[This Transcript is Unedited]
Radisson Hotel Reagan National Airport
2020 Jefferson Davis Highway
Arlington, Virginia
CONTENTS
Welcome and Introductions
Agenda Item: Welcome and Introductions
DR. CARR: We'll take a minute or so to look at the charge. Don Steinwachs, I think he is coming. Who else are we missing? Marjorie. I'm Justine Carr, Caritas Christi Healthcare, Member of the Committee, Co-Chair of the Subcommittee on Quality.
DR. TANG: Paul Tang, Health Medical Foundation, Co-Chair of the Committee, no conflicts.
MR. QUINN: Matt Quinn from AHRQ.
DR. WARREN: I'm Judy Warren sitting in - Standards is going to keep an eye on you guys.
DR. W. SCANLON: Bill Scanlon, Member of the Committee and the Subcommittee, no conflicts.
DR. MIDDLETON: Blackford Middleton, Member of the Committee and of the Subcommittee, Partners Healthcare in Brigham Women's Hospital, no conflicts.
(Introductions around room)
DR. CARR: I think we have at our place the various charters and if you want to just turn to page 3. We dramatically truncated the detail that we had before because we want this to stand as something that will translate year over year. Maybe just take a moment to look at it, just the three paragraphs, and see what you think.
DR. TANG: It talks about looking at emerging data needs including methodologies, technologies, database and networking and perhaps data collection instruments. I think in fact that's what we've talking about the most like PHRs is one example. Maybe other ways of doing quote surveys but anyway that's sort of missing.
DR. MIDDLETON: My mod to Paul's suggestion would be to leave off instruments and just talk about data collection.
DR. TANG: Data collection is a noun.
DR. CARR: This is in quotes because it's taken exactly from the charter. I'll look at the charter because there may something on data collection already, and then the third paragraph.
DR. TANG: The concept of collections is also missing perhaps.
DR. MIDDLETON: We can stuff it in the tracking stuff.
DR. CARR: The individual participates as a fully engaged participant. Any other suggestions?
DR. MIDDLETON: A word that is missing, it may be intentional, is sort of community whether a person-centered view of health includes –-
DR. TANG: It has community. A set of personal, professional and community activities.
DR. MIDDLETON: I'm sorry. There you go.
DR. CARR: Hearing no other recommendations I'll amend this and this will stand as our charge for the next two years.
Welcome Marjorie and Debbie. I think we're now ready to move to our focus, which is the hearing that we propose.
DR. TANG: My recollection of the high-level concept we talked about before the hearing, to cover in the hearing, where one novel data sources about individual health (including health behaviors and health contexts), privacy and security of data repositories in exchanges for personal health data, data standards and certification, Medicare PHR pilots, and an existing methodologies to collect data on individual and population health.
DR. MIDDLETON: Can you go through the whole list one more time?
DR. TANG: Sure. Novel data sources about individual health (including health behaviors and health contexts), privacy and security of data repositories in exchange for personal health data, data standards and certification, Medicare PHR pilots or other demonstrations, and existing methodologies to collect data on individual and population health.
PARTICIPANT: What was the last one.
DR. TANG: Existing methodologies to collect data on individuals and population health, and the population's health I guess.
DR. MIDDLETON: Can I just raise a terminology issue, which I didn't get to raise yesterday, and is of interest to me? It may pertain. When we talk about population health my knuckleheaded thought is that we're really referring to, in most cases, a cohort, which is defined by a disease, state, or condition as opposed to broader perhaps community-based measures of health, and wellness, which are not the disease stage specific, or what have you. Do we mean to imply or can we include both?
DR. TANG: I think we included both because clearly the existing talks about demographic whether it's geographic or age or gender, disparities - things to uncover disparities. And then in fact I think less so we were talking about the disease specific or context specific, but I think that's included, both included.
MR. QUINN: I think that one of the paradigms that we're looking at is a lot of our measures today are healthcare measures, claims, derived from the healthcare system today whereas there could be other sources whether source by consumers or source about consumers, their setting, their behaviors, et cetera, that they could inform.
DR. MIDDLETON: For self-report data and aggregate of those data.
DR. CARR: If I may I would just like to put out a kind of hierarchical model because I find that these four things each would take us down a different direction and I think that we would be naïve to think that we are going to do something that's all four of these things. I would like to begin by making the case for novel data sources. I think that this comes from our discussions even going back a couple of years where we talked about in the quality what was in the quality work group, anticipating where the puck will be as Carol called it in reference to population literature. Rather than just looking at what is now but look ahead. Where are we headed and how do we anticipate that and add value? I think the committee has had a long history of anticipating where things will be. I think that's the role that makes sense especially in this environment where now suddenly there are so many different groups looking at health information technology and also quality.
My point would be that if we really seek out what are some of the novel ways that individuals one, manage their health not in an autonomous fashion but in some kind of data driven fashion, and first understand that. That it's not just a data element. It's not an A1C but it's two-dimensional, longitudinal data with interventions and response to interventions. I think is very important and exciting and meaningful data. One would be to understand what's out there.
The second thing would be what happens with that data. One let's just say worst-case scenario it makes health better for the individual. That in and of itself is a terrific thing. But beyond that if we wanted to think about is there information that, as Carol would say, data by-product of that. That could then be shared with a physician if appropriate or with other groups and then that brings in all the issues about standards, privacy, security and all that. I'm interested in sort of looking at what some of these novel approaches are, their effectiveness, and if we think that they are something that has traction then say okay now as we think about this in a model where the data could be then used, transmitted, and aggregated, what applies. I'm just not ready to jump into security and they wanted to even know what we're talking about.
MR. W. SCANLON: I would agree with you completely but I wouldn't move the Medicare to personal health pilots up into the novel data. I don't think you can start to ask the questions about privacy and security and of course standards until you know what we're talking about. I don't think we've got a strong grip on what we're talking about. If I were to read the paper that's going to come out of this, what I would expect is the first section would be the description of these novel data methods and then we would deal with the implications of that. It's kind of hard to write the whole paper in a hearing without having done the research about the first part.
MR. QUINN: What I was going to say is that we're looking at the data sources going up. Another way of looking at this would be to look at our abilities as a nation and to measure health today and say what's working well. What's not working well? Where are the gaps? Where are places where we can improve timeliness of data, comprehensiveness of data, accuracy of data and then seek out sources that are different than we have today for that? Then exam what the challenges or applicability of those are to support what we have. That's looking at it from a top-down perspective as opposed to a bottom-up.
DR. CARR: I like your idea in bringing us back to remembering identifying the gaps. The gaps are in relation to population health, the health of the nation. What are the methodologies that we have today with really just survey in many ways creating the shrinking budget? What's the gap and then going to this what are other novel data sources?
MR. W. SCANLON: The gap is one reason why I go back to the charge and to the Full Committee. Part of the charge is the national health information policy. I'm going to maybe interpret this in terms of IT and the whole issue as IT is out there and that we have a concern even if it's not about developing macro, aggregate information. It's about the flows of information that may never be aggregated but we worry about them from a privacy perspective. We worry about them from an efficiency perspective, which brings us standards. Even if we don't and a building up from this we still need to be concerned as a committee because we've been given this additional assignment beyond those statistics.
MS. GREENBERG: First of all I completely support trying to focus this because otherwise I think we're going to get just a little bit of everything. It won't be satisfying. It won't lead to something concrete that you feel anything to make some findings and recommendations on. I guess I'm still kind of feeling this conflict between the ideas of the novel approaches, although yesterday I was saying why limit ourselves to novel. I can go with that fine. It would certainly be interesting and I've think we've learned from it and can lead to other things. But then what is novel? I assume when you're saying the novel you're talking about not these PHRs that are fed by claim state or whatever so why are we bringing in the Medicare demonstrations? I'm interested in those but those aren't so novel.
DR. CARR: I guess you're raising a very good point, which is, do we have a day and a half on novel. Is there enough novel to talk about? Probably not. Maybe we think about a spectrum of novel.
MS. GREENBERG: Also I think if we're trying to do this in a collaborative way we serve the needs of the privacy and security subcommittees was mainly over there if we go a little beyond novel. Novel is maybe kind of on the cutting edge but it's not that common where is I think they are trying to get a handle on some of the privacy and security issues related uses of personal health records that goes beyond just what's on the cutting edge but the broader spectrum of personal health records. If you're not just sticking with novel then I can see the Medicare demos coming in.
DR. WARREN: I guess one of the questions I'd asked since we're looking for gaps, are we also looking for what we should be collecting? To me we don't always collect what we should because it's too hard to collect and no one knows how so we collect what we can and then use that as proxy. At some point instead of just looking at gaps maybe we should look at the shoulds. That might help us then identify what could be novel methodologies as we look at that.
MS. GREENBERG: Surely the issue is not the gaps because we probably collect them.
MR. LAND: Maybe besides novel data we ought to be looking at novel uses. It might be director of data that has been used in a population sense in novel ways.
DR. MIDDLETON: I was trying to think actually of organizing principles of which to address the problem. Which are the novel data that might be of interest and which of those data might be the most valuable. I'm struck by two charges which stick in my mind forever. One is the one, I'm trying to find the reference, which shows for a thousand people in a community on any given month 200 have a symptom of some kind to take note of, 9 see a physician and 1 gets hospitalized. A great deal of stuff is happening outside of what we ever are exposed in healthcare delivery system.
The second is one that I mentioned yesterday of course Larry Green's great work showing the attributors, the determinants of premature mortality, behavior, genetics, social environment and healthcare and again healthcare is the minor component contributing to premature death. With those two principles in mind I think we need to look across the community landscape and understand and what signals do we wish to detect from a community perspective for all those folks to not seeing the healthcare delivery system. Even among all people how do we measure in behavior, genetics, social and environment in determining health outcomes, or status, or utilization, et cetera to frame this conversation? This may not be a one-time shot. This maybe something we have to deal with over time.
DR. TANG: I was probably a lot with Judy in the sense of just like we don't want to be tethered with what is. We don't want to be tethered novel either. I think it's much more normative to think of what data are required to assess the health wellness of an individual and the population and then work backwards. That's just a slight amount of what you said. Instead of focusing only on novel why should we ignore the existing methods or the existing to get at them? It's almost like in a sense privacy legislation should attach restrictions to the data not to parties, people who would touch the data. Likewise it's the data that's abused in assessing in population health and then the things attached to data, the standards attached to data, the privacy restrictions and practice attached to data, et cetera, rather than chasing any –
MS. GREENBERG: It's kind of early in the morning. I like Garland's suggestion about novel uses also of maybe not such novel data. We were talking at the BSC, the last time or time before, about problems of getting good race and ethnicity data and death certificates. I suggested that one way to address that would be to have the person report this. Before you think that I'm really novel the idea was not after they died but while they were still alive.
DR. TANG: Why not do it when they are dead. That would be truly novel.
MS. GREENBERG: That would be novel. That would be definitely novel. I said I'm not that far out. The person is taking more ownership of their health information. We know that certain things are best reported by people about themselves; this idea of people gathering certain information or compiling or storing certain information that could then be used in all sorts of records including their death certificates. To me it's sort of an interesting idea.
DR. MIDDLETON: Just a piece of evidence that you might find interesting. In the patient's site application and care group in Boston, which John(?) et al run. You know that the amount of data that is actually submitted by a patient is on the order of three percent compared to all other data, which is in the record, the structure data, et cetera. I don't have high hopes actually for people. The third of Americans don't balance their checkbook. You know that kind of stuff. I don't have high hopes for people submitting a lot of structured data. Our experience is similar actually.
MS. GREENBERG: I don't know if they would submit it. They would store it in some conditions that it would be accessible.
DR. MIDDLETON: Let me finish the thought. I think the acquisition of those data is really the interesting part because someday they may come from the patient but not require any manual entry by the patient. That's kind of the trick.
MR. W. SCANLON: What about pre-death certificates? I would go along with Paul. I would add to this idea of assess I think when we originally started talking about this there was the idea of influence the health of individuals. It's how things are mentioned no data can assess and influence the health of individuals. Keep both of them on the table because I think there is a public interest.
The other thing, I don't mean throw a wrench into plans, I think we need to focus on an agenda first and then decide the amount of time that you need to have for a hearing. I really see this as potentially sequential. That you come to grips the answers to some questions and you think about them and then you go on to the next. You can't anticipate that you're going to be ready to go on to the next set of questions at the beginning without having done the homework to really investigate the approaches that you're going to be looking at.
MS. GREENBERG: I don't know if you can do something in January. Then maybe you could also do something linked to the February meeting like the day before or the day after so it would give you a little bit more bandwidth.
MR. QUINN: I really liked what Garland said and I think that the definition of novel is relative. In the realm of vital statistics information there's a lot that's novel, not to belittle the work of my colleagues at ark, maps, and other things. If Walmart or google or a lot of organizations really want to understand the behavior of their customers, potential customers, web surfers, they are not going to bond giant survey that requires people to fill out forms and collate them and 2 years later you get a pretty cool report. They are going to do it in novel ways and ways that aren't necessarily new data. We've been scanning purchases for a long time. We've been using the web for a relatively long time. It's really about thinking about the data sources and thinking about what they want to measure and then figuring out what the best way of measuring it is and understanding the data governance, issues around it, the data sharing issues. I would say that there's a big piece of this that goes into the broader secondary use category around health information exchange, around PHRs, around information from more than one practice.
DR. CARR: Just to recap. If we talk about and begin with what are data needs about how. We'll start with that. Then we say what about novel data collection, data sources, data uses.
DR. TANG: Let's not restrict to novel.
MR. W. SCANLON: I don't understand why we would throw up perfectly good measures. It's like doing new technologies when off the shelf pots would work.
Isn't there a link to IT and the fact that we're talking about data that's been assisted by IT as opposed to kind of really old methods?
DR. TANG: Actually the birth certificate isn't a faciscous remark. If it's already there and it says your ethnic background why throw it away and go chasing after when you're dead?
PARTICIPANT: It's actually no. It's like the mother's ethnic background and the father's.
DR. TANG: It still works even today.
DR. CARR: Data needs about health and then data resources assisted by IT and then it could be data collection, data sources, data uses.
DR. TANG: Data? Collecting that data, protecting, transmitting, stewardship.
DR. CARR: Again, I feel like everything has just come back on the table.
MR. SCANLON: When we started this discussion at some point back, months ago, it seemed to me we were really talking about this individual, a novice individual would be assisted by data in terms of maintaining health and then there could be a by-product of that which would be that we would be able to assess or monitor health for the individual as well as the aggregated for a population. We started with this individual and then thinking about and that's where the personal health record comes in most directly. That's the most prevalent link not between IT and the person.
DR. TANG: That's just an example.
MR. SCANLON: Right. We're talking with expanding it. Expansion Justine just mentioned is that we're now into this –-
DR. TANG: We need to understand the data that can assess and influence, assess individual health then how do you collect it. One old fashioned way is to go look it up on the birth certificate that's already there. A new way is a PHR but that's just an example. Another is what Matt said there's behavioral things that we know, your credit card. There are all kinds of stuff that's already there in electronic form and in fact classified like credit card. How do you get the data? But then we also have the privacy aspect of it. How would you make sure that that's used for this and only this purpose? There's the how would you understand it, the standards piece of it. We're not tethered to the novel thing. There is a novel thing which is we worry about individual health with population measures as by-product. That was our novel concept I think. We're now chasing after the emerging data needs to satisfy those needs to the extent it's encompassing in these other areas. I think that's okay but we're not chasing down –-
DR. MIDDLETON: Maybe we should try to redefine exactly what the objective is right now. Are we trying to define the series of presentations and what not that will inform the thinking. Are we trying to revisit the charter?
DR. CARR: What is a worthy topic to bring folks together to advance our knowledge along these lines? Again I'm very concerned that we not get entangled in privacy, stewardship and standards before we have a picture. We had this talk when we did the secondary uses thing and it came up with various things. You can have a data set that's extremely important, valuable, value added. You can have a Netharian who takes that and does something bad with it. That is certainly true. That will always be true. If we have an opening sentence say there's data you can get here and we go right into privacy we're never going to get the conversation forward. There's something that you just said you know if we really think outside the box and we should have Carol and Paul here.
If we wanted to know about health and behaviors and we have credit card purchases and websites and queries on google about flu symptoms, we actually could today paint a picture that incredible about the health of the nation. In the hands of netharians privacy would be violated and standards would be but we're not there yet. Let's just say we live in this IT world. We're using a typewriter and we have a computer. It's sort of like that. We have tons of information about health and health practices that's out there. I guess that's the part that I want to say. I want to be unconstrained by standards and privacy but just say what information is there. We have no money for surveys. We have all these challenges and yet we have this richness of data.
MR. QUINN: I would frame the goal of this testimony or perspective, and this is my perspective on this, is seeking to better understand the emerging data needs for measuring and tracking population health in the 21st century and informing policy around funding, how to collect this data, funding additional work into examining what are the implications of sourcing it both from a security perspective, a standards perspective, and other things but to unearth those things. Today we measure population health this way and we have these resources and metrics were doing it. What should we do? As additional data sources become available and we additional data needs what should we do?
DR. MIDDLETON: I was stimulated by Matt's comment and yours, Justine. In a way it might be useful for us to kind of floats and travel on to see where they fit. I'm struck with the question about what is the data, which matters the most. Playing a pediatrician just for a second, for example, my wife will for sev(?) on screen time for the adolescent. How much screen time the kid has in front the TV, in front of the computer? How will be possibly getting that piece of data even though it's one of the primary determinants of obesity and adolescents. It's there.
DR. TANG: That's the Garland thing. What you described was like you said the google. It's already there.
DR. MIDDLETON: I don't know if that one's there.
DR. TANG: It's almost like the IRS, and I'm sorry to use that metaphor, people report stuff of private financials to this agency that does protect the privacy. I'm just assuming. If there was an agency of the government that collected all this stuff and was able to hold it private just like the CDC does and yet extracted the stuff that we can learn about behavior and create the correlation, et cetera, at no extra cost. That would be a novel use and novel collection.
DR. CARR: Let me just say one thing. I'm picturing back to your grids. On one side we have where are our needs. The conversations that we've had about what are our needs for population health. What are our resources for that information, the conventional and the novel? We go through them one by one as we want to know about obesity. We want to know weight. We want to know activity. We want to know screen time. You can call and ask a hundred people on a block that or you could think about gathering it. We unencumbered ourselves about stewardship, privacy, all of that. We simply say there's a whole world out there and it exists and people are using it now so to say well we can't do that because it would be privacy it to encumber ourselves before we have the model.
MR. W. SCANLON: Another question. One is I think that this capacity already exists. On Monday we got a letter from a manufacturer from Vinaigrette dressing saying that it was now going to be stocked again at a grocery store. It was a personal letter. I think it was triggered by the fact that when we checked out at the store and we used our consumer card they recorded that. I get things from warranties that I never registered but somebody has identified that I own this thing and we're having a recall and stuff like that. It exists.
We have to think of ourselves I think as government here. We are a government committee. There's notion in government is thinking about, talking about, the idea that we could monitor kids screen time. For my grocery store example we could be sending notices saying you're buying too many Twinkies. They know that you're buying Twinkies. If you buy yogurt they know what flavor it is. This has always been one of my examples which is the grocery store knows about your flavor of yogurt, Medicare can't tell you what services you use for a year. The plan takes a year to process the coins. It's that kind of a disconnect. I think we have to be careful about how much we save. We would contemplate government's involvement in this advantage of this capacity. Think about it on talk radio being discussed about the government committee is sitting down and contemplating looking into your most private aspects of your life.
DR. MIDDLETON: I think what we're talking about is data needs and data management as opposed to who is doing it.
MR. SCANLON: We have to think about what are the public needs and how the public needs can be satisfied.
DR. CARR: This is going on and we are defining it so that we can precisely move forward with privacy and standards. This is happening. I think you define the problem before you solve the problem. If we define that there is this universe and then we say it can't be like an old-fashioned person that says I don't like those new fangled things because those new fangled things are here to stay. To say you don't like them because it's not what you have. I look at my kids and the world they live in. It's a different world. It has all different privacy. They know everything about what people had for breakfast in California. That is the reality. We are naïve to not harness it and we're not naïve to not manage it. I think that it begins with define the universe of things and then what can we learn. I keep going back to that google flu epidemic. If google knows before CDC knows that there's an outbreak on Eastern Long Island isn't that telling us something?
DR. WARREN: There are things like that. Within the Web 2.0 Tools there are things called mash ups. It's where you take two independently available data sets. One of the things that a lot of police departments are doing now is with their county they are doing a mash up between a google map and crime. Anybody can go in, google their address and find out what crime is occurring around their home, or sex offenders. We're also beginning to see disease outbreaks so looking at who's got pink eye, who has the flu, things like that. Those are things that are occurring right now that local communities are doing that I bet you they haven't figured out they can use.
DR. CARR: We spend all the time talking about how are we going to advocate for more money in the budget, for more surveys to be had, and we've heard year over year it is getting less. Why don't we continue that avenue but then amplify and think creatively?
DR. TANG: Just as you were saying it would be naïve assume these things don't exist, it would naïve to think that the American public would like us to use it without adequate protection.
DR. CARR: That is what we are all about. This is why we are cross cutting then we begin the privacy conversation, then we begin the standards conversation. It's an important issue to manage and to problem solve not to say we can't do it but how do you put things around it to make it gain the value without compromising the privacy. I think then it becomes a very important –
MR. LAND: I think our problem has been the surveys, as aestheticians and epidemiologists, we're very precise. We do it in a very precise way. We're ensured that it's good science. Now we're moving into an area of using data that is a little bit fuzzy but it's being used in the corporate sector very effectively and the scientists we're not too sure if we want to go that direction. When I read about the google announcement I saw two things. First of all I saw that CDC was involved in it which I don't know if they are or not then I also saw some public health people challenging it because it's not the way we've always done it and we're not really sure if it really is credible information. If we can bring out the framework or what's going on out there then it challenges us to see if it really is useful information.
DR. CARR: Does an NCVHS need to create the platform for these conversations to have the important conversations and that's what we're doing. Marjorie, because you are the Executive Secretary please give us our concluding comment and then we'll go to next steps.
MS. GREENBERG: This is so like too what we're trying to do to update the vision for the 21st century about statistics. They are completely linked to each other. The one gnawing thing in my head is is this fully or even really what we mean by persons centered health? I thought to some degree that the person-centered health was what is important to people. People are checking things out and keeping things in their personal health records and checking things out on the Internet. I guess we'd have to assume that's important to people. What information also are people in the best position to provide as opposed to information being filtered through clinicians, et cetera. That's just a question I have. Is this really what we need?
DR. CARR: Right. So you raise a very good question. I think that we're not prepared quite probably for a hearing in January. We are maybe looking at tagging onto February. I guess we need to just set up a call and frame these things and define it. Do we have a chance to record out today and see what the Committee says?
MS. GREENBERG: It would be good to have Larry as part of this discussion.
DR. CARR: You are right. We have person-centered as kind of one approach and unless we veered off into something else.
DR. WARREN: I don't think we have veered off to me. What we talked about is all the different kinds of data that are out there that contributes to their health and well-being. What really excited me was when you turned that around and said, our task would be to create the platform for future discussions.
To really lay out that territory what are all these things are popping up? Whether it's going to your grocery store and scanning through so that you're managing their resources but by the way we also get data about what kind of food you are buying. It's all these other things that are coming out because of IT. Once we get our arms around that then we can go forward and talk about standards and privacy and the rest of it. We can't have those discussions until we know the what.
DR. CARR: I think we're onto something. We'll have some discussion. It'll be an interesting segue to the de-identify data conversation that we'll have that we're going to update on that hearing in there. This is kind of a little bit like that.
MR. QUINN: As a next step why don't we have a call in January to discuss this.
DR. CARR: You've got to do it way earlier
MR. QUINN: Do we want to have this meeting in conjunction with our February meeting?
DR. CARR: I think we need to have a call the first week in December? Could you set that up?
MR. QUINN: I will not be here.
DR. CARR: You'll be on vacation?
MR. QUINN: Yes. I will be back on the 14th.
DR. TANG: So you're assuming a 1984 scenario and asked for a way to control it is to do it safely.
(Miscellaneous comments)
DR. CARR: When we were in 1970 and we described 1984.
DR. TANG: We now have knowable information. We, one, have not linked it yet, or at least the good guys haven't, and two we don't know how to (laughter)
DR. MIDDLETON: I almost wonder at a level Paul, given where we are at Justine, at this maybe heretical. Do we need actually to have sort of just a visioning session just to get this nailed? What is the thing we're trying to paint a picture towards? In a way it's getting the baseball diamond built. Build the field and then they'll come, security, privacy, everything else.
MR. QUINN: Paul and I put this together prior to this and it's overview goals and just a straw man agenda and presenters.
DR. WARREN: I got the grid part. I didn't get the word part. I read it at the meeting.
MR. QUINN: Let me send this around and on this one, I wrote it just based on our previous conversation. I started with the Healthy People 2010 as a start and went to Healthy People 2020.
DR. TANG: Why don't we edit it before you send it around again. There are multiple versions and I think we need to go another round.
DR. CARR: We kind of have a small huddle to kind to integrate what we've heard today and put out a couple of straw mans so that we can respond to what we plan.
(Whereupon, the subcommittee adjourned at 9:35 a.m.)