[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
SUBCOMMITTEE ON POPULATION HEALTH
November 19, 2009
National Center for Health Statistics
3311 Toledo Road
CASET Associates, Ltd.
Fairfax, Virginia 22030
P r o c e e d i n g s (3:40 p.m.)
Agenda Item: Introductions and Future Subcommittee Agenda
DR. STEINWACHS: Let’s do introductions.
DR. STEINWACHS: Bill, do you want to restate our mission in this hour and a half.
DR. SCANLON: Thank you. I am a little bit at a loss but at the same I can’t let that stop me. My sense is as if you were to think about the goal being something to say in June 2010 – what are elements of what we want to say. Again, saying this morning’s the letter is model, think of it as objective to some kind of means. I think we heard a lot about objectives in the last two hours, and there was the question of how do we organize those around themes that can linked to means. Because one of the things that Harry pointed out is you have got a disconnect what you are giving as background and what you are making as a recommendation, that often doesn’t help you sell the recommendation. I think we have to think about that.
We have to think very concretely about the means. Judy, I am glad you joined us because to me – the notion that reform could open up elements of HIPAA and administrative simplification, as well as it just seems that if we were to move quickly we have to think about what are the most immediate sort of opportunities. To me the claim is sort of an immediate opportunity. Claims are not going to go away regardless of how advanced health information is going to get. There are still going to be payers and they still are going to demand information. That is going to be a reality.
The other thing about – Tony, I guess in terms of the states and the health information exchanges, I think we are going to experience incredible variation in the progress and the ultimate position that those states are on. When I was at GAO, we would do these 50 state surveys to kind of check on different things. I can remember the most vivid ones were in terms of preparedness after 9/11. You just kind of got back, well, we have done all kinds of things. Then you get the responses, well, we are thinking about it, we are talking about it, we are going to plan to talk about it.
That is what worries me is that we could have very significant gaps for a very extended period. The part about – I confessing that I sort of care about some health care policy, that a part about making health care policy, you have got to be able to reassure kind of a very broad set of interested parties in congress that you are not hurting them. We have to have a good set of information to be able to make these changes.
I think the other thing about health care policy is one of the sort of significant factors in it is provider’s perspectives. We have to be able to reassure them. If I were a provider, I would not be reassured if you came to me with the data that we have now in terms of operating a program, and telling me it is all going to fine we are going to change the system. I am going to do this on the basis on this limited amount of data and I have designed this new payment method or whatever, and you will be okay. I would be very skeptical if I were a provider, knowing sort of what the programs have had. So I think if we are going to move in that direction.
To go back to the population statistic side of things, never ever, at least in mine and I don’t think in anybody’s mind, was the idea that we are envisioning any kind of curtailment. In fact, I have been the liaison for the BSC. Our principle focus of the BSC has been we have got to put more resources into collection of the population to health data. They have written two letters to the Secretary saying we need more money. Almost like a staple of the meetings is here is where we are in terms of budget, here is where we are potentially going to have to make sacrifices. And then there is this kind of universal, this is unacceptable kind of discussion. So that there is nothing there about any idea that we could curtail things.
There, and this is in the context of something I was actually involved in, the last meeting there was a report on long term care data that NCHS collects. An honest assessment of the long term care surveys of NCHS is they are covering only a fraction of long term care. That if you really need sort of information about all of long term care, you have got to think about a strategy to deal with that.
The strategy in part, because resources are not – is to think about how can you substitute at times administrative data for components while you go out and do more extensive original survey data collection. The administrative data are not going to fill in the gaps everywhere. There is no way existing surveys are doing the job. It some kind of combination strategy about that.
I guess my sense is how do we outline where we want to go in terms of objectives and how we think about methods? There are methods in terms of recommendations and then there are methods in terms of operating as a committee in terms of gathering new input that we don’t have. Don and I have had conversations, we do have a lot of input if we look historically at what the committee has heard and then again, we all are out there too, in this world gathering information of our own and what we bring to this too. That is why we were appointed to this committee – we can utilize that as well. I will stop.
MR. RODGERS: Can I ask a question?
DR. SCANLON: Sure.
MR. RODGERS: We have delineated some of the population health data needs. If we look at claims data, I can tell you – a lot of other people can tell you too – what of these data needs could be addressed with claims data versus where you would have to get clinical data versus where you would have to get data that may or may not exist currently in an electronic form. May exist in medical records, maybe in health assessments, et cetera, that long term you could develop strategies to get that data electronic and then to flow.
Is that what ultimately the deliverables have got to have in it? We know that claims data exists – are problem is aggregating it. Getting it in an aggregated form that could be used for population health analysis, but that could be done. But the question is it doesn’t answer a lot of these other areas like I would question our ability as claim data for functional status. There are some data elements, and with 5010, ICD-10, there may be ways that you can look at – get some reports. That seems down the road.
I could say right now we aggregate probably would not give you that but what it could give you is prevalence of certain conditions based on progress. Is that what we want to start to look at?
DR. SCANLON: Maybe sort of dealing with an example that is in my mind, let us take the National Health Interview Survey, we go out and survey these people. We ask them – I hope I am characterizing them correctly – we ask them have you ever been treated for a condition or has anybody ever told you you have a condition. That is the diagnostic information we have got from them. We have had sort of in the past, some linkages between – I think we have had with NHIS with Medicare claims, which we have gotten independent diagnostics sort of information there – so that right there is kind of like how can we augment the survey. Not that we would eliminate the survey, but we would augment the survey through the claim.
Then this is kind of the question for the future. Should the claim have more information than what it currently has, which is essentially diagnosis and procedure?
MS. GREENBERG: Some demographics.
DR. SCANLON: Some demographics, but I think in terms of richer information.
MR. RODGERS: But is the value of that for the claim or is the value of that for some other purpose? Because if it is for some other purpose I would suggest that we look at another way to collect the data before we change claim systems.
DR. SCANLON: I would argue from a health care program perspective that there is value to change the claim. When we had a hearing two years ago about quality measurement; administrative versus clinical data, two important things we heard. One, the unbelievable burden in terms of trying to supply quality measures because there were so many variance out there in terms of if somebody would come in and say, I deal with 17 payers, they all want my diabetes numbers in a slightly different forms. That is a big problem.
Second thing we heard, which to me was significant, was the idea that I change in a significant way the predictive power of risk adjustors if I had simple things like lab values – and I say simple advisably, but it is like I don’t have that in a claim today. There is a question of kind of like if I had lab values, if I had X, Y, and Z, how do I change my risk adjustors?
When we talk about right now in the bills, episode payment, pay for performance, all those ideas, pay for value, all those ideas are in the bills. If you look at the analyses of those we always come up against this barrier. We have explained this much and then we have got this very large share that is not explained.
Can we do a much better job if we have a claim that has much richer information and that will serve the health care program and then maybe we can turn around and match it with a survey and serve the population statistics?
DR. MIDDLETON: Yes.
DR. SCANLON: That is my conclusion too. The question is how do we get there, in my mind.
DR. GREEN: Two comments, Bill. I really like the way that you are thinking here. I put just a question mark by the claims – that the claim is going to be with us forever. Some sort of claim is going to be with us forever. In 2009, I bet I have been to 40 or 50 meetings of some sort, about redesigning the health care delivery system. In 90 percent of those you will hear someone stand up and say, until we get rid of fee for service payment there is no hope for redesigning the health care system.
They may be wrong.
DR. SCANLON: No, but in response to that I would sort of think about what is happening now with Medicare Advantage. We are creating a reporting requirement for Medicare Advantage plans that gives us basically the equivalent to a claim even though we are paying on a per capita bases.
DR. GREEN: That is my point – the claim has got to morph.
DR. SCANLON: But the data reporting is nuts.
MR. RODGERS: That is the problem because you have to have an encounter to know what the transaction occurred that you are aggregating to. So if you want to look at the number of diabetics, you have to have an encounter that is giving you that claim information so you can say, here is a premise that my diabetic is in Hispanic population or in this population, or whatever. That has been the only source of data for us to look at populations like that.
The question in adding data for the purpose of claims is a way the way that claims are processed. The timing of the process and where that other data is going to come from and what is the timing of that. All those things will make it complicated. That is not to say that we don’t have to solve that problem. I am just saying that when we solve the problem, know who is going to pay the price of solving the problem. Because that is going to tell you what your resistance is versus I personally think that if we could identify data aggregators at the state level that are going to be the sources where all data will eventually reside, whether they are HIE’s, Health Information Exchange organizations, with data warehouses or whether they are certified individual data aggregators, we have got to identify the who as much as the how. If we do not identify the who, then no one is going to do it. I don’t think claims themselves are the answer.
DR. WARREN: That is my point though. You have got to be careful about what you start adding to the claims because remember who the data input person is. It is going to be the nurse or the doctor. I already know a lot of people that I design systems for, they don’t want to enter one more thing unless we pay them for it. They want to be paid for data entry.
DR. GREEN: So much of the redesign care is asynchronous. So much of redesign care is asynchronous, which gets back to your very first point that it is not a face to face encounter. The data we need are not even there. They have been somewhere else and they go completely undetected.
I wanted to ask a question, what is our thinking about whoever fills it out, whatever this thing is, what if there were a data set or data standard, that goes with it?
DR. SCANLON: That is this morning’s letter.
DR. GREEN: There has to be.
DR. STEINWACHS: There is only one way you measure.
DR. SCANLON: Those of you who are clinicians tell me if this is wrong – I am the physician and I have electronic health record and I am seeing a patient and I am entering information about that patient in the electronic health record – that is the data input. This is an ordinary encounter. This is instead of writing it down I am putting it in directly.
Can we construct records that then I press a button and certain elements flow? That is what I think we are talking about here. So there is nothing incremental in terms effort to get the information that we want. We are not saying to them that okay, you have got to collect 15 other data elements because we want you to give us those, too.
DR. WARREN: But you could.
DR. SCANLON: You could but that is where we are going to get the resistance.
DR. MIDDLETON: The by-product question, it is the value calculation that has to be done at the individual level for the clinicians to say sure, I will gather you these 24 more things if I know I am going to get some value from it. We may have to say that there is population value. The problem would be sort of rational myopic analysis from the docs point of view is he or she may never see the value themselves. However the public health value or other process value, even in their own clinic, may overwhelm that calculation so it is the right thing to do for the whole clinic even though it doesn’t feel right for the doctor.
When you are ready I have another topic idea.
DR. SCANLON: On that point, I think one of the things you have got to do in terms of thinking about minimizing collection costs, and that goes all the way to the level of physician collecting incidence, we have to decide what is the amount of quantity of information I need for a population health from a public health perspective versus what is the information that I need from an individual perspective?
Because I could – it could be that sampling gets me everything that I need to know about the public health sort of perspective. The question is what is the most efficient way to do that because this idea that we always need univere data that is going to get rejected.
DR. MIDDLETON: When you say universe data, what does that mean?
DR. SCANLON: Information on everybody.
MR. RODGERS: CMS is going to aggregate eventually both the Medicaid data and the Medicare data. If we just had that could we extrapolate it to the general population? Okay, so what are we missing?
DR. HORNBROOK: You are missing the uninsured group –
MR. RODGERS: After health reform? Okay, so that is not a large enough sample size.
DR. SCANLON: But it has merit of its own.
MR. RODGERS: Okay, so we certainly could write some standard suggestions to CMS as they begin to aggregate this data, what is going to be useful for population health. It still comes down to data aggregators. You cannot do this on an individual EHR basis. You may pull the data elements out of there, but someone has got to aggregate it at some level, whether it is the claim or clinical encounter.
DR. HORNBROOK: The aggregation can be at the source, in the middle, or at the other end. Three different places you can aggregate it.
DR. SCANLON: Warn us when we are getting to far a field.
MS. GREENBERG: I want to put a process question on the table. Even though I said June was only the beginning, it is the beginning. The question that I have is do you think that the committee, I am not saying the subcommittee but the committee, with obviously significant contribution from this subcommittee and the rest of the committee because we have agreed that this is cross-cutting whatever we are talking about and I am not sure that we agree on what that is, but it certainly is cross-cutting whatever it is. Do you think that something could be put together by June that would be interesting, exciting, contributory, et cetera, and you sort of alluded to this, but based on all the work that the committee has done over the last 10 years, and the brain power around the table in the committee, plus staff, and maybe some consultants? Or will we need to hold – and then maybe based on that – one could hold some hearings to start the feedback at the symposium and then have some other hearings or are there some things that we need to either gather or test or whatever, through the hearing process, because if the latter is true, we have to do those hearings in January in order to contribute.
We cannot leave this building tomorrow without that answer. We can’t defer it to a teleconference. It is getting into the holidays between Thanksgiving and Christmas and Hanukah and Kwanzaa and everything else, just getting everyone together. I really need your assessment of that.
DR. HORNBROOK: Can I ask one very basic question here in response to Marjorie’s question? Is NCHS held accountable for a national ??? (52:49) such that there is a premium on a national sample and if so, do we believe that if we used the samples of convenience because the data could be extracted more comprehensively and cheaper, that we could use that through comparative effectiveness modeling techniques to predict the people who were not available. This issue is exactly the same as what would Medicare patient’s have cost if they enrolled in the other system – estimating the counter factual. We can’t randomly assign people in these two systems – how do we estimate across?
MS. GREENBERG: So what was your question – it started with NCHS?
DR. MIDDLETON: It is a suggestion.
MR. RODGERS: NCHS, you have the problem of estimating for the whole country. We could go ahead and do a national sample and tell you how best to do the national sample in populations that happen to have electronic and those that only talk to you because that is all they have got.
MR. GREENBERG: As Don says we have a lot of experience accounting for or trying to model around non-response and people who don’t have telephones and all of that.
MR. RODGERS: Are we willing to go to the next step of modeling from those who have comprehensive electronic data, inferring from them what the people who don’t have electronic data look like?
DR. MIDDLETON: There will be an error assessment based upon sampling only from electronic data which people need to know about. Whether or not that is kind of the big question on the table right now, I don’t think so. I think it is important – I think the question was very helpful to Marjorie because it was framing.
In a way, having been in my short-term tenure on the NCVHS, the last vision statement of this group was 2002. That is not that long ago. I’ve kind of lots of ideas about the big picture vision of where we might want to go with population statistics and all the rest of it because after all it is the 21st century and that is going to go all the way to 2099.
What is different right now than the last time any statement was made? If we took a very near term view of what is different right now, it is going to be focusing on how to build upon the 21st century vision with some insight and assessment of how to do the real fine grained assessment of value and its relationship to both individual and population type decision making. So far I don’t think we have really done that.
Value of course, will interface comparative effectiveness research and all the rest of it and there are a whole set of questions that arise there. It is not the long range vision perhaps of where are we going to go with respect to what is particularly of interest to me is where are their knowledge gap? You say information needs – to me it means the same thing as knowledge needs. How do we assess knowledge needs? How do we assess preferences of providers in patients? How do we assess community status in our fine grain way to reflect behavioral influences and constructs that have to be measured for individual health and wellness? And what are the meta-data – oh, by the way, that we want to think about across all population data types, et cetera, that allows us to use them together and combine them in interesting ways.
DR. SCANLON: I go back to Marjorie’s question for a moment. I don’t want to be too absolute about this but I am not sure that we need a hearing.
MS. GREENBERG: I am not going in that direction.
DR. SCANLON: We have already, Dan and Gib, we have surveyed – we won’t call them stakeholders – team informants about the division, which as you say, is not too old. We got their perspectives in terms of its relevance for today. We have got the collective experience of members of the committee, as well as the prior hearings that we have had. From then it sort of seems possible to put together – in some ways it is already in the three recommendations you gave us – some kind of summary of what are the key areas and then those can be elaborated sort of in a report. But then the next step is okay, what are we going to do about that? That to me is kind of is the key.
In the quality letter – I had to keep using the quality letter as an example but it really did seem like it is going to be déjà vu all over again, we had a hearing and we heard a lot of things. In some respects you could call this a lot of rich information. The question is what part of it did we use?
As opposed to how did we step back and say, alright this is a fairly good characterization of a problem. It is a fairly good characterization in terms of what we heard about all kinds of activities that are directed at that problem but not necessarily sort of solving it. What is it that we could say that might make a contribution toward making it possible to more effectively attack that problem? I think that is similar to what we are talking about here.
Which is what are the different components. We could have our letter about there needs to be a bigger investment in the population surveys. There is no reason that the BSC are the only people who get to say this. We have heard in the past about – even when fully funded how inadequate – we know that from our own experience how inadequate some of the surveys are. That could be sort of a part of this.
Then there is this question of how do you merry other data sources with surveys? We can talk about that. This is where I think Standards and Privacy can play a big role in this because I feel like were in some ways, the academic subcommittee, okay. We can give you sort of this ivory tower fantasy about what we want –
DR. MIDDLETON: Quality is the academic subcommittee.
DR. SCANLON: They are our counterparts. Then we have got people on the committee that actually deal with the real world. They deal with either the people that are going to produce the data or the people that are going to complain about the data that is being delivered. Those are two critical components of what we have to say. If they are not weighed in what we are asking for we are dead in the water from the get go.
That is what made this a full committee activity in a way, because Quality and Populations can get together and say, oh, this is what we want but –. We have got to get practical, we have got to be very realistic about what it is we are telling the Secretary that she should do in terms of meeting our goals.
DR. GREEN: I am personally very happy with the review that these two characters at the end of the table have done. On page two of four –
MS. GREENBERG: Which document are you talking about?
DR. GREEN: Just before the potential issues to be addressed – the seven things, letters underneath. The question to be considered within an extensive time period, there is clearly indications of how do we get there from here. What is wrong with that?
MS. GREENBERG: Where are you here?
DR. STEINWACHS: Page two of the Potential Issues and the Seven Data Needs.
MS. GREENBERG: This is what I am thinking we have got sort of all the background and everything, and as Bill said, we can articulate some of the key problems – a lot of them. We don’t really have to go out and get them anymore. Everybody has their own view. So what is needed is to kind of really capture that in a compelling way across the whole committee and then some type of a beginning of a roadmap it seems. Some people don’t like the term roadmap.
You are not going to hold a hearing and say what should the roadmap be? That is not going to be successful at all. But if you could begin to sort of lay out the components of the roadmap and the priorities and present something – I think that can be done. I don’t think you need hearings for that. Then I think you might want to push it out to some different audiences or hearings afterwards.
If that is kind of where you think you could be in June then I frankly don’t see the need for hearings either. We might want to have some, just as they did their key informants, some selective consultations. Even though I think the most brilliant minds are on this committee, there may be some other people out there that we have not had a chance to put on the committee yet that would be really useful to talk to. We have that. We don’t have to talk about a hearing to do that. We can do that through new questions. We can do that through consultants or we can do that through – somehow engaging with them. But we need a plan, obviously.
DR. STEINWACHS: Bill promised me because I have to leave, that I would get all the work. What I was hoping was that Dan and Gib were going to get all the work and I would help them.
Being a person who does see some value in concrete examples because if we could pick something – when you talk about major health threats, well obesity comes to the top of that list maybe. You could pick any one thing and use it as an example. Set out a set of principles or in a sense, these seven areas are identifying, use a current pressing example, obesity or whatever we wanted to pick, and then talk through it – at least from our point of view, what is missing and how could electronic health records be a resource that is valuable in augmenting what we know. I still think that we need information sources at the local community level because the action on obesity is not a national action in great part.
MS. GREENBERG: I loved what Judy said.
DR. STEINWACHS: Then let’s go with Judy.
MS. GREENBERG: What are the questions we really want to know about health lets say for the baby boomers. Are most of the decision makers, they are a little younger than baby boomers now, right? They are moving along. That is so compelling because all those things you mentioned many of them don’t even get collected in surveys, let alone administrative data or in potential electronic health records. And yet, that is where we all live on the quality of our life.
DR. GREEN: I think Don’s point though is a good one.
MS. GREENBERG: You are leaving us?
DR. STEINWACHS: I am afraid I have to. I got double booked in my life.
DR. GREEN: The seven questions can be answered for an academic community in a fairly technical sort of way. But they could also be asked around questions like Don and Judy are asking to illustrate this sort of stuff so that it begins to seem real to normal people, sort of thing. I don’t hear that as a contest is my real point.
MS. GREENBERG: No. One includes a lot of the things just in academic terms like you were mentioning.
DR. GREEN: I do want to argue Marjorie, these are good questions and they resonate with what we have been working on and the nature of the discussion that we just had this afternoon for me, needing some sort of framework in some way to go forward and get into action mode. I don’t understand why we don’t just use this. This is really pretty darn good.
If someone says we are missing this, that we really need something different from that – I don’t understand why we don’t use this.
DR. MIDDLETON: And the target might be instead of a kind of new vision, document, whatever, but what do you do, as an addendum, or an update or something maybe not as quite a profound a level but still extremely useful and timely, given where we are? I don’t know if you can make the target specification before we depart here today or tomorrow?
Then by way of process, there still could be a hearing, or a couple of days worth, to flesh out this stuff that I thought was on the table.
DR. W. SCANLON: There is the big question of whether we have a meeting of ourselves to flesh it out.
MS. GREENBERG: That is what I am thinking. Maybe supplemented with some of these people who have not gotten on the committee yet.
DR. W. SCANLON: The February meeting, we need to think about a half day ourselves.
DR. MIDDLETON: Is it a budget driven decision or a science driven decision?
MS. GREENBERG: No, it is not budget.
MR. RODGERS: Thinking from a policy perspective, when I look at the issues of population, I look at first my age cohorts because each of these questions are different within the age cohort that I am talking about. So have we got agreement on the age groupings, the sex groupings, and then framing our questions that we need to validate with either hearing or whatever, that we could say – quite frankly, functional status and children, lets say the ages five through 16, has an implication in terms of the cost of education, the cost in terms of level of disability, et cetera, long term costs for that person.
Functional status and a baby boomer has a different I guess you could say, timeline of cost. The reason that I bring this up is that at the end of the day there is two things; return on investment for the resources I am going to mobilize to address the population health issue that I see in my population. So when I see a large percent of adolescents who have adolescent diabetes, I ask the question – what is the intervention I can take to reduce my long term cost associated with that population.
First is the question, what was the data? What is the framework of what adolescence is? And then what is my expectation of my ability to address both the preventive interventions, but the ongoing care interventions that will be most effective?
The reason why I am thinking this way is that from the standpoint of view of what population health information should be used for, to help policymakers to help those who have a liability for the cost of care, to help those organize resources at the lower community level. You have to have the age cohort identified in which the data will be used so that there is standardization there. Then you have certainly the sex, and then the questions within that as relates to environmental factors, functional status, behavioral risk, because early drinking in adolescent populations is a great determinant of future health risk.
If I knew that I had a high risk of that and I see that in my population, what intervention should I take now and what is the cost effectiveness of those interventions for reducing my cost later. The only reason that is important to Medicaid today is that we see a future where we will have certain populations that will never leave Medicaid under health reform. That there will always be Medicaid. That has never happened before. It is always they move in and they move out. There is going to be a population that will just stay with Medicaid all the way through from their childhood to their adult to their aging out, if you will.
So if that is true, it effects how we look at population health. Can we agree at least on the cohorts in which the questions will be asked?
DR. WARREN: Let me frame that just a little differently. I think down in the leads of data, all I need is age or birth date. Then I can create cohorts as I want because I have already got the data. To me your argument is what do I do with the research after I have got it. Where what I think that we are discussion is what data do I need and what should I have and can I aggregate it once I get it?
MR. RODGERS: I am trying to categorize the questions within cohorts versus just general questions as a functionality.
DR. WARREN: If I have got the data then I can ask all kinds of questions.
DR. W. SCANLON: Right. My sense is, particularly for June, that we don’t go as far as identifying cohorts. We maybe identify the need to be able to identify cohorts. In some ways what we are dealing with for June is a framework that is relatively generic. We need to know – have the ability to define cohort or a sub-population that is relevant because of X.
Don’s example of obesity – I would see that as maybe a text box or a sidebar where we are talking about for example, I want to know about this.
DR. WARREN: I need a birth date, I need gender, I need weight, I need height. If I go really good, I need a waist and hip measurement.
MS. GREENBERG: What about race and ethnicity?
DR. WARREN: Well, you need both.
MS. GREENBERG: Socioeconomics status.
DR. W. SCANLON: Right, and part of what we have been talking about is that we need all of these things and we are not necessarily going to get them all from the same source – we need a strategy that gets us the necessary information through some means and different amounts of information for different purposes.
MS. GREENBERG: You just said the most important words. We need a strategy.
DR. W. SCANLON: That is what this is about.
MS. GREENBERG: I was just at a meeting in Brisbane, Australia. I will misquote it but somebody said, do never underestimate the power of one person or one organization that essentially has a strategic vision that has a strategy to get somewhere.
DR. GREEN: Last meeting, maybe it was the meeting before that, but at some point and it may have been in Charlottesville, we were recalling how the National Committee on Vital and Health Statistics envisioned the National Health and Information Network before people even said NHIN.
Just arguing by analogy, it is as if we need to do that again for the new health statistics. But it is to some degree not a function of holding a hearing to collect more data, as much as it is a failure of the imagination about what is possible. Another magical thing I think you said a while ago, you said what is different now? One of the things that is different now is that we actually all believed that there will eventually be a National Health Information Network. It may take forever and who knows how functional it will be.
Things are different now then they were when –
DR. MIDDLETON: Two years ago.
DR. GREEN: Again, trying to take the charge up seriously here about how do we get from here to June with something. I agree with the drift of the conversation, I don’t know how we would construct a hearing. I have been listening to if we have a hearing what are we going to do – and I am not getting very far there. Getting this committee back together and possibly with some help, I could imagine us addressing –
DR. MIDDLETON: That is big vision, Larry. I wonder if that is the tactical 2010 goal or is that the 10 year, decade anniversary goal for the next statement of 21st century vision. Do you know what I am saying? Do we want to do the big vision thing or –
DR. GREEN: I just would not polarize it. There is no such thing as a little vision. Things that are little are not really visions. They tend to be tactics and that is the next step to do that. In envisioning what is possible – when you look at the background stuff and you think about the shoulders we are standing on, something is trying to happen here.
DR. MIDDLETON: I like what you are saying. I am just trying to get the right frame.
MR. RODGERS: I kind of look at the questions of the what, the how, the when, and what is the value of what we are going to do. The what – is this the what? Has then been defined?
DR. W. SCANLON: I think we need to expand them some. I think there is a great start here, okay, but we need to think about how to expand. We need to think about how to organize in some ways around the questions; the what, the how, the why, and the when stuff. I think that is key.
I don’t want to be too – this is sort of an issue of strategies or tactics – sort of too grandiose in what we say here because in terms of what is different now, there is a real opportunity here. We are not just trying to condition people to think about a broad and different future. The future is right now and they have got this money. What we want them to do is spend it in a way that achieves a vision that we have already seen. We are going to talk about changing that vision at the edges.
If they don’t spend that money well it is going to be a lot longer before we ever achieve that vision. I don’t know how prevalent this is in the real world, but here in D.C. we talk about elevator conversations, how do you sell somebody in an elevator ride? Thirty seconds – that is what we have got to have. We have got to have something that is sort of compelling about a decision that they are going to make. Keep it concrete but make sure that our vision is not violated in this process.
DR. HORNBROOK: Statistics Canada gave the nation the ability to measure quality adjusted life years as a way to evaluate their health care system. They have done it over several years now, it is a utility weighted health system, and they can compare across provinces, they can compare across age and gender groups. They have the vision of saying we need an output of a healthcare system for our country. We don’t have that.
DR. WARREN: One of the things because I was thinking back to Tony, what you were saying kept bothering me because I tend to get too specific, that maybe if we start thinking about all the things that Tony is talking about in the way that we talked about this morning with the quality letter. What we are really looking at are groups here. So each one of these A, B, C, and D’s, are really kind of groups of data. And not get so worried about what it takes to measure them. Although at some point we are going to have to.
What NQF has done that I think if phenomenal, is they are harmonizing all their data elements so that you can collect that data once and then put it into any quality measure that you need to have. That is the idea, that is the dream, it is very hard.
Maybe that is what needs to happen here at this time, and then we get down to some of the questions that you are talking about.
MR. RODGERS: Does it matter to us whether the data broadly is in digital formats now? For example, we know we collect functional data on all of our long term patients, that is part of the eligibility process – it is digital, therefore it can be shared. Now it may not be harmonized but it can be shared.
If data is not digital it is going to take time to put it into digital format. Is that enough for us to be able to say, and this data is available or not in the future EHR because it is not broadly adopted yet. What are the data that will adopt to digital that is not digital now, that tends to be in records that you have to pull or whatever. Would that be useful as part of the report?
DR. MIDDLETON: I am reminded of a NCQA diagram from literally t 10 or 15 years ago, that describes kind of the sources of health care information and that which is in a computer system versus that which is not. As you know better than many, much of what we might want that might be interesting isn’t digital. This is actually a good point, key point. What do we say we want to point toward to digitize? Which data? Because it is not going to be one record sourcing all this data. And really the comment this morning — I will underscore it again — the system to system idea.
MS. GREENBERG: It is a subtext, though. I think eyes glaze over when they hear statistics. Just think how they glaze over when they hear digitalization.
MR. RODGERS: Tone other change is going to happen. Once in it is avaialbel, you have to add it again. So the cost, today’s cost is that we do have to fill out that form over and over again because it is not digital, we don’t share it. In the future collecting data at eligibility, for example, something I preach in my state, the health assessment collected at the same time we collect our eligibility information, digitizing that, can we give that to the health plans and say this person has this history of health care problems. If you take that same data, not digitized yet, that could be part of where we want to drive the initial contact. Once you know, the family history doesn’t change –w ell, it might change a little.
DR. SCANLON: Well, the demographics may not change. There have been recommendations to commit demographic data. SO we have every provider in the country, for everybody they see, submitting demographic data. My concern about that is then what do you do with all of the inconsistencies? We to report four years ago now, I think it is, unbelievably complicated, when you really get down to really defining it well, and we found providers should do this? And they should submit it? And what are we going to do when we have the billion Medicare claims and all kinds of inconsistencies?
Collect it once, this is one of our principles, collect it once correctly and share it. That is another thing about our current system is that we have been so tolerant of sloppy behavior. That is a major source of conflicts. You try to deal with administrative records and you find out so much is just inconsistent and nobody goes back and says we have to improve on how we do this.
DR. MIDDLETON: Sounds like a key message, digitization, data integrity.
MS. GREENBERG: Can I ask — prerogative of the chair — I would love to hear from Gib and Dan and also from Garland, since they I think, of the people in the room, who are not staff, live the most in the population health data space. Is that okay?
DR. SCANLON: Yes — I thought you were asking them if they would talk.
DR. MIDDLETON: In other words Marjorie is saying everyone else is blabbing.
MS. GREENBERG: No, no. My fiduciary responsibilities since I brought these guys in, I want to hear from them. Also Garland has spent his whole career in the population health space, I would say, generally. So I just would like, if you have anything to say — you don’t have to.
DR. FRIEDMAN: Sure. Somebody mentioned obesity earlier, perhaps taking an issue specific lens to some of the questions. For me that is an interesting example because certainly in terms of what I think of when I think of population health, this was the cheapest and most graphic way of addressing the problem. What would I do to address it and what kinds of data would be helpful?
When I think of what kinds of data, what would I do to address it, frankly I would deal particularly at the individual encounter level. What I would do would be to adopt the kind of strategy that thirty years ago people said was completely unrealistic around cigarettes, which is suing the hell to of the companies, and suing the hell out of McDonalds, and collecting data, driving the price up until — thirty years ago you could buy a pack of cigarettes a day —
MS. GREENBERG: A hundred dollars for a pack of French fires.
DR. FRIEDMAN: It is not as stupid. Thirty years ago everybody would have said this is idiotic, you are never going to do it. Tobacco companies are too strong. Well, the prevalence of smoking has plummeted over the last thirty years. It has been a huge public health success. A lot of that was done through policy changes and through legal changes, and through making it incredibly uncomfortable for tobacco companies.
So I guess what I would say, if you want to take a few issue examples, one of the things that I think would be helpful would be also saying, in addition to treatment data at the point of care, ask what other kinds of data that are not individual record data at all, regardless of what kind of individual record data, would be helpful in trying to understand the problem and address it from a population health point of view. That’s my two cents.
DR. PARRISH: I like the vision idea here and I think that you could continue to have a longer term vision here with some short-term tactical goals. I think you shouldn’t rule one out for the other. But I think it will be important 0– I guess I would try my best to frame the longer term idea first, the strategy. Then within that look at it carefully and then what are a few things that we could tick off tactically that could be suggested that are short-term approaches, whether it might be something with the claims, or with some other aspect of this, that fit into that larger vision, because we have still until 2099.
MS. GREENBERG: And out grandchildren can figure it out.
DR. PARRISH: And I agree with Ann. I spent a while a few years ago actually trying to draw out a picture of all of the interrelationships of smoking in terms of the factors that have brought us to where we are, admit is really a complex set of interventions from many different places — tax policy, second-hand smoking policy, advertisement policy, suits of the companies themselves as well as counseling by individual physicians of people not to smoke. It is a complex picture and obesity would be a similar situation. It is a complex set of social, economic and other factors that have been brought to bear to bring about that increase in obesity in society. It is beyond the individual level.
DR. MIDDLETON: So actually it is interesting, crossing my mind, is the notion of sort of looking at the smoking story and analyzing in that whole multi-factorial canvass of stuff what were the data that were relevant post hoc? But then turn around and say, well, we have this idea from the smoking experience. Going forward what would we do in obesity?
DR. PARRISH: And actually the Office on Smoking and Health put together a group that gradually compiled over time data, state laws, related to smoking and smoking policy, state taxation policy related to cigarette sale, and they actually started to piece all that together and that allowed people to then look at what is in fact the sensitivity of cigarette consumption to tax level, and things like that.
MS. GREENBERG: Then there were surveys and clinicians started asking people — you can see the different pieces. It is interesting.
DR. PARRISH: But there may be some smaller pieces that can be done in the short term that can add to that larger picture.
DR. MIDDLETON: The only thing that is missing, though, was there — I should know but I don’t remember — was there a value-based purchasing part of this smoking story at all? Or coverage decisions based on — in other words, the value question. How did that apply in smoking?
DR. PARRISH: Value in terms of insurance?
DR. MIDDLETON: The interest now in global payment or bundled or thirty day readmit — whatever. It is for this amount of money you are going to get to do this much and no more. If you do any more, it is up to you, in a simplistic point of view. So how do we use, or was there an analogous kind of thing in smoking? Maybe it was implicit — your premium.
DR. SCANLON: The only thing I would think about was premium. Premiums were explicitly tied —
DR. PARRISH: Not only health care premiums but life insurance premiums.
DR. HORNBROOK: Perhaps your company liability policy if you have smokers in the workplace.
DR. MIDDLETON: The other thing you didn’t say was kind of the community standard thing, which maybe was the result of public health messaging on a whole variety of levels. AS soon as, you know, bars, and everybody said no smoking — you are not sure which came first.
MS. GREENBERG: Now we want to star discriminating against fat people.
DR. MIDDLETON: It is already happening.
DR. GREEN: The obesity thing really does take us a lot of places, doesn’t it? It is a pretty good idea, I think. What happens there in that exchange is, to get theoretical just for a second, is we are saying that a biomedical model has been powerful in some ways but hasn’t gotten us to where we want to go. It has to be augmented by a socio-ecological framework. Our data systems are not sufficient to move to the socio-ecological framework we are going to have to be if we are going to solve problems like obesity.
If that is the case you have to know distance to fruits and vegetables. If that’s the case you have to know if there is a playground. If that is the case you have to know if there are gunshots still being heard at midnight. Otherwise kids will not gout and play. It is that sort of stuff that we are talking about needing to get into if you are going to fix the obesity epidemic in early childhood.
DR. MIDDLETON: And even nearest neighbors in your cohort. Everybody in your friends group and everything. You said something just then, Larry, in a way another framing kind of idea is moving sort of population statistics from the biomedical model to whatever you want to call it, actually, this bio-psychosocial new world order where we know much more about the behavioral and contextual influences on health.
DR. GREEN: To get something to replace it, the opportunities for augmentation and completion and balance.
DR. HORNBROOK: You can find out probably pretty carefully from clothing shipments, size distribution, coming into what zip codes or what census blocks.
DR. GREEN: I can’t buy my size coat most places except Africa, because of the clothing shipment problem. I understand that.
DR. HORNBROOK: Then the location of fast food service areas because they know what their service area is. Then the location of the grocery stores and of course the people are using their shopping cards and you have got how much fruit and vegetables they buy. This is all without asking the person.
DR. SCANLON: Garland’s turn.
MR. LAND: You mentioned that we only collect demographic data once. Just sharing something, the basic demographic data on a person is all on the birth certificate.
MS. GREENBERG: That should be the bookends, the birth certificate and the death certificate.
MR. LAND: We can never use the birth certificate as the fundamental block for demographics because it is isolated into its own. If everybody had to present their birth certificate for the first visit you would have their date of birth, you would have where they were born, you would have their race, you would have all the information —
MS. GREENBERG: You would have their race or their mother’s race?
MR. LAND: You would have their mother’s race, which is really about all you can work with any more because of the multiracial things. So the basic demographics are there, but it has never been considered as an opportunity. And we have the software already right now where you can actually look up somebody’s birth certificate. We’re doing it from a security purpose right now, but it could be used for a medical purpose.
The way public health views them, and this is an enigma, we identify what are the most pressing health problems and from a data perspective then we develop registers to insure that we understand that people who have that health care problem — we don’t develop registers on everybody. We have registers very specifically targeted to where the greatest needs are in terms of morbidity and mortality.
I wonder if that concept needs to be considered. We are talking about all of this data that has come from encounters. There is obviously millions and millions of encounters here coming out of this. Most of which probably aren’t all that important to even be tracking and be worrying about. But we ought to be looking at, from an age group point of view, and it is going to be different for each age group for morbidity and mortality, where do we need to be focusing our resources to track people and understand what is working and what isn’t working?
I really appreciated the discussion this morning saying that we can improve our health care system and not improve health. Man, I think that message is not well understood by Congress. That is why the whole health care debate is going on out there, because everybody is saying, the general populous is saying we have the best health care system in the nation. Then we come along and say, yes, we have the lowest infant mortality, we have the lowest life expectancy. They don’t understand those concepts. So it is only a group like this that is going to continue holding up that we have to be collecting data on the health, besides the health care. That needs to be coming out very strong in whatever we propose, because we are the only ones — I mean, public health is trying to say that. CDC is trying to say that. So we are not the only messenger, but it is a message that has to be pushed forward out there and since we are looking at it from a data perspective we ought to be looking at it with how that data is tacked into the health care data and has equal importance.
MR. RODGERS: One of the new audiences for this data, if health reform passes and if there is a mandate and if you have the opportunity for products that are guaranteed issue, is what has changed is that commercial health plans can price you out of the market. So they don’t have to worry about your health status. They will just up the prices until you leave or whatever. Or they can deny even covering you if you have an existing condition. If that goes away they are going to be extremely interested in population health, because they are going to see that as a huge liability coming their way. People are going to pick the best networks, the best — I call it the brand equity problem that some health plans have. So if you have diabetes you are not going to a plan that doesn’t have a good network.
So we have a new audience and I think we should say that. I think if that is going to be true we should say that future payers are going to be extremely interesting broad population health because it is going to affect their long-term financial stability.
The mechanisms we use on the payer side, in managing, are disease management, which is focused primarily on — well, first it is case management, which is focused on an individual who has got a particular case that we are trying to make sure is managed appropriately. Then there is disease management, which is a little more population-based, et cetera, using certain interventions, et cetera. Then you have care management, which is the activity of the plans, evidence-based, clinical practice, the network design they set up. Then finally you get into public health, broader population health.
Within that we have never really focused on behavioral risk factors. My point being that population health has never been of interest to the commercial people, other than academically or as it relates to risk adjustments, et cetera. But it is going to become extremely important to them to be able to determine their liability in managing a population and what interventions are going to be most effective.
There is a new funding source, potentially, if we can articulate it in this document, the importance of this population health data to these entities that are now going to be responsible for providing health care coverage in this new environment — assuming health care reform goes through as it is currently designed.
MS. GREENBERG: We’ll have two versions of the report.
MR. RODGERS: It just seems like that has been the problem with population health. There has been really no effective funder user.
DR. MIDDLETON: A business case.
DR. SCANLON: Right. I guess in some of what I have been hearing there are two issues in my mind. One is the cost of acquiring information, and two is what you can use it for. They both need to be taken into account. One of the issues is that we are cheapening the cost of certain information. That is what the future involves. Plus we are trying to expand the number of uses, and I think that is sort of a key. Some of these uses get us down to a very micro level, which means we are going to require a lot of information. So we need to be thinking about that.
DR. MIDDLETON: Small cells, you mean?
DR. SCANLON: Small cells, right. You deal with physicians and you talk about pay for performance, and the immediate thing you hear is that a program as big as Medicare, the observations are they don’t have enough elderly diabetics, they don’t have enough elderly COPD people. So you can’t measure me on that. That is the kind of dilemma we are facing here, which is you worry about that. So that is a couple of things, some of these principles that need to be in our work.
The other thing, the element I would put more in here is the health care component. And even if we can talk about health care being this small determinant of health, that is the direct impact. I would argue that we are moving more and more toward the indirect effects of health care influencing sort of people’s health, because right now the biggest budget item in states is Medicaid, and it is not education. I mean, we are crowding out the other determinants of health. We are going to be crowding out all these other things that we value unless we get a grip on the spending, okay? So we have got to think about the essential information from that perspective as well.
I am wondering as a strategy for the future of this endeavor whether we should take Tony’s questions — the what, the how, the why, the when — and think about where do these things fit into them, and what we would add to them. Then we try to put those together. These are questions, but there are also potentially principles that are associated with these. We have talked about we really want data that is digitized, which to me — the word retrievable and manipulable are — what good digitized means. I want digitized narrative that I can’t work with.
We want data integrity. There are various principles that we would weave into our discussion. We think about identifying those at the various sort of elements. Put that together. Have a call to go over that as a sort of first step in this process.
DR. HORNBROOK: Does this vision also perhaps think about preventable diseases?
DR. SCANLON: Absolutely.
DR. HORNBROOK: Because preventable is a lifetime perspective.
DR. MIDDLETON: That is the only way you can get at the cost factor you are worried about.
DR. HORNBROOK: If you look at osteoporosis you are going to start at age 18 or 15.
DR. SCANLON: There are two issues. One is that there is necessary care that is required because people are in poor health given a lifetime of bad behavior. Then there is the issue of costs that are due to sort of services that we would regard as unnecessary. There is a real believe that there are a component of them out there and the question is how do we ever get to them?
SPEAKER: What was the second one?
DR. SCANLON: Unnecessary services.
DR. HORNBROOK: Overuse.
DR. MIDDLETON: But let’s not just say it is all bad behavior.
DR. SCANLON: No, it is not all behavior, but there are multiple parts — there is necessary care that we are not going be able to ever eliminate, and we wouldn’t even think about trying to attack that and reduce that. In fact, we have evidence that we have underserved in many instances. So we want to think about how do we sort of eliminate that? And these data drive that, too. If we have the information about an under-service, then we can think about the intervention. It is when it is hidden is when we are in trouble.
DR. HORNBROOK: The other thing that is hidden is that we won’t pay for some interventions that will save a year of life at $50,0000 but we will pay for interventions that will save a year of life at $3 million. We cannot socially ever get that affirmed.
MR. RODGERS: You can’t politically say that. I hear what you’re saying. Unfortunately that is exactly what everybody is concerned about. But you are right, if we could ever get the maturity. I personally think the problem is low health literacy in our population does not allow us to have this discussion because they are victims of the systems and they come at the system like my doctor has to do whatever he has to do and anybody who gets in the way — not being part of that. One of the things we have done with our Medicaid populations is test the level of health literacy based on a standard test, and it is shocking what people think and what they know about how to manage their health care. That is part of population health, the fact that we have so many people at such a low level of literacy makes it very difficult for them to engage and become responsible. Because what they have learned they have learned form people who don’t know anything.
DR. MIDDLETON: I am sure you mean doctors and patients.
MR. RODGERS: Some of them, too.
DR. SCANLON: One of the things I have encountered on many occasions is some providers are a part of this problem, which is they have instructed their patients and created some myths. I have someone I know who got a bill from a hospital, questioned it, and the response from the hospital was Medicare makes us do this. So we sent them back the intermediary guideline which says Medicare is not going to prescribe how you bill this. It is up to you and your systems. That is the kind of thing. You will have a home health agency saying Medicare says I can’t do that for you. Well, no they don’t. So it is those kinds of things that contribute to this health literacy or illiteracy.
DR. GREEN: I was just thinking — it won’t happen, of course, this is all politicized — but it would be so wonderful if like this whole thing that has gone on this week about mammography could be a teachable moment, as opposed to the advocates lining up on both sides. It is such a microcosm of so many things and it relates to so much of what we want to maybe be able to do in the future.
DR. MIDDLETON: But we are not going to shy away from political issues, right?
MS. GREENBERG: Of course not.
DR. MIDDLETON: I am dead serious.
MS. GREENBERG: Except for the unique patient identifier.
DR. MIDDLETON: I am not sure I even agree with that, frankly. Isn’t NCVHS supposed to be the unbiased, nonpartisan advisor —
DR. GREEN: And we are supposed to be ahead of the policy makers.
DR. MIDDLETON: We need to speak leadership to truth and say what needs to be said.
DR. GREEN: Either way in this instance.
MS. GREENBERG: But you don’t want what you say to fall on deal ears. You can say anything you want but you would like it to be heard.
DR. MIDDLETON: I will sleep better at night if we say the right thing as well as make recommendations about what can be done. Not just the latter.
DR. SCANLON: But it is how you say the right thing.
DR. MIDDLETON: I am with you.
DR. GREEN: Back to strategy and not leaving the building until we know –the tow conversations now, this one — I want to ask about privacy and how it would relate to our work in moving to June. Don’t we have to move in concert with our privacy and confidentiality communities and how is that going to happen?
MS. GREENBERG: That really relates to the health literacy thing, too.
DR. SCANLON: I think it is a big part of the how, both privacy and standards. This question of if we are talking about here is what we want and here is how we are proposing that we get it, that that is the role of those two subcommittees in terms of reality checking. There is a means to do this and that is potentially feasible, that it will be politically tolerable. I think we have to worry about the fact that —
DR. MIDDLETON: What are they going to do, fire us if we say something politically inappropriate?
MS. GREENBERG: But you want to be effective.
DR. SCANLON: They are going to ignore us. That is the problem. If there is A and B and A is the best but B is close, but B is sort of feasible —
DR. MIDDLETON: I jus think there is a new opportunity in the current climate.
DR. SCANLON: The question would be maybe we need to have one call to deal with these questions that are formatted in a different way. And then maybe we could get Judy and/or Jeff to sort of join us in a call and Leslie and John at the same time. We did try some coordination before this session. It is difficult to do coordination across the subcommittees. The Executive Subcommittee had a discussion of this in its call a month ago.
MS. GREENBERG: The way the Standards and Privacy really fit into this is first of all, if you don’t really push the standards agenda this data isn’t going to be useful, certainly for population health. It won’t even be that useful for any kind of aggregation of data. So it is critical.
Then on the other hand, if you don’t have the policies that make people feel comfortable sharing their data or trust whatever the process, and that things that need to be kept confidential will be, you won’t ever get the data anyway. They are so related and of course quality is just one piece of it.
DR. GREEN: So can we as a committee assume that the Executive Committee and you will figure out how to sequence things between this committee, Privacy and Standards and we don’t have to care about that any more?
DR. SCANLON: If there was someone else to assign it to I would say we should do that. That is what it is going to do. It is going to be up to Don and I to try and take a lead.
DR. WARREN: From my perspective this committee needs to kind of generate the discussion. So I like repackaging the questions in the format that Tony talked about. Then what standards can do is say these standards exist that can do these things for you. Here are the gaps. So some of our recommendations may be to go back to those SDOs and ask them to fill in those gaps so that we create the machinery to transfer all of this stuff. We could respond much better than we could try to coordinate. It would have to come from you.
DR. SCANLON: That is what I am thinking and we need to talk about it in terms of what we are trying to accomplish. Then you can give us the, well, you better go back to the drawing board because here are the limitations that you need to be cognizant of.
DR. GREEN: Can we do that tomorrow morning?
DR. SCANLON: IN Quality?
DR> GREEN: No, I mean can there be an interaction of the committee tomorrow? In the group there is an indication of the direction of the direction we are thinking about.
DR. SCANLON: We are certainly going to report to the group. I guess the question is the agenda tomorrow.
MS. GREENBERG: Everyone will report back and we are talking about the sixty —
DR. WARREN: Was your question is there time to work on it tomorrow morning?
MS. GREENBERG: It depends on who is sitting on Quality ro Standards.
DR. SCANLON: I am going to Quality because their letter is being revised.
MS. GREENBERG: It seems like we have sort of an agreement to build on this document.
DR. SCANLON: I was thinking how do we take questions one to seven? With Don’s help we’ll take a crack at this — take one to seven and put them in the framework of Tony’s questions.
MS. GREENBERG: Which were —
DR. SCANLON: What, how, when, why, and then maybe try to add some, to the extent of what we think at this moment, maybe principles.
MS. GREENBERG: So you are thinking that you and Don could do that?
DR. SCANLON: It is not rocket science.
MS. GREENBERG: Do you need any additional help to do it? Do you want Tony to work with you?
DR. SCANLON: I would be happy to —
MR. RODGERS: Tomorrow morning?
DR. SCANLON: Tomorrow morning I am going to go to Quality.
MS. GREENBERG: A teleconference?
DR. SCANLON: Yes.
MR. RODGERS: Do we want to measure, as part of the parameters — who do we want to measure?
MS. GREENBERG: Whom?
MR. RODGERS: Well, I —
MS. GREENBERG: I didn’t mean to correct you. I was trying to make sure I heard you.
DR. GREEN: Can these two guys continue to be pressed into service?
MS. GREENBERG: We can if they are willing to be pressed into service. I will be really honest here. I didn’t really know what direction this was all going to go, but we asked them to come. Our contract with them sends them home today.
DR. FRIEDMAN: Tomorrow, actually
MS. GREENBERG: And that is the end of that phase. Tomorrow, whatever. But yes, we can certainly continue our engagement with them, with Gib and Dan. The question is if we can make clear what the engagement is and then if that is something that they feel comfortable with, and that that is how they want to spend their time on it. If we are keeping a population health focus, obviously bringing in these other areas, my sense is they are more interested than if we were just really going in a very different direction. Obviously we have quality, we’ve got the standards, we’ve got the privacy and those other areas aren’t their area as much. I don’t want to put you guys on the spot, but are you going to stay with us or not? I’m kidding.
SPEAKER: Are you going to pay them?
MS. GREENBERG: We will pay them. We have the money to pay them, maybe not enough to make it worth their whiles, but we do have money to pay them. The contract ended with their coming to this meeting because we didn’t really know what the next steps the committee wanted to pursue were.
DR. SCANLON: We need to quickly put this together. Then you will have a better sense of where we are going to do and what the role might be. You also raised this whole issue of the three other subcommittees. We need to kind of have their buy-in, too.
MS. GREENBERG: But I will say that it would be very hard to get anything together, even if we don’t have hearings or whatever, in time — and you can’t get it together the night before the meeting. You have to have it ready to go. I know people on the subway are preparing their presentations, but we are not doing that.
So it is going to be very difficult for the committee to pull off something that is even ten percent of what I think we all see as the ideal without some outside help.
DR. GREEN: I concur. I also want to take it a step further. I really think the privacy discussion is central to everything we are talking about. And there may be the equivalent of these two gentlemen that we may need to engage in privacy. We may already have it there, I don’t know.
MS. GREENBERG: I think the policy stuff has to come from the subcommittee itself.
DR. SCANLON: A combination of this subcommittee and the Quality Subcommittee.
DR. HORNBROOK: Do we assume we have to have an individual model health? Can we go to a more aggregate —
DR. GREEN: We make that explicit right now. I thought that was pretty clear. If we need to make it explicit we are absolutely going forward realizing that health is won and lost in the context of one’s family and community and your social environmental circumstances. It is silly to plan for 21st century health statistics any other way, right?
MS. GREENBERG: So may instead of talking about patient-centered health we should be talking about community-centered health?
DR. GREEN: Yes.
DR. HORNBROOK: And look at all the environmental statistics you don’t get from the individual. You get them from geographic units. There is a whole set of environmental factors — built environment, air pollution, water pollution, density, poverty, income and education, tobacco, consumption of alcohol, crime — all those things can be geographic — geospatial data.
So are they in?
MS. GREENBERG: If they want to be in, whatever in is.
DR. MIDDLETON: Or you could charge a little subcommittee to decide where exactly the scope of work.
MS. GREENBERG: We can’t just say to them go forth and make all these disparate thoughts we are having into something that makes sense. But as I see it this is going to require a lot, within reason, of involvement of the committee members, certainly the people in this room. There is just no way to avoid that. But I don’t like writing by committee and I don’t think that is very successful. I think we need some dedicated resources for putting it all together. I would love to just quit my job and do it myself — not that I would do as good a job probably — but it is so exciting. But it is going to require a commitment of all of you, maybe more than you have had over the last six months. Some of you have had a lot of commitment, I know.
DR. MIDDLETON; We’ll have a retreat in Aspen for a writing party?
MS. GREENBERG: We could. I was thinking myself that we may have to do some first drafts or whatever and then have a little retreat — good food, good wine and lock the doors.
MR. RODGERS: Within the scope of this I hope that we will look at the issue of the advanced medical home and its use of population health data. I do think we have to articulate the users a little differently in the future, because that is what you want your medical home to have, some of that population health data, and to be part of the strategy of addressing community health issues.
The second part is I really think we have to think about the issue of population health as it relates to schools and their effect. Then even clinical decision support tools. Because if we can articulate users I think we articulate value.
DR. WARREN: So maybe it should be the health one, elf-literacy.
MR. RODGERS: But we are also setting the new vision for how the data is used.
MS. GREENBERG: To whom is it useful?
DR. MIDDLETON: The new members of the committee actually wrote the first draft.
MS. GREENBERG: That is a policy we have. Actually we should assign the whole thing to Don, who isn’t here.
DR. SCANLON: When you said he wasn’t going to be here and he was going to get all of the assignments, he agreed.
MS. GREENBERG: I am sure he writes well.
DR. SCANLON: Yes, I think he does.
DR. HORNBROOK: One of the things that has happened in Kaiser is the recognition of they are more than just a cottage industry of physicians doing their own practices, and that they are accountable for their mistakes because the mistakes come back and hit you in the pocketbook, and that the cheapest patient is a patient who is controlled, and that a brand new patient is more expensive, and that the fact that you patients are going to age with you means you better treat them well early because they are going to get sicker and sicker and sicker and you are going to have to take care of them. The best thing to do to prevent ER visits is to go out there and get to them before they get to the ER, even if it means going out to the house physically.
MR. RODGERS: The other part of that that we have added is that if you think about the patient’s engagement with the physician, if they have the same level of health literacy a year from now, after being with that physician that they have today, that physician is not engaging the patient. So it is the new responsibility that we see the primary care and the medical home having in the future. I think this is where perhaps that new user group —
DR. WARREN: Somebody was talking about looking at what is happening in England with the whole phenomenon of the health district. So there is a health district for the community and they are notified that somebody in their community is admitted to the hospital, and they visit them in the hospital and then go get the home ready for them to come home to. They make sure all the resources and everything are there.
DR. SCANLON: Any other topic that anybody wants to raise?
MS. JACKSON: Yes, could I clarify? I am really excited about the direction of what I have heard earlier. The synthesis of so much that has come out from the quality report, the materials from Don and Bill, everything just seems to be pulling together in some amazing statements. But I am kind of excited and terrified at the same time.
DR. SCANLON: Me, too.
MS. JACKSON: There is a structure of expectations. We are looking smack dab in the middle of a dead zone from December 15th through January 5. So just so you know, that means a lot of touching base, connecting, almost weekly sessions. There is so much that has come up and percolated that has got to be grappled with seriously before we can try to engage anyone. Right now it is trying to prepare a scope of work of what are we doing, who is touching it, where are we going? There is just a lot out there.
So I would like to also find out if your idea is to have another session like this at the February meeting? You have only got one full committee meeting before June, and that is February.
MR. RODGERS: You will have to have a lot to present at that time.
MS. JACKSON: This session was so helpful in giving everyone altogether the points that Don and Bill pulled together to put feet to fire to these subcommittee and really bring these points out on a lot of these issues. It was amazing.
MS. GREENBERG: It may be the only time that one subcommittee told another subcommittee to do something and they did it. And I was very happy to see that event he Population Subcommittee did something.
DR. MIDDLETON: You actually made them look bad by having nothing under their subcommittees.
DR. SCANLON: Can we make a decision by December 10th on whether or not we are going to have a meeting besides the breakout session? I would propose that we try to do a little bit of what we talked about today and then make a decision on whether we want to have more than a breakout session at the February meeting. I wouldn’t want to make a decision right now that we want to have a half a day at the February meeting. I would like a strong statement of what we are going to do and we will make a decision by December 10th.
MS. GREENBERG: I agree with Debbie that part of our strategy has to include how we are going to — I think we have pretty much agreed we are not having any big hearings. That was very helpful. But how are we going to use this time until February? How are we going to use the February meeting, all of that. Then how the other subcommittee is going to use their time.
DR. SCANLON: Those who have been a participant in the quality letter know that you can have a lot of meetings without anybody coming to together.
DR. MIDDLETON: Which actually led to some cycles that weren’t always productive.
DR. SCANLON: I know. But it is like there was a lot going on. There is the question of could those cycles have been avoided? I don’t know.
DR. HORNBROOK: Blackford’s notion about the geopolitical. You think about the peanut scandal and the quality of food and where that food went. If you could link all those peanuts to places in space, the eat local movement, and then you need to link local food to the soil toxicity and soil reports from the Agriculture Department Department. You need to go to the health departments and get the sanitation reports, the restaurant safety reports. Who is failing? And put those again on a grid, put all that together. You haven’t talked to any individual patient who went into a restaurant but you know the distribution of dysentery and diarrheal diseases and you know where the restaurant is.
DR. MIDDLETON: You know what might be a little bit fun at some point? I don’t know when or where, but just to may get some of the real wizards from some of the other industries — Wal-Mart knows where the truck is on the road, what parcels are in it, and if sales occur more over here they send it over there. Real time analytic stuff.
DR. GREEN: Coke and Pepsi know more about their clients by far than any health person.
DR. SCANLON: I put the wrong expiration date on an Internet purchase three weeks ago. In five minutes my phone was ringing saying someone is trying to use your credit card. This is what is possible.
MS. GREENBERG: I thought they would just reject the entry.
DR. SCANLON: They rejected the entry, but on top of that —
MR. RODGERS: Do you know why they spend money to do that? Because you are not going to be liable for that cost.
Scan: I know.
MR. RODGERS: And that is what we have to tell them. You are going to be liable for that cost — health plan, medical —
DR. SCANLON: The payer, in not having that information is liable. Well, thank you very much.
(Whereupon, at 5:30 p.m., the meeting was adjourned.)