if you are available after this for a couple of minutes, but to help us with sort of the umbrella macro perspective that I think you might have in mind because, you know, that this is orchestrated is news to me and I think we really need to emphasize that. Then also with the people, the staff personnel.
DR. STEINWACHS: Simon is here. Welcome, Simon. I notice you are traveling under the name of Stan Huff. So, we do expect certain kinds of — and you brought your Kleenex box. I hope you are feeling better this morning. We missed you last night at the dinner.
DR. COHN: Actually I am a little slow to jump in on this. The mind is willing this morning, but the flesh hasn’t been quite as resilient as usual.
I just wanted to add — and Jim, I think, said most of what I was going to get ready to say. Actually, Gene, I really liked the way you are framing this and think that this idea of more efficiently and productively using resources that already exist is I think really — it sounds — it has a nice chord and sound to it. So, I think that is a very useful thing.
I do think there is a fair amount of overlap and we need to sort of leverage synergies and I am thinking yesterday of the conversations that Jim Scanlon was having about sort of reuses of data, which is really — this is just another frame of reuses of data. I was also thinking of the quality discussions yesterday afternoon, which was once again a conversation about — I mean, there are pieces here that touch all of it. So, I think there is a really very general applicability on this one. I mean, I think we have tried — I am just reflecting back with Vicki Mays and others over the last several years — we have tried to dig into this one, but we keep getting bolloxed up with the lawyers and everything else. So, if we can find lawyers that provide clarity, which is —
DR. STEINWACHS: Notice we have got John here at this meeting. So, we are prepared.
DR. COHN: I guess I need to be careful with my comments.
Given the right resources and the right systems, I think this could be immensely valuable.
DR. STEINWACHS: Other guidance? Because it sounds like we have some next steps laid out and, Jim, with your help —
DR. STEUERLE: Could I just — literally give us
— you know, read the thing, add one paragraph about — you know, Irma, the examples you gave. Let’s put them on there. I mean, again, this is just white boarding at that point. So, everybody who has an example, let’s at least get them on the table.
DR. STEINWACHS: What we will do is — sending out some reminders to people to try and prompt them since you go home with good intentions, but always — don’t always carry through.
The other thing I wanted to ask just in terms of timing because staff are here and you are all here about — in shooting for a hearing or a panel discussion, whatever this is is roughly win and I know it takes time to set these things up. So, are we now looking at — this is late February. Is it May, June, the earliest — I am looking at Marjorie and Audrey here as my — I understand we have to get some things done first, but I am just trying to get a sense of — from the point of view of being able to arrange and set up things, roughly how long.
MS. GREENBERG: I had mentioned, you know, possibly September. But that might be too — you may not want to wait until then. I think the important thing to know is when — what your sort of time frame is as to what work needs to be done before this could be set up and then project to that. Then we need to poll people for dates because I would think we would need — I can’t imagine we could do this probably before late April or May. I don’t know.
DR. STEINWACHS: Part of the — you know, sometimes you both work the time frame forward and work it back and then — so, Cynthia had — we polled, but the polled ended up with a time that we weren’t prepared to really move ahead on. So, why don’t — you know, I will look at Gene and Nancy and others to sort of help us, but it would be good to pick sort of what we think might be a time when we could hold the first of these and I am assuming it is maybe more than one because I liked your idea of using possibly the September meeting, either as a follow-up to or as a second set of sessions on this.
MS. GRANTHON: I have on my calendar that June 21st is the beginning of the NCHS Data Users Conference.
MS. GREENBERG: July.
MS. GRANTHON: Then I have the wrong date.
I just wasn’t sure how big an audience — just think in mind also in terms of perhaps other activities that will also create an audience for this or not, but —
DR. BREEN: There is that and then there is the September 14th — what was the partnering with the Board that you mentioned?
DR. ELO: The Board of Scientific Counselors.
DR. BREEN: Oh, for NCHS.
DR. ELO: Actually what I was thinking when you mentioned that, that it might be more productive if the hearing was before that and then we could discuss the hearing at that meeting rather than have the hearing at the meeting and then we would have to have another meeting as a follow-up. If that were possible, then I think that might be more productive.
MS. GREENBERG: That would speak to having a meeting early summer or something and then that would be a
— discussion by the two groups.
DR. ELO: I think that meeting would be better if there was actually something concrete to discuss so that it would be — we would actually make some progress rather than to —
DR. STEUERLE: To be very blunt, I feel really crucial — and Jim left, I think, a really crucial step is whether Jim finds this person or two within the staff, who are really substantively knowledgeable on this subject with whom we can work. I mean, because I think there is the organization of the meeting, which is one whole level of things we have to do in context, but there is also just — I think just fleshing this out better, which is probably that is never completely done, but I think that is a really crucial step. So, if we get somebody quickly and they can work with us quickly, I think that makes the timetable a lot shorter than if we bounce around at the next meeting and not being very clear about it.
Basically, I am saying I think Nancy and I have limitations on how much we know and can do internally within HHS, among other things.
DR. STEINWACHS: But we have never noticed those limitations, Gene. So, I wouldn’t worry about that.
Why don’t we do this? We will be in contact, Audrey, Marjorie and I with Jim and see if we can pin down just tentatively — and I think your caution, Gene, is good. Let’s try and make sure we are in a readiness to go situation and then if we could, either in June or maybe early July have a hearing or whatever we are going to call this. I will call it a hearing for — with the idea then in September we might take advantage of the joint meeting, you know, if Irma and others agree and then in terms of looking at people’s availability, also to look at availability of the Board of Scientific Advisors in addition to this committee for whatever dates and why don’t we try and push toward that?
The reason I was trying to find a time frame in part was trying to find a deadline to try and push toward because it helps me anyway to sort of think this is where I want to be, recognizing that we have to change course because we aren’t ready yet, then we will.
DR. STEUERLE: I just feel like there is a lot of work at both ends, both the stuff we are talking with Jim and what we are asking Audrey or Marjorie, whoever it is, to do. Organizing these things are not simple.
MS. BURWELL: We will be working with Jim and the Data Council and other on subject matter people.
MS. GREENBERG: I really think we should poll for a date probably in either June or early July and then we will — I think with Audrey and working with Jim and I think we can get someone involved from NCHS. That is a reasonable target.
MR. LOCALIO: This is Russell. Just one word of caution. My calendar and probably others gets filled up way in advance. So, we should probably try to pick a date as best we can soon well down the pike that we can all then hold for this meeting.
MS. GREENBERG: That is exactly my — what I am saying. Let’s pick a date and then assume that we can get this organized. You are dealing also with federal agencies, primarily. So, it is not — we are not going to have to travel in a lot of people. So, I think it is feasible.
DR. STEINWACHS: Marjorie says it will fit in the budget.
DR. COHN: Just a final thought if we are talking about late June and July, you might actually — Miryam was talking about the NCHS Data Users Group. This could be a very exciting meeting to sort of leverage somehow that activity just because if you think about the interest of the people who might already be here around all of that might make it a very exciting session. I liked your comment about trying to leverage meetings where possible.
DR. BREEN: Were you thinking of having it like before or after, Miryam, or —
MS. GRANTHON: I wasn’t thinking of a time. I just remembered that it was coming up —
MS. GREENBERG: That is the 10th to the 12th, though, and we could poll for the 13th and 14th — well, the 10th is a Monday. There are pluses and minuses of that, like the NCHS staff are completely occupied with the Data Users Conference. So, whether they want to go into a meeting and take up the rest of the week, I don’t know. The week before is probably not good because — certainly for NCHS people because everyone is preparing, but I think what is important is to get some dates from the key members, who — I mean, from —
DR. STEINWACHS: Well, the subcommittee and the Board of Counselors.
MS. GREENBERG: Cynthia, if you can get available dates from Don and Gene —
MR. LOCALIO: Who is going to be requesting dates?
DR. STEINWACHS: Cynthia will be sending out a request shortly. What we will do is try and identify at least three alternative two day combinations and poll and see if we have, you know, something that looks like a clear preference and then just as you are suggesting, Russell, is we will select one and shoot toward that.
Any other advice on this? Gene, Nancy, thank you very much. Thank Jim, too. Jim escaped on us, but we will find him again.
Why don’t we turn to the second area, which the committee identified as the second priority of equal importance and that was — Bill and Kevin have been giving us an education on what surge really is. Some people initially were not certain they understood this concept, but I think now right on top of a surge and surge capacity. We are also in the developmental process here trying to think about the nature of the contribution and having hearings, what kinds of questions would be and who would be there. I thought Bill and Kevin might give us an update where they are. We could have some group discussion.
This does tie even a little more closely, Simon, to some of the quality workgroup issues because the surge capacity of the health care system to deal with natural disasters or manmade bioterrorism, other things is critical and I think the general sense is as that capacity is dwindling and the other issues I think Bill has raised before Kevin, were how do we measure it. I think it is measured differently in different agencies.
So, let me turn to Bill, Kevin?
DR. SCANLON: Don did recap some of where we have been for those few — joining us for the first time. This came out of Katrina and the idea of sort of are there information needs that we should be thinking about and deal with disasters regardless of kind of their source and what our focus was was first institutional capacity and then Kevin made the contribution that we should think about this in the surge concept. It is not just an issue of empty beds or staff that are idle. It is the question of how do you adapt when you are facing sort of a significant increase in demand.
Then kind of what are the consequences of that adaptation, I mean, because you may be serving people but you are serving them in a different way than you did before and that the quality can be affected in significant ways. Again, Don, you suggested that people are concerned about what has been happening and I think there is really a strong basis for that concern, that we probably don’t appreciate how efficient, in quotation marks, hospitals have become over the years in terms of reducing excess capacity and that when we — if and when we need them, that capacity is not going to be there.
Katrina was very unique in some respects. 9/11 was unique in different respects. There are other types of disasters, such as SARS or the bird flu that would be of a different type and of a different magnitude. So, therefore, there is the issue of preparing for a variety of types of disasters and the question of sort of how one can best build the capacity to do that.
Kevin and I have had discussions and in particular we don’t — we recognize our role is not as an oversight committee for the Federal Government in terms of how prepared is it and what it should be doing in terms of preparedness, but more from an information perspective. What do we know and what can we do to improve what we have in the way of information. What we have found is that there really is a fair amount of activity as well as a fair amount of interest in terms of different agencies, as well as the private sector sort of in this question.
In some respects maybe sort of creating a prime opportunity for the committee because there is an issue of coordination, particularly I have heard from the private sector side, which is that, yes, we are inundated with requests for information and different measures of capacity, this way, sort of that way and it would be very nice if we weren’t so inundated. So, that is in some respects presents an opportunity.
It is not all coming from the federal level, though there is certainly sort of at the federal level — I mean, even if we were to restrict ourselves to HHS, you have got the potential of sort of multiple actors involved. In addition at the federal level, you have got the VA and the DOD and DHS also being involved, but states and localities also are on this train and they sort of are both gathering information, sometimes because they have federal money to help them plan and help them gather information, but other times it is just because of their own role in public health that they are doing this.
I think we have encountered certainly enough interest and enough variety that we think that it would be useful to proceed with some type of informational meeting, whether it deserves the dignity of being called a hearing isn’t sort of clear —
DR. STEINWACHS: Simon will guide us through the proper wording.
DR. SCANLON: Certainly sort of a further explanation. I am sorry that Jim escaped because actually sort of one of the contacts we had suggested that Jim would be a prime sort of resource for us because during Katrina, he was incredibly involved with the Department’s efforts from an information perspective. We wanted to tap sort of his knowledge sort of here.
DR. STEINWACHS: Maybe one of the things we could do — I assume Jim got pulled away — that we might be able to set up a conference call with him.
DR. COHN: Our first session at 10 o’clock is Jim talking about the hurricanes and data.
DR. STEINWACHS: So, he had to step out to prepare for this.
DR. COHN: So, one could innocently ask about any learnings around —
DR. STEINWACHS: Simon, thank you again for making sure our agenda moves ahead. We appreciate leadership.
DR. SCANLON: Kevin, do you want to —
DR. VIGILANTE: No, I think you have summarized it well. I think one of the fundamental questions is how do we frame the question so that it is relevant to this committee and I think Bill pointed out the notion that we want to be sensitive to the fact that this is not an oversight role and we are not here to say, oh, you know, we have — you know, DHS should be doing that or CDC should be doing this or they should be measuring or counting this way. But I think — so, I think we want to — if we do have a hearing, it shouldn’t be — it should be framed in a very sort of non-threatening way, but as an informational gathering exercise and I think that — but there are clear deficits in the way we even think about measuring this capacity, large gaps in sort of the methodology to approach it and then differences among different agencies in that thinking.
But, clearly, an important thing to be measured because as we have seen, we really can’t manage our resources unless we know it is at our disposal. So, I think it is very complementary to the notion of biosurveillance in which we are investing heavily and which is an important part of digital data collection to understand what is happening in terms of a threat, but then the complementary to that is to know what kind of resources there are at our disposal to respond to that threat in something akin to real time so that you can make those responses.
So, I think that — but, of course, that can’t be done on an ad hoc basis. The informational infrastructure has to be built ahead of time to do that and I think this is part of the thinking here and, of course, very relevant to population health.
Now, a corollary to this is the notion — and this is maybe another thing to explore in collaboration with the quality committee, there is — in preparedness circles, there is the realization that when something happens of significance scale, it is not business as usual. It is going to be searching and the staff ratios, the standards of quality are not the same in that environment. How do you grapple with that?
AHRQ has been talking about something called the sufficiency of care standard in these environments. Frankly, it is a vague concept. I don’t really know that much about it myself, but it would be interesting to think about quality measures in the context of this kind of environment and what kind of information supports those.
So, I think as Gene said before in some ways we don’t know what we don’t know about what other people are doing and we need to, I think, dig a little further to understand — to really flesh out what our role might be here, but after doing some initial probing, it seems like there is enough to suggest that another level of probing would be appropriate. But we would like to hear other people’s thoughts.
DR. BREEN: This isn’t an area I know a lot about but I just wanted to remind us of something that was on the conference calls, which I thought was a useful additional approach, in addition to looking at federal agencies and what they have been doing in this area. That is that some states or possibly some counties, the more local entities or cities have particularly good systems in place to do this and I think New York was brought up as one of them and there were a couple other examples that I probably have written down in notes somewhere and I don’t remember where, but it struck me that that would be a useful way to go, too, because I think that oftentimes the fed can learn from the local and vice versa.
So, to have that comparison available is useful so that we are looking at, you know, wherever we can find helpful models.
DR. VIGILANTE: If I could just draw an analogy, to our discussion yesterday, when we were talking to NHII’s subcommittee about what are we going to look at next. One of the things was, you know, what are the priority elements of data that one would report on in medical information. I think this brings up the same point. You know, in a locality or in a region, to understand — because disasters have these ripple effects that go beyond regions. If something happens in Louisiana, it is very important to be able to talk in Texas and Mississippi and Arkansas in the same ways about resources and what are the data elements that are critical to be collecting in these places to know what your resources are and some harmonization of this at the state, local and national level would be very useful things to be thinking about.
DR. COHN: I agree with what you are saying. I was actually responding to your second comment where you started expanding it off into the quality sphere and I was just going to — I think I was reflecting and I know we are both emergency physicians. So, I think we —
DR. STEINWACHS: So, we are safe here?
DR. COHN: I don’t know about safety, but I think we can directly relate to these issues of — I don’t know, I can think of them almost in terms of controlled chaos or controlled pandemonium or whatever, when you really have a massive disaster.
DR. STEINWACHS: I was wondering why you are such a good chair of this committee.
DR. COHN: But I guess I would sort of — as I was reflecting, I think that actually the items that you brought to the table here, I think, are very valuable and some of them not at all — I am somehow not at all — not at all certain as I was listening that delving deeply into the quality indicators of controlled chaos maybe is quite as high a level of issue. That is really what I was responding to.
MS. GREENBERG: I am warming to this topic.
DR. STEINWACHS: It has taken Marjorie awhile. We have been using all sorts of therapy on her to try to warm her up to this topic.
MS. GREENBERG: This is very relevant and topical and in a positive way. I mean that. No, I mean, it is not just that it is the topic of the day. It has relevance for the biosurveillance use case you mentioned, for quality, for a lot of the high priority items right now in the department.
Have you been working with any staff person on this or has this pretty much just been the two of you?
DR. VIGILANTE: Bill has done all the work. The answer is that we don’t have someone specifically who has a fondness for this area. Audrey is always there by our side.
MS. GREENBERG: Oh, of course, because I think on the topic we discussed about linking, I think we have some pretty good idea of at least where we need to draw from agencies and Audrey, I think, my sense is you would be comfortable kind of pulling this together also, I mean, with your experience in statistical agencies and all that.
This is an area as you said that there isn’t — the committee has not been involved and I don’t know that the Data Council or really there has been much identification of people to work with the committee. So, I think that is the first challenge. I don’t know whether it — I know Steve Steindel was here before.
DR. COHN: He had to leave, but I was going to say that CDC would be the —
MS. GREENBERG: I think CDC and HRSA, it seems to me, would be two key agencies. And possible AHRQ and I think that would be our first thing to try to pull in
DR. BREEN: What about FEMA and Homeland Security?
MS. GREENBERG: I would start within the Department because we are talking about health at this point.
DR. SCANLON: One of the reasons that I had actually asked Jim to come was that there is some activity going on within ASPE and some of the ASPE staff are looking at some of these issues. We don’t want to duplicate that, but we would sort of draw upon it. So, I think there is certainly a potential to coordinate with.
MS. GREENBERG: That is something we could talk about with him at the next session.
DR. STEINWACHS: Try and get Jim’s guidance and HRSA certainly is devoting — who in HRSA —
DR. VIGILANTE: In particular, the National Hospital Bioterrorism Preparedness Program, which is a $450 million grant to the states.
MS. GREENBERG: I was thinking the next step might be a conference call of Bill and Kevin with either Jim or someone from his staff, somebody from HRSA, somebody from CDC that our colleagues can identify, to kind of explore — okay, you know, where could the committee be helpful because I agree that the committee is not oversight on whether the country is sufficiently prepared, but you are oversight on statistical aspects of that, the measurements, the data aspects of that. That is your role. So, I think it is very appropriate and it might well be welcome.
DR. STEINWACHS: It sounds like you are starting to warm up a little bit, Marjorie.
MS. BURWELL: There is an Office of Emergency Preparedness that — you are talking about surge. They are really bringing on a lot — in the Department. So, I am sure Jim would touch on that in his presentation, but they definitely should be involved.
MS. GREENBERG: But I was thinking that conference call just to do some more exploration, which is fine. I mean, that can be done in that kind of environment because you are really trying to gather information and then come back to say, okay, well, we think it would be good to have a hearing or we think — you know, something. Maybe we don’t need a hearing because at this point I think we are committed to the one hearing in the summer. So, we might not be able to have a hearing on this until the new fiscal year. But that would work.
DR. STEINWACHS: It is good. I am just laughing because I enjoy you. I always think of you as Mother Marjorie.
MS. GREENBERG: I have a sign in my office, Mother Greenberg.
I don’t think we can plan for a hearing on this in the summer, but I think maybe, you know, later in the calendar year, the new fiscal year, also the budget might support that.
MR. LOCALIO: I just want to jump in here for a second. If the information that we are looking for is not currently readily available to anyone, then there is a need for having some type of a conference or a hearing because it should be readily available to anyone. That is at least my opinion. It seems as though you are not talking about stuff that should be hard to get. It should be easy to get and maybe that should be one of the focuses of a hearing as to how can this type of information be made readily available to people.
DR. STEINWACHS: Another piece, I guess, NCHS does do the facility surveys that, you know, certainly are not a source of information that talks to right now, but does talk to I would think trends over time and changes in the capacity and use of those facilities.
DR. COHN: I was just — I mean, somewhat belated reaction to Marjorie’s comments and I guess I was going to observe that, obviously, it is my interest to move things along in a reasonable fashion if we can. Of course, we do have budgetary issues. An option is always to sort of add a day pre a full committee meeting. This particular issue might lend itself very well — I mean, this doesn’t feel to me that in the first hearing you would need a day and a half. I was actually thinking of even a June meeting potentially as a pre-meeting day. Just throwing that out.
DR. SCANLON: I was just going to add that we — it was circulated at one point, but there was the supplement to the 2003 Medical Care Survey about bioterrorism. I think that thinking about those kinds of vehicles as one connection and then there used to be the NCHS inventory that went beyond surveys, but that they really were looking at the universe. I think we have to ask ourselves a question here. At what level of detail do we really need this information? We need to get down to localities and to, you know, sort of the state at least, there is a question of whether we need a different kind of data collection effort that maybe this would be an NCHS function again.
DR. STEINWACHS: It sounds like we have got a next step here. Jim, while you were out, your name came up repeatedly and we were told that in your formal presentation to the committee you would answer all questions that would come up because you were out actually preparing for it. It was mentioned that within ASPE there are some individuals sort of working on this area of preparedness and that Marjorie had suggested, which sounded good, that we needed your advice about putting together maybe a conference call with Bill, Kevin and if there are other interested committee members, but with someone from your office that would be appropriate, maybe someone from HRSA and CDC and to have that as a discussion to try and sort of help focus on, you know, where the next step is and who to work with on this.
MR. J. SCANLON: As Bill said, there was a — early on in the process the questions were added to the emergency department survey, I think it was, at NCHS and on readiness and preparedness and then one of our agencies actually gives grants to the states for hospital preparedness and I think they did a web-based survey. It is hard to know what the denominator is, but — so, they had some data as well. But I think we do need some more systematic look at first of all what are the standards, what are the best practices for preparedness to begin. This is hospital preparedness.
Then what are the measures and what are the data sources and certainly at some point there it is useful to have sort of sample aggregate information but I will talk later, at other points it was really important to have literally facility specific, very basic information on operational status of the various services and the staff, which is kind of a different — somewhat of a different issue that is not a statistical issue as much as a data issue, but I think we are at a point in HHS where this would be looking at this if the committee would be willing to look at this more systematically, that would be very helpful.
I have one or two of my staff looking at very basic literature review and industry best practices about what exactly is hospital preparedness and for what and how do you measure it. So, they could help as well. But we are — fortunately, some work has been done through the surveys. We have some estimates, but it is a growing area. We need preparedness measures for states and others as well, which is a whole other matter, but starting with the hospital preparedness. The hospital as you know is usually the focal point for emergency preparedness and response. Usually, you don’t have a whole city destroyed. Usually you might have one locality, a couple of hospitals and capacity hurt and then it is a surge issue. It is a preparedness — with Katrina, it was a whole other matter of not having — virtually having the whole infrastructure wiped out and people gone.
But that would be very helpful and we would be happy to help.
DR. STEINWACHS: Jim, I want to reassure you that this committee is well staffed, I understand and we have two emergency physicians here, Simon and Kevin. So, there is surge capacity.
Other comments, suggestions? I think we have a sense of momentum and direction here and I feel like that is good. We are on the move.
Well, why don’t we let you have a five minute break before the main meeting starts and thank you everyone. Appreciate it very much.
[Whereupon, at 9:55 a.m., the subcommittee meeting was concluded.]