[This Transcript is Unedited]

Department of Health and Human Services

National Committee on Vital and Health Statistics

June 12, 2014

Hubert Humphrey Building
200 Independence Avenue, SW, Room 705-A
Washington, DC 20024



Agenda Item: Call to Order, Review Agenda – Clarifications, Updates

DR. GREEN: We need to do our formal customary convening at this point and
have everyone introduce themselves, declare your conflicts of interest. I have
heard some beeps as people joined on the phone. Before we run the table and the
audience, let me just make some comments about how the agenda has been adapted
and adjusted.

In light of an unprecedentedly rapid approval, development and approval of
letters yesterday, so if you are calling in for the action item at 9:10 about
the HIPAA report and the data standards letter, I have good news for you. They
are done and approved. You could do something else with your time. We certainly
are going to do something else with our time and that is primarily to keep this
conversation is now well under way. We are going to come right back to it and
continue it. I suspect that this will take us right up to the break at 10:30.

We also have just another item or two of housekeeping. Walter, why don’t you
start us off with introductions and we will go around this way.

MS. JACKSON: We have run the table generally. That is already in the record
and it is all part of the full committee block.

DR. GREEN: We only need to hear from people on the phone or who are here who
were not before. We will start with Lynn.

DR. BLEWETT: Lynn Blewett from the University of Minnesota, no conflicts.

MS. MILAM: Sallie Milam, West Virginia Health Care Authority, member of the
committee, no conflicts.

DR. CHANDERRAJ: Raj Chanderraj. Member of the full committee, no conflicts.

MS. BOOTH: Vickie Booth and I am the lead of the Population Health Metrics
Team at CDC in Atlanta.

DR. COHEN: Bruce Cohen, member of the full committee, no conflicts.

DR. VAUGHAN: Leah Vaughan, member of the working group, no conflicts.

DR. MAYS: Vickie Mays, member of the full committee, no conflicts.

MS. SQUIRE: Marietta Squire, staff to the committee.

MS. HOGAN: Katie Hogan, Department of Veterans Affairs.

DR. GREEN: Do we have anyone else on the phone?

Debbie, report the weather airline cancellation news.

MS. JACKSON: We were very productive yesterday, as Larry just mentioned, and
clarification about the speakers proposed for the morning. We got a notice from
Dr. Karen De Salvo that she will not be available for the 11 o’clock
presentation, but we are working on trying to get her in for later this
afternoon. She is traveling this morning and she will not be in for the morning
session, but I did give communication with the co-chairs for the Working Group
Data Access and Use, so that if she is available this afternoon, we are looking
at about four. We are targeting four. Then she will come in and present to us.
Apparently, Dr. Charlie Rothwell from NCHS is not available. We have some open
space. We will continue discussion for the morning session.

DR. GREEN: We will continue our adaptive approach. The working group plan is
going to need perhaps to be a little adjusted this afternoon. Unless you
object, we will take the position that if Dr. De Salvo can get back to
Washington, DC and if she can include us in her agenda at some point during the
day, we will aim to suspend our other plans and take advantage of the
opportunity to hear from her about her priorities and have a dinner
conversation. Any objections to that plan?

Is there any other business we need to do, Debbie?

MS. JACKSON: I think the agenda has been clarified for the record so we can

Agenda Item: ACTION – HIPAA Report and Data
Standards Letter

DR. GREEN: I would like to transition back and I want to ask your indulgence
here. I heard people talk about the power stories. I want to contribute to the
discussion I just heard by telling you a story about the way my life has gone
since we adjourned yesterday.

Colorado. It is a lovely place. It is on fire in terms of its interest in
population health, redesigning health care delivery, changing the way we pay
for it. And the federal government is present. AHRQ is there. AHRQ has this
lovely RFA out to develop extension service opportunities with a target on
dependent variables with cardiovascular risk and health indicators. The CDC is
there with various projects trying to figure out how the health departments can
use the resources they have to help communities figure out their problems and
get to the data —

NIH is there funding CTSAs and community engagement. The community
engagement folks are six years into developing community liaisons. There are 17
of them now scattered in and about the state with very interesting
relationships sometimes being housed within not for profit entities within
communities that nonetheless are trying to connect through and figure out how
to do the business arrangements to allow them to do their work without being
employed by the university or the state or whatever.

The state is totally committed to submitting a proposal to CMMI in the next
three weeks, focused on integrating the care for people in Colorado so that
some time in the not too distant future about 70 to 80 percent of the
population has access to integrated care when they have emotional and
behavioral problems.

State health department is elbowing its way to the table trying to define
where they fit in with all of this stuff. We are trying to define where the
agricultural extension agents that are scattered all across the state, which
AHRQ’s work is modeled when it comes to extension. How do they fit in? The
university is not a pure thing out there. We have a lot of institutions of
higher education. And Colorado State University, a land grant institution —
shouldn’t we do this? No, this is ours. Health Science Center says shouldn’t we
be doing this. You know what. It is a lot of really good people with a lot of
good ideas trying to do the right thing. We are having a hell of a time pulling
ourselves together into anything approaching coherence to try to move along

Some of you talked about the connectors. I just want to insert into our
continuing conversation that I am relatively certain that there is an imminent
and urgent need to have some countervailing force to fragmentation,
fragmentation of effort. When I listen to what Vickie said a moment ago, this
convening role that we can play, this safe place where the people who need to
get together and talk to each other to figure out how to do stuff, to discern
where opportunity lies, that seems like something worth continuing to
contemplate that we might be able to do something useful there. I have not
heard exactly what that is yet, but I am hearing contributing ideas.

One another thing. Nobody out there agrees what the metrics are or the
measures are that we are going to manage to. But almost everyone agrees that we
cannot improve what we cannot measure. There are definitely data definition
needs that will be harmonized in some way in some way or another. A lot of this
is over our space. Leslie’s comments about — one of you were talking about
access to the data that you actually need and using it properly and not messing
up a community.

I just want to confess that I live in a space in my day job where virtually
everything that has been said is not theoretical. It is not impractical. It is
not nerdy. It is what the hell is trying to happen. Colorado. Of course, you
know that I am not authorized to speak for the State of Colorado. But I can
tell you from where I live and work and play that Colorado could use some help.
I do not think we are alone.

I think this conversation we are having is hugely important and very
relevant. It is targeted and it is timely. Between now and 10:30, we can begin
to form some consensus about where we want to go. The next year or so of our
work I am convinced can be useful.

Go back to the Population Committee’s enterprise.

DR. COHEN: Great. Walter, you, in a second. I just want to address Ob’s
question, which was are we making any assumptions. My reframing of the question
will be are we making any assumptions about where communities are at in terms
of their activation and understanding of the problem.

MR. SOONTHORNSIMA: One and then also engaging the members of the community,
in other words, the community itself, not the community leaders, not the health
policy people, public health officials because that is really the — I
understand that these are the targeted audience, but beyond the targeted
audience is really the committee —

DR. COHEN: I guess as we move forward we are going to have to make certain
assumptions so that we can progress, else our enterprise will be too diffuse.
But I do not know what those are right now. That is the way I would answer it.
We need to focus on a particular situation. For me, part of it is understanding
the context of what we are doing. There are already a lot of existing tools
that help us understand context, but we are looking for our sweet space.

MR. SOONTHORNSIMA: That is what I am asking. Whether that could be —
because everything else may have been done. Are they really connected to their
communities? Are they really going after the right problems? Are they really
engaged, the community member? Maybe that might be a sweet spot. I do not know.
I am asking the question.

DR. COHEN: I think we need to discuss it. I would say our strength and
expertise is more technical than the interrelationships between community
groups and how they represent communities per se, but that is my initial
thought. I could be dissuaded.

DR. STEAD: Let me just make a suggestion and it may put Vickie slightly on
the spot. I do not want to interrupt running the cards — making a process
statement for when we get through that.

What Vickie shared with me and mentioned if there was time that she would be
willing to speak briefly to and this expanded time this morning might give us a
chance to do that. Surrounding the established process in the community health
of both needs assessment and improvement planning piece, which is a structured
process. As I read the National Quality Forum, it essentially has the same
process. As I looked at the toolkit being developed by Privacy and Security, it
has a very similar process and certainly, in the convening and assessment
stages, it is really identical. We might mention that because that is a context
that exists. We are actually going to use that context as a framing for the
work we are doing with the framework tomorrow. We are sending the use case in
it just as a way to target it. I think to let Vickie have a little time after
we run the tents might be very helpful in just bringing that in front of us.
That is just a process suggestion if that is okay with you and Vickie.

DR. SUAREZ: I was going to tell a story, but it will be a much shorter
story. In an earlier part of my life, I actually worked quite a bit on
something called community-based participatory research, which indeed is very
much what Ob was pointing to, the idea and the interest in ensuring community
and when we say community, we do not mean specific individuals in the
community, but the community at large are engaged from the start. In some ways,
some of the conversations that I have been listening to and hearing still lack
a little bit of that concept of a community-based participation in this
element. I wanted to try to bring that back and try to — we are always going
to I don’t want to say struggle, but at least have to ensure and have to find
new and different ways to engage directly the community in these processes. I
just want to put a plug for that of exploring new and more engaging ways to
involve communities and communities as in the real communities, if you will.

The second point I wanted to make is really about conceptually we have been
talking about the health of communities. And the word health has always been
part of the evolution of where things are and where we are going with our
system here and our environment. I have always been saying basically that we
have moved from an industry that deals with diseases and manage disease to
focusing on health and improving health to ultimately and this is where I want
to really see as going is wellness, the wellness of the committee. And people
many times confuse and mix the terms health and wellness.

What I wanted to put a plug on was to conceptually expand the scope of this
to really look at wellness rather than health. We talk about the health of the
communities and healthy communities and all these things. We are ready to move
to the next level of conceptualizing this to look at the wellness of the
community and not just health and healthy community. Community engagement and
wellness of the community are my two messages for this effort.

DR. WALKER: I want to tell a story to express my concern about our talk
about data. First of all, I think it is spectacular that the government is
releasing data. If you wanted a quantitative, systematic review of treatment
for heart failure or something, you need the raw relatively unprocessed,
patient-level data to do that systematic review.

What we have found is that once someone has done that systematic review,
publishing it does not have all that much — disappointingly small impact on
physician behavior. We said the thing to do is to teach evidence-based health
care medicine to doctors. We did that. That did not have very much effect on
anything. We said what we need then are journal clubs where doctors will get
together and review evidence together and that had disappointingly small
effects. People said the ACP and others said the thing to do is to produce
reviews of reviews and turn it into optimally readable, actionable reviews of
research. That maybe had some impact. Maybe where we are now is saying actually
if we built that into the electronic health records and other information
systems then clinical decision support and other things would have an impact.

My concern is that for doctors who are not the dumbest people and have some
training in the use of evidence, it turns out that what we call data is rarely
of any use at all and even very systematic and careful efforts to turn the data
into pretty focused answers to highly relevant questions is often only a start.

When we say data, I assume we are meaning relatively raw, relatively
unprocessed information. There is only a certain community, even fairly
sophisticated communities, only a certain subpopulation that that is what they

DR. NICHOLS: I want to pick up on Larry’s enthusiasm about Colorado and how
it affects our world here and just say I am as much as a tired guy my age can
be, I am equally enthused about all this. I want to try to focus us therefore
on what we really want to happen in October because amazingly October is a hell
of a lot closer than all of us acknowledges.

I loved the way Vickie described this three-part conversation, but then as I
thought more about it, I got the following kind of queasiness. I think if we
have the federal government teach us in public in October what they are doing,
it is embarrassing. We need to learn from the federal government before then.
We need the federal government in the room in October and maybe they get to
respond or something. I think two things. We are the federal government. We
need to learn from the federal government before in public. I am willing to go
to CDC in August and do whatever.

The second thing I would suggest is that go back to Ob’s question. What kind
of community can we really get there in October? Larry may be acutely aware of
the fragmentation in Colorado, but I can name ten. He can name 30. There are 30
places in Colorado that have come a hell a lot farther than most places in the
country. And what I am trying to say is the places we can actually organize it
came from the list we all gave Susan and others. Those are places that have
done it. That is to say done it to the level where they wanted to go. And what
I would suggest that would be most helpful is to have them come and tell us how
they did it, why they did it, what they used and as we learned what they used,
then the feds could say you should have checked this. But I do not think we
want to learn from the feds what our capacity is in public in October. We
should learn from that on our own time.

DR. VAUGHAN: Good morning. I really appreciate all the very interesting and
thought-provoking comments thus far. I did want to loop back to the issue
around and question around data literacy because I do feel strongly about it if
you do not want to call literacy data engagement and to frame that as another
social determinant of health. While the purpose might not be to create a
universe of data — that will be fine or ninjas or whatever you want to call
it. There may be somewhere we least expect them and we should not presume that
they are not there in the community, but to at least have people who are data

I think that there is a huge problem right now around data literacy and it
is not necessarily where we assume it is. My leading example would be vaccine
preventable diseases where the group at highest risk for data illiteracy or
higher socioeconomic people and the majority. That there is this huge unmet
need where all sorts of mischief happens and to take that seriously.

I found it at different levels and I thought I would give three quick
examples to speak to that. Within technologists, there is sometimes a casual
approach to what is health. This would be a great idea. I will do health things
now without any real understanding of the data’s limitations or its risks. But
there is great promise there I think especially in partnerships.

One recent example I enjoyed was with Health Technology Forum and San
Francisco Department of Public Health. They recently came together and had a
behavioral health code-a-thon and did a real deep dive on data with all the
teams, curated many data sets, which were already public. A lot of them were in
PDFs and had to be reframed and digitized and Socrata was a wonderful partner
in accomplishing that as well.

The second example goes back to Chicago with Smart Chicago, Dan O’Neil, and
Chris Whitaker doing amazing work in partnership with the Chicago Department of
Public Health, particularly Raed Mansour. The initiative is called Smart
Chicago. For their civic apps, they actually convene community panels to
approve or disapprove the panels. Anybody in the community can sign up. They go
through a little course. They get some lunch. They get carfare, transit fare,
and then they review every civic app that comes forward and thumbs up or thumbs
down and comment on it. They are doing that armed with information. They are
doing armed with respecting their time and participation. It is very exciting
what they are coming up with. The food apps, food poisoning detection algorithm
is very interesting to work with, which I have had the pleasure of doing with
Raed. But I commend that initiative and partnership to you.

The other very fun recent example with actually last weekend with about 110
17-year-old girls in Oakland, California. Their hack-a-thon was called Love is
Respect. It is to reduce teen dating violence and prevent bullying. The girls
had a lot of great ideas and were very passionate and determined what they did,
but also working through data with them. Number one, it was an interesting
process for all the teen initiatives, which are wonderful. Most of them were
not written certainly for middle school grade level. The most important people
who needed to understand the information had to have it distilled. But walking
through them with data and their idea of getting stats was to Google it. We
walked and we worked through basically what is the beginnings of a curriculum
on how do you find data, what is good data, and what do those numbers mean.
They got it right away. It was very exciting and very fun. It was an exhausting
and exhilarating weekend.

I did want to add some additional partners that I thought might be
considered to be invited to underscore just how often I hear from people in
state and local government, what about us, and to include them. But to also
think about other agencies. For example, the Department of Education is doing
some amazing things. They have very much to do with health. It is just that it
sits in education. And some of the other agencies are doing similarly fine work
as well.

To look at the safety net and I have spoken with you about the safety net
provider networks that are emerging and that they are on the front lines and
responsible for gathering much of the data. But they also have great ideas, but
would welcome the support and being heard and I think would add a great deal to
the conversation.

I would also encourage the inclusion of librarians who are often the
community data brokers and universities, including people within public health
who I also think are not necessarily being heard and not necessarily aware,
kind of vaguely aware that there is this data thing going on, but meanwhile are
plotting through their human subjects review and this kind of whole other
system that needs to stay intact is not the only thing and resource available
to them. I think often times their expertise and wisdom has not been used to
the greatest advantage. Thank you.

DR. COHEN: Fantastic ideas. Thank you. Susan, can you follow up with Leah
and get some of those contacts and names. That would be great.

DR. BLEWETT: This is Lynn Blewett. Is this a good time to add my comments?

DR. COHEN: I see your tent is up, Lynn.

DR. BLEWETT: I mentioned this at our last meeting and I just want to make
sure it is still on the table. The National Health Interview Survey, which is
our nation’s leading health indicator survey has a lot of population health
information and currently only has released this data for about 32 states. And
when I started in this field, it was at 20. They have increased it to 32.
Somehow, it would be great to have a statement or some kind of — I do not know
what would be appropriate for this committee. If we are talking about community
health data, that just seems to be so fundamental that we have the National
Health Interview Survey available in all 50 states and the District of Columbia
— those estimates and work that they are doing on small area estimates as
well, but at minimum the state’s representation in that survey.

DR. COHEN: Thank you for reminding us about that there are a lot of federal
data partners. The BRFSS folks are pushing stuff down to the county and many
states in the BRFSS community are really trying to focus on sub-county
estimates as well. I think there is activity going on among federal agencies
around providing data, approaching what I would call the community level that
we really need to capitalize on and we should definitely make sure that NCHS is
actively involved in these discussions as well as other federal agencies.

DR. NICHOLS: Can I just ask a question, one of clarification? Lynn, this is
Len. We pronounce our names the same, but spell them differently. Is it the
purview of this committee to make recommendations about things like 50 state
and so forth? Isn’t that sort of what we do?


DR. NICHOLS: Then I would like to second Lynn and say maybe we can put that
on the table for serious consideration like what we think the highest
priorities would be in this universe. Obviously, we are not ready to do that
today, but it might be a smart thing to do in the fall.

DR. COHEN: That would be a great outcome, I think.

DR. MAYS: I want to pick up where Walter was starting and it kind of weaves
in also, to what Ob was talking about. I think we have to be prepared for what
it is we are talking about we want to do. It is prepared not only by who is at
the table, but it is also kind of what you are saying is. Let us not embarrass
ourselves. And the let us not embarrass ourselves is that a lot of times when
you talk to the community about data, they want data differently than we are
talking about. It is much more in the context of what Walter is talking about,
which is this wellness notion. We are focusing on things like our standard NHIS
and NHANES and things like that. They are trying to get at the other data. They
are trying to find out about police violence and how long it takes for 911 to
answer calls. To them, the health outcome of somebody dying is that there is no
trauma room anywhere nearby.

They are trying to figure out how do we make a data case in order to get
ourselves a trauma center back in our local hospital. They are trying to figure
out when the police arrive at somebody’s home and they shoot first and the
person is mentally ill and wasn’t there some other way to do this. And they
always say can you tell me how often this happens. When we go to the mayor and
we go to the city council, the whole notion of the social determinants of
health and this is where when you are saying let’s not embarrass ourselves.
That is not the stuff we often are bringing to the table. That is what the
community often sees, as this is the data that I need to go to Congress, to go
to the city council and be able to say this has to stop. We will be a healthier
community if we can get our pollution taken care of. It is all those social

I think a very significant part of the hearing probably should also focus on
exactly what Leslie said is that we need to hear from communities about what
data do you want that you have asked for and you cannot get. What data do you
want that you think makes the case to link up the health that you have because
of the context in which you live? That means some of the usual suspects in
terms of the federal government is probably not them. It is probably a deep
dive into CDC and to say tell us about your violence, the National Violence
Index. They often do a lot of the community data in prevention.

It may not be the population surveillance data just that we want to focus
on, which is what we usually do, but we need to think about some of the
community data that actually comes both from the big NIH studies that out there
and available as well as CDC. It is the standards people over here are talking
about the community-based, participatory response, which is just wonderful.
This is like the ultimate of what Larry is going for as the integration that we

DR. COHEN: Some great points. To pick up on this particular thing that both
Walter and Vickie are talking about. When we talk about health data from the
traditional sources, what we are really talking about is disease and illness
information where we are not really talking about health or wellness.
Government has never really focused, as far as I know, on community assets and
resources and in any systematic kind of way. Government has been really good at
the federal level and to some extent at the state level of collecting
information about illness and disease, but never about health. That would
really spin the paradigm on its head if we try to focus on establishing
structured ways to get government involved in collecting information about
health, wellness, resources and assets. That is one thing that we should

Leah said something that really resonated with me. What do these numbers
mean? I guess that is the fundamental question. For data geeks like me, there
are a lot of numbers. We can generate them as the building on Jim’s thought. Do
doctors, communities use data in its rawest sense? Is it our responsibility to
provide road maps so that communities know what these numbers mean? I have been
in lots of community situations. Sometimes they want the data to tell them what
to do. More sophisticated communities integrate the data into a larger
decision-making process based essentially on their values and their beliefs. It
does not mean that data are not important. It is just data don’t answer the
questions. But hopefully they can be one input into helping communities make
better decisions about what they want to do.

DR. WALKER: Just one sentence. What I was trying to say is what Vickie said.
If you start with their questions, answers to relevant questions are a
different thing than data.

DR. COHEN: That is a great frame. One more thought for me. I loved the word
engagement that Leah said. I liked the concept of community data engagement. I
think it is better than readiness and I think it is better than literacy. I am
putting that on the table as a possibility for us to frame our enterprise
around engagement. I think that gets to the bidirectional nature and the need
for a conversation about answering. Maybe it is community information
engagement rather than data. We can parse those words in our continuing

DR. STEAD: I am just trying to get sort of like time check where we have
another 40 minutes before break. I am hearing a lot of convergence in where
people’s heads are. I just wonder what we need to do to now begin to get this
tightly focused for your statement at the start. You just began to get at one
part of that, which is how do we define the space. It seems to me we may need
to begin to structure our time now to get a clear definition, at least the one
we are going to run on for the moment of the space. Then why are we doing this?

I liked Larry’s passionate story around counter force to fragmentation
because there is a lot of stuff going on. And the last thing we need to be is
one more thing like the stuff that is going on. In essence, what we want — I
believe our purpose is to align. We want to define the target and then we want
to be the force that aligns and enables bringing together — I keep forgetting
what the word from the Stanford thing of collective action, collective impact.
What is the essential difference that allows that to happen and what
conversation can we have in the fall that will be useful to that purpose?

Then I think the thing we really want to walk out of the fall, I believe, is
if we have a stated thing. We know we are trying to align people. Our
particular thing at the end of the day needs to be a draft of the federal role.
If we could figure out how to design something that walked us through that, I
think we might get narrowly focused and be more targeted. I do not know if that

DR. COHEN: Thank you for getting me back on task. Everybody whose tents are
up is welcome to continue with their thoughts. I would like us to spend the
next half hour trying to focus on what it is we want to do in the fall, how to
frame this what we are calling a roundtable or a discussion probably for two
days. And then I would like to do that for half an hour and then leave 15
minutes for Vickie to update us about what is going on at CDC.

DR. FRANCIS: Shouldn’t we hear from Vickie first?

DR. COHEN: Sure. Is that all right if we hear from Vickie first?

MS. BOOTHE: I may have slides.

DR. COHEN: While she is preparing that if some folks had some last —

DR. SUAREZ: This is a comment specific about this. I prefer to wait. This is
a comment specific about how to organize the workshop in the fall.

DR. FRANCIS: I wanted to tell the story about why I raised the question of
what data and not being able to get data. There is an area in Eastern Utah
called the Uintah Basin. Recently, there was quite a lot of publicity about
really high infant mortality and miscarriage rates in the Uintah Basin. Now
there are many possible contributing factors to that. Obviously, it was very
disturbing. This is aggregate data, very disturbing to the community to realize
that their infant mortality rates were way out of line.

Here are some possible explanations. One possible explanation is people get
— it is an impoverished area. People do not lead very healthy lives and they
get into prenatal care late.

Another possible explanation is that there are very high ozone levels in the
Uintah Basin, increasingly high associated with oil and gas extraction
activities that have been going on in the Basin, including fracking.

A third possible explanation is the quality of health care out in the Uintah

If you can get data about some of these like the unhealthy lifestyles of
people who live in the Uintah Basin, but not about others, you are going to get
a very skewed picture. Actually, when these high infant mortality rates were
published by a local journalist, they immediately got we shouldn’t even talk
about that because fracking is so important. A lot of push back about not from
the people in the community, the people in the community who are not happy
about the high infant mortality rates. That to me is an illustration of —
maybe the federal government actually has a role to play because the federal
government can be a stimulator of getting data in a way that local folks might
not be able to do. That is the story that the EPA could play a role in all of
that. That is the story that led me to say I think the issues here are more
than just are you good at using what you have.

MS. BOOTHE: And I have slides. Let me just start by saying I am a data wonk.
I spent 14 years at EPA as an environmental engineer and 13 years ago, I
thought — EPA is wonderful, but what they are allowed to do and what they are
charged to do is very specific and prescriptive. It does not always lead to
improved health and improved population health. I thought I want to go to CDC
and do the soup to nuts kind of thing. I went to CDC as an environmental
engineer and decided — at NCH that is narrow. Then I went into healthy
communities and overall population health. Data has been at the heart of
everything I have done since I joined the federal government 27 years ago.

I am sure a lot of you know that the IRS now requires non-profit hospitals
to conduct a Community Health Needs Assessment every three years in order to
maintain that non-profit status. There is over 3000 of them in the United
States. Not only must they conduct this Community Health Needs Assessment, they
must engage with the community and they must engage with the state or local
health department. This non-profit status represents $22 billion a year to
them. There is big financial incentive in this.

I have spent the past three years — my team and three other offices, people
in three other offices within CDC have spent the past three years trying to
figure out how to prepare and leverage this requirement and how we get
everybody on the same page. The health departments are doing community health
assessments too as a prerequisite for accreditation. They are on a five-year
schedule. Hospitals are on a three-year schedule. The hospitals have to reach
out to the health departments. Health departments are already stretched and
cannot really manage the responsibilities that they have. Your example of
fragmented systems. I have seen it up close and personal for the last three

What we have been trying to do and this is with our policy office, our
office of state, local, tribal, and territorial support and with the Division
of Community Health and Chronic that is doing the communities, putting
prevention to work and the community transformation — these groups have come

A lot of what we have done and are working with comes out of the
community-based participatory research arena about how to meaningfully engage
the community. You see — meaningful community engagement and shared ownership
is the underpinnings.

Now having said all of that, I have been mostly doing trainings for the epis
and the states and the locals trying to help prepare them to stop, not stop —
they have been doing restaurant inspections. They have been doing outbreak
investigations. They have been providing clinical services in some cases. This
is a whole new arena.

The take home point for the trainings that I teach is that data is good.
That data does not give you the answer. As a matter of fact, we do this — I
always do this exercise where I have Littleville in small town and I actually
give them the data and these are epis. I have them go through and analyze the
data. It includes air quality data and it includes water quality data. It
includes violence and crime. I have them go through. And then I have them go
through it. We teach them prioritization schemes so the Hanlon Method,
multi-voting dot matrix. I have them come up then with their top three
priorities. These are all epis, all from public health, all with the same
background, their priorities inevitably are different. And then I tell them
they all have the same data. It is really horrifying to them because they
walked in just knowing that the data was the answer.

And then I have them think about this is you guys. You all have similar
backgrounds. You do similar work. You went to similar schools. I said now think
about how you are going to be prepared when you engage community-based
organizations and community members themselves. They all come with their own
backgrounds and personal experience. How are you going to use this data
intelligently and present it and have a dialogue with them and understand what
it is that they want and need? That is the point of my training.

I just want to say that — this is the focus. I think you guys — it was
amazing to me how close you guys are to where the thinking is around this whole
community health improvement process. We teach our epis that four things come
out. There are four products that come out of that community assessment. It is
the analysis of secondary data. What are the most prevalent, severe, and
important health outcomes and health determinants in your community? I will
show you the data that is recommended by the ten seminal guidance documents on
community health in a minute.

You do your secondary data analysis to come up and you look at trends and
you compare the benchmarks, peer counties to state averages to healthy people
targets. You get your shorter list from doing that. You go back through it and
you look for important disparities or you look at mapping life expectancy at
the census track level. That will paint a story like nothing else. In
Baltimore, as you drive across Baltimore, you lose 20 years of life expectancy
because you were born, lived, work and play three census tracks west. All that
is data driven.

But then the other two products are what the community thinks important.
What are their priorities, not just in health outcomes, but in their social and
physical environments? Finally, what are the assets? I am always up there
saying not just health system assets. I am always emphasizing. We are always
emphasizing the difference between empowering a community for sustainable
action where everyone has accountability and ownership and that is doing in a
community health assessment with a community as opposed to doing a community
health assessment on the community. Drop in. Do your little study. Funding goes
away. You leave that job. And then the community is left there probably worse
off when you started.

We talk about assets, not only of the health system, but what are the assets
within the community that they can bring to bear and that you have these
things. These are critical inputs before you start to say these are our
priorities and these are the things that we are going to act on.

Having said all of that, when I first started talking to the hospitals that
pretty much stay within their walls in the community when they think about
their mission, what it is they are doing. When I started talking about
meaningful engagement with the community, another look of horror. And then
understanding the data on food deserts, but air quality, water quality, on
crime, on no streetlights in a neighborhood and the impact of that on
population health. They have been on a big learning curve. They have come a
long way.

I think the only other thing that I want to say. The framework that I had up
there a minute ago. It goes all the way back to an effort. Some of you may have
known of this. Way back in the 1980s it was patch a plan approach to community
health. Jim Marks is now at Robert Wood Johnson. It was out of the chronic
disease center back then. There were lots of efforts. Public health got into
this community improvement. Then I believe Congress decided, no, let’s go back
to categorical funding. Let’s go back to chronic disease. You are over here and
here is your line funding. Infectious disease. You are over there.
Environmental health. You are not. But this framework. And then there have been
many that have followed. There is NACCHO’s map process, the ACHI, the Catholic
Health Association. They all have these common steps.

Having a common framework and a common language and understanding and
talking about where it is that you want to get to and thinking about upfront
what are those assets and how to engage the community members, that is the
important stuff in this arena. The data is important too. Having that dialogue.

I used to work at ATSDR. They have to do the health assessments around super
fund sites. Dialoguing with them about cancer rates in the community and the
different ones and what we know about the pathways and learning how to
acknowledge and hear what they say and understand, but then having that data
that you then go through and learn together in a back and forth and telling
them the limitations of the data. We just do not know kind of things. That is
the hard things, I think, in the process.

The data is easy and you do not know how many people have told me let’s
build a system and the non-profit hospitals will log on and maybe they will
have a data wizard. It will access all of this great data. Then it will spit
out the top three priorities that are going to get the greatest increase in
life expectancy and QALYs. We will measure with QALYs and DALYs and HALYs. That
is what I fight is because the answer is not in the data.

With that, I am going to stop and see if there are questions.

DR. COHEN: Thanks. Could we get a copy of your full set of slides or is this
from something else?

MS. BOOTHE: This is from another presentation. I am happy to pull some
together that I think would be —

DR. COHEN: That would be great. Thank you.

MS. BOOTHE: I am sorry. I was going to do one more thing. When I first
started doing this, the hospital said the only data we have are for people that
come in through our days, hospital discharge data — there is no data about the
community out there. The public health people said I am overwhelmed with data.
The — county-level estimates and environmental health has environmental health
public tracking network and infectious disease has — now we have biosense. It
is in a cloud. We do not know where to start.

To answer that question, we did — I am in the division with the community
guide and systematic reviews. I am doing a systematic review in meta-analysis.
We tried to apply some of that same thinking here to this question. We did a
systematic review of all of the guidance going all the back to patch up to the
2011 IOM report about performance and communities and came up with the 42 most
frequently recommended health outcomes and determinants for community health
assessment. And what I want to point out is the social factors, the social
environment. The numbers that are behind — we do have leading cause of death.
That is important. We do have injury-related mortality and motor vehicle
mortality and that is important.

But I want to show you the balancing that is going on here. The numbers
after one is how many of the ten guidance documents recommended that particular
topic. They have smaller numbers, but they are continuing to increase. We have
social capital and social support, violence and crime, violence abuse and child
abuse, domestic violence and child abuse, poverty, air quality, water quality,
housing. We use this as a guide. This is what we have been working on.

Chris and I cross all the time because we are working in very similar lanes
here. We have worked on community data that complements what they are doing and
complements the county health rankings that has county level or sub-county
level data for all four of these topics that are going to come out.

These are important. We say that you should consider all of these and
probably some more things too. We have added access to healthy food and places
for recreation to our list because of the body of evidence that has evolved
since I finished this in 2011.

MS. JACKSON: Thanks so much, Vickie. You can see how similar that is to our
21st century health statistics graphic that we have been using for some 10, 15
years and how Marjory was very helpful in bringing that voice in for Vickie and
others from CDC. Thanks so much.

DR. NICHOLS: Can we ask questions. First of all, thanks. Thanks again. That
was really good. Have you done or have you begun to do what I will call a
meta-analysis of the CHNAs that came in. These numbers are from the guides.

MS. BOOTHE: I am working with our OSTLTS Office, State and Tribal. And I am
working on the hospital one. This is my passive surveillance system. I have a
Google alert on Community Health Needs Assessment and CHNA. Pretty much every
other day or so I have a new Google alert that announces the release of a new
Community Health Needs Assessment. I have analyzed 41 of the hospital ones. A
few of the things that I can tell you that are interesting that will be coming
back with feedback on is obesity was the top one. I thought it was going to be
access to care so access in quality to care is there. I was very pleased that
access to care I think ranked second or third. I do not have this with me. It
was not number one. Mental health was up there. I was pleased with that. I have
only done 41 and I am going to try to get to 90 of them before — I am happy to
share the results if you guys are interested.

The other thing that I wanted to share is the hospital — the rulemaking
around Community Health Needs Assessment actually happened in the middle of the
process. Things were grandfathered, especially how to do engage with the
community and who, how to engage with public health and who. It was involving.
I think about 50 percent engaged with public health. But the ones that did were
much better in meaningful engagement with the community. That is the big one.

DR. NICHOLS: I would just like to add — wherever you are, 41, 96, 124 by
the time we make our site visit in August, which I am quite serious about. That
would be a wonderful thing because that will help us shape our conversation.

DR. COHEN: If we could, I would like us to focus our last 20 minutes on
thinking about priorities for the fall event. Walter, I thought you had some
suggestions there.

DR. SUAREZ: Thank you. Yes, I did. Well, first of all, Vickie, thank you so
much for this. I am very excited about the work. I have said before here in the
community to me the future of a lot of what we do and how successful we are is
in population health management. This is truly something that helps us all look
at that dimension.

Very quickly, one comment I think from the — that we have had and this is
Kaiser Permanente that I represent with some other work. Certainly, there is
the goal of having these community health assessments done on a hospital basis.
Interestingly, in hospitals is not really where wellness happens as I was
earlier mentioning. It is not sometimes where health happens either. It is very
much — but certainly, it is the one focal point that one can point to a
non-profit entity that receives certain task benefits from a status.

What we have found is that while we can create our own plan, there is only
very few communities at the state or local level that actually get together the
various hospitals, for example, in that community and try to coordinate across
hospitals a more integrative plan, if you will, or harmonize or whatever the
right word is. In many cases, what is happening is sort of everybody trying to
do whatever they need to in order to meet the basic requirements. Sort of what
meaningful use is all about.

Going back, thank you so much again. I am very excited about this. Going to
the comment or the question or the point about the fall event, I want to again
emphasize and Vickie, thanks for bringing up again this concept. I think not
only when we think about health in the context of what we have been talking
about, we are perpetrating still in my mind some of the concepts of health
being simply the nonexistence of disease. It is in many cases still the same
challenge that we have. I want to try to emphasize the concept of wellness by
actually creating a core set of domains that will be the focus of the full
event around wellness. As Vickie pointed out, the communities in reality are
interested about all the things what we probably more as the physicians and
executives around this table probably think about what the communities want.
They really want to know about home and safety and food and occupation. When I
say all those things, people think those are not health. They are actually a
component of the wellness domains.

My suggestion is to create a picture that shows what are the domains of
wellness and health including all these elements and then try to focus on what
are the communities interested in terms of the data for specific domains and
bring in examples of how communities are looking at issues of safety in their
neighborhoods and housing and environmental conditions and other things that
are more encompassing of wellness. But then trying to focus too much on our
traditional view of health and all these factors about mortality and
morbidities and all these things. That is my suggestion.

DR. FULCHER: Thank you, Vickie. Nice to see you again. Just a short summary
about what we have been doing with CHNA in terms of implementation with the
work with CDC. Our team worked with Kaiser Permanente and built their national
or their CHNA platform. That was outside of community commons. They liked so
much the public good utility concept of community commons that they gifted back
the code and said you need to make this CHNA tool available for every hospital,
non-profit hospital in the country at no cost. Again, it is making all of these
data and these indicators available.

What we have is through the work with not just with CDC Division for
Community Health, but with ASTHO, NACCHO, ACHI, Catholic Health Association,
Kaiser Permanente, the county health ranking folks, United Way worldwide. A
large group of people got together and worked quite hard at making this public
good utility available this CHNA tool. It is out there. People are using it. We
get a lot of good feedback in saying we do not have money to go to private
sector vendors and spend $30,000 plus dollars a county where we can get this
publicly available data to take us on that first step of our journey.

The Community Health Needs Assessment is just a report. It is a starting
point for that community process. That is what we engage CDC further around
this digital journey ideas. How do we engage communities along a process that
uses these indicators, these reports in making meaning of them? That is the
second point.

Finally, we have been working with community action agencies for a number of
years. Community action agencies are really around poverty, alleviating poverty
around human services. The Administration for Children and Families part of HHS
is really focusing on community needs assessments, but it is very disjointed
from the community health needs assessment process. The names of the people and
they are hospital centric, but if we are really going to look at community, we
need more of an integrated health and human services approach around expanding
this needs assessment framework.

I just want to add that I really appreciate the work with CDC and that there
are tools out there living that are publicly available. The challenge is in the
public sector, we do not have a lot of funding for marketing. Actually none.
How do we get the word out? It is mainly through friends and family, but that
is not good enough. How do we again align our messaging to help people know
what is publicly available for them? I just wanted to comment on that.

DR. COHEN: I wanted to follow up with you for a second. Given what CDC has
done and what you are doing and what Kaiser has done, where do you think the
sweet spot for us would be in the fall to build on what has already happened in
this space?

DR. FULCHER: I think the sweet spot would be really not working at the data
level, but the indicator and making meaning of that and pulling the right
parties together. It can be consumable. So much of what we are doing around
data is not consumable for communities. A sweet spot could be around bringing
together those. Am I understanding your question correctly?

DR. COHEN: Could you just give me an example of what that would be?

DR. FULCHER: Again, bringing community action agencies, some NGOs that are
working around community health improvement. Another tool we worked on with CDC
or through Community Commons is this target intervention area tool around
improving community health. It is really elevating the conversation and
bringing groups together to say what is missing in what you are looking at with
your suite of indicators that really can round out our understanding of
community and take us further. The sweet spot is filling in the gaps with those
non-traditional organizations that need to be around the table.

DR. SUAREZ: If I may just jump in very quickly to say just to follow up on
that. I think one example specifically of that would be the handful of
communities where the actual agency that is coordinating all the hospital
community assessments is taking an extra step and coordinating the work, not
just of the hospitals, but also the rest of the community participation in
creating a community health assessment that is really not hospital based or
hospital centric or one by one hospitals, but actually more comprehensive.
There are actually a handful of those communities that are taking this a step
farther than just simply receiving community health hospitals.

DR. COHEN: Comprehensive health planning. I was involved in that 40 years

DR. FULCHER: With the united ways, for example, is that — how do we look at
community in a very broad health perspective? It is really moving away from the
word. Just a traditional health approach, but a more comprehensive community
needs assessment that is inconclusive of these different sectors.

DR. COHEN: Vickie is next and then Leah and then Susan.

DR. MAYS: One of the things Bill said is let us start moving toward the
specifics. I think we are starting to hear some. I want to build on those.

I think the notion of first for us the work is defining how we are going to
approach health in a community context, not health as we do it within data
collection. But if we are going to the community, it is health within that
domain within their context.

And then I think the notion that Walter is talking about in terms of the
domains, I think we want to go and think about social determinants. A lot of
the communities are trying to build from that perspective of social
determinants, not from our — again, the data perspective.

I think if we do that, what we will see is some intersections with the
federal government. They are the ones we really want like what Vickie Boothe is
talking about. There, you can bring in somebody like a Jonathan Fielding who
helped to actually chair the committee, has to do this in a day-to-day basis in
his job, can really translate, for example, how that happens.

I think the other thing we want to do is to not bring questions to the
table. I am changing before because I had said something about this. But I
think that at this hearing, we do not want to hear problems. We want to hear
solutions. We do not want to bring in people. It is not what they did not have,
what they cannot do and all of that. I think we want to stay in the domain of
picking the solution makers and figuring out from them what are the ways in
which you made this happen and how did you work with the federal government.
How did you work with the state and local government? What do you need to keep
going? Then I think in the room, the synergies that will happen will be
incredible if what you are bringing are the solution makers. Some of them are
going to be in different domains. You are going to have people who are great at
doing it in the hospital, great at doing it at a lower community level, great
at doing it in a neighborhood.

I think in addition to having the social determinants as domains, we want to
also start to slice this across different places. I think we need to dig down
and find somebody who does this at a neighborhood or small community level.
Pick somebody who does this at the hospital level. I think we need to think
like the slices, who is in the room, and making sure that also we pick federal
partners to come. I am trying to deal with the embarrassment here, but pick
federal partners to come in that have something to offer. We are going to hear
from SAMHSA later, for example, in the work group. But to pick the partners
that also are in the midst of doing this. Of course, that is going to be CDC.
That may be the CTSA people. But I think we should really keep this on —
everybody has problems, but right now, we are trying to maximize solutions.
That would be the design stuff.

MS. BOOTHE: I just wanted to — you asked the question about the sweet spot.
I thought I would tell you a little bit about where the processes are falling
apart that have the data link. And then I want to say one more thing about

The first place where things were falling apart is — the non-profit
hospitals and the health departments really — they are setting priorities.
They may not be the perfect set of priorities. They may not have gone through
the perfect process. They may not have engaged everybody. But they are coming
up with a set of priorities. Understanding and doing enough analysis to support
those in pretty rigorous processes and review of available data is happening.
Kaiser is a great example. They used 80 indicators. That was the indicators.
They benchmarked this. It is knowing what interventions are out there that are
effective for their populations that they can afford that they have the
expertise and that are acceptable to the community. We have an obesity problem.
Who doesn’t have an obesity problem? It is continuing. What do we do about it?

And then going back to what was said about being measured and Pat Remington
said about knowing where you are going and your goals. Having data. The BRFSS
data. In our system, it is the seven-year rolling average. That is fine. That
is what you need at the county level to get stable estimates. You are not going
to use that to see how well you are doing in whatever intervention you have
taken because you may move the needle and it will take ten years to see it

The small area estimates. Fabulous. I have one of the leading statisticians
at CDC on my team doing that work. You start trying to explain the modeling
technique that you have that, the limitations with the model, the assumptions
that were made to the community and into your lost at that point. We have
indicators. I will show you. I have 300 of them in my system. We have
indicators galore and we pulled them. It is coming out of that and being able
to think about how you are going to move forward, what data you are going to
need, and what evidence-based interventions there are for the problems that you
identify. If you identify a problem. Alzheimer’s, fifth leading cause of death.
And functional status and definitely not associated with wellness for the
individual or their family. We do not have a clue as to what to do about it.

CDC. There is no better place for them to be able to tell you the origin and
the trends and leading causes and multi-level analysis. The proportionate risk.
Population attributable — there is not enough emphasis on what do you do about
it and how do you monitor that and figure out and do mid-course corrections
when what you have selected is not working in your population.

DR. VAUGHAN: It is a huge issue.

I want to loop back to one of the things that Chris has said, but also
touches on Leslie and the other Vickie, Vickie Mays. And framing data as being
consumable for communities. And to loop back to — maybe it is the engagement
piece again and maybe it is framing it as a design issue, but to be thoughtful
about how we are presenting data and making it available. For example, growing
the APIs that are available that we are not recreating a new silo because for
regular epidemiologists to access an API is often met with a quizzical look.
And yet their data are rich ninjas all on their own. As a community member
wants it as a PDF, it is expensive. We should not let that be in our joy in
moving forward in growing these other communities to not leave the others

People accessing data. Some really interesting folks I have met actually
both within civic tech community here in DC that I have met with, but also a
wonderful ePatient whose name is Fred Trotter. And their way to approach lack
of access to data was to FOIA things. And that has led to some really
interesting results.

The DC community, FOIA, the 911 response data for the District. They were
working on it and these were all tech folks, not health folks, but they had a
personal experience. That was what had driven their inquiry and the FOIA and
all this little group of folks pulling stuff together, but they had already
done some interactive mapping and different responses by neighborhood. It was
very interesting to see what folks driven by their own personal experience who
had some of that knowledge and could not access the data, but were DC folks who
FOIA thinks got it done.

But that there is also another really rich interesting community around open
access, which is not per se open data in the same way that we are talking
about, but also has a great richness around data transparency and accuracy.
Some of that goes to clinical trials and then gold diggers work of what gets
published and what does not and how that introduces extraordinary bias. But
also, the World Bank has been doing some really interesting things, open
knowledge foundation, and public library science in terms of publishing those
data with your literature.

And then one of my other favorite examples of actually getting providers to
use data around their communities is from the Robert Graham Center and health
landscape, which I think has done a wonderful job over more than a decade now I
guess and just getting information to providers about their practice and
letting them add their data. There is the local knowledge and tools in those
hands, which I think is a great example.

DR. COHEN: Thank you. A couple more comments. I do not want to over shoot
our break time.

DR. STEAD: I want to actually propose a way to close this down with the last
minute if you can save me that last minute.

DR. WHITE: Vickie, solutions. Solutions that exist are what you are
describing. I am wondering how much solutions that are needed is part of that
discussion. It is potentially embarrassing for our discussion. But I think if
you only look at what exists, you will make some progress.

The final thought as I am listening to all of this is I was reflecting very
hard on what Jim said earlier as somebody that has been developing decision
support stuff based on systematic reviews. I was reflecting on why people do
things. I was casting sidelong looks at Len because I was thinking about
incentives. I am wondering to the extent should we be able to define what
people who care about population health are incented to do and why they are
incented to do that and how that might drive things.

DR. WALKER: A quick specification. I want to join the amen corner, Vickie. I
want to propose a specification of it. I think if we get those people who have
created solutions, two of the things we would want to ask them is what
information did you need and how did you get it.

DR. STEAD: This has been a rich discussion. We will try to mine this. I
suggest the following process. Several of us are going to be together tomorrow
for the framework workshop. The framework workshop is going to be discussing
structures that could help with this and with concrete examples. We will move
the discussion.

And then we have reserved the last block of tomorrow’s workshop. It was
entitled to incorporate the frameworks into the plan for the roundtable. I
think we will modify this to — I think that Bruce and Susan and Vickie, our
facilitator are meeting at five tonight, I believe, to vacate a last cut at
getting ready for the workshop. We will figure out in that how to go in
tomorrow afternoon with a draft of the plan for the roundtable. I think we have
to get out of conversation and we need to actually be working from a draft. I
think we will have something.

And then informed by the framework, we will use the last block — the
framework discussions, we will use the last block to edit that and to answer
and to list questions that have to be answered before we close it down. Maybe
that would move the ball.

DR. COHEN: Before your last comment, Susan. Again, I totally agree. This was
a very generative discussion, but we need to move past generative discussions
into a concrete action plan. On your trips home wherever they may be, use a
little of that time to translate these ideas and the specific suggestions for
the roundtable because I heard stuff that would take us two months to cover and
we really probably will not have a two-month roundtable.

MS. KANAAN: Simple plea, which I am transmitting from several of my busy and
important informants, which is please get a date on my calendar ASAP so I can
be there. Several people said that, some of whom are in the room.

DR. COHEN: Do we have time to do that right now?

PARTICIPANT: Before we leave maybe we should put anything we know we cannot

MS. KANAAN: You can see that I put the week of October 6 in my notes trying
to point us in that direction. That is largely because Bruce and I looked at
the alternatives and concluded that that is about the only week that is
available in the fall. That narrows it down. Pat Remington said he was
available that week. I think Len told me the same thing. We need to almost
arbitrarily and as soon as possible select those dates.

DR. GREEN: Debbie, can we leave this in your hands to say we want to target
the week of October 6? Let’s go find the best day to do it.

DR. COHEN: We are thinking two days. Thank you all very much.


DR. CARR: Is the start of the Full Committee now or did we already do that?
I was late. Was the Population Subcommittee and this is the Full Committee or
we already started the Full Committee?

DR. GREEN: Yes, yes, and yes.

DR. CARR: I should introduce myself. I am Justine Carr, Steward Health Care,
outgoing chair of Working Group on HHS Data Access and Use. Where is Vickie?

DR. GREEN: Vickie has introduced herself? Do we have anyone else on the
phone? Do we still have Lynn? Do we have anyone on the phone?

DR. CARR: I am just saying if we could send someone out to find Vickie
because some of this will act in her incoming role.

DR. GREEN: That was a pretty interesting discussion guys.

MS. JACKSON: I agree. Folks might still be stunned, but it is amazing and
quite a testament to how condensed the committee has been working since we did
not have that time originally set up and the fact that we just used it, grabbed
it and took advantage of it. Quite a testament.

DR. GREEN: I am not sure what we were commenting about over here in the
corner a while ago — all across the room. We really have some urgency in
setting the date. We are going to jump all over that. If necessary, we will
have an executive committee meeting to make choices and get those decisions
made. We will keep the momentum going.

Justine, why don’t you commence. I am sure Vickie will be in soon.

Agenda Item: Working Group on Data Access and Use
– Update/Briefing on June 1-3, Datapalooza

DR. CARR: Thanks. I had the opportunity to co-moderate a panel with Damon
Davis at Datapalooza. He really did a terrific job in configuring it because it
was — folks who had used HHS data kind of like what we were talking about and
then the providers of the data. It was very rich and very fun. I could spend
hours telling about all the different things

But I thought I might start out by talking about how it has changed and
building on this convergence theme, which just is striking me in every
conversation that is going on today. Although I did not get use the new app
that Vickie gave me last night on how to make slides dance. I did put together
some thoughts that just also resonated exactly with what we have been talking
about. It is a little bit of a historical perspective of Datapalooza, but not
just Datapalooza.

It starts with on the left, top-down mandate. And some of us were here when
the top-down mandate came from Bush 2 talking about electronic health records.
That preceded Obama who obviously announced the data liberation and has gone on
to say data, not only needs to be liberated, but needs going forward to be
machine readable.

Now, what we are saying as new programs develop like the insurance exchanges
and so on, we are anticipating what is the data we will need as opposed to just
taking the data that is by product. I think that is very exciting.

I was in a reactor panel that Greg and Todd I think had upstairs right at
the very start of this. It was memorable because all of the HHS data leads were
sitting around the table. Todd had brought them together to talk about what do
you have, what are you doing with it. It was remarkable as I think back on the
body language as well as the concepts because it was still very early on. I
think everybody was in a different place and understanding what would be the
value of it. Some folks let’s say who had data like OIG data. They are like who
is going to want this. Why are we here? That was what I was reading. It was not
this sense of we together will do it, but rather these silos had things. One
would have information about children in this setting. Someone else in a
different agency has children in another setting. But those two agencies
thought very separately about themselves. Again, my personal observations.

When I think about that moment of Todd being the cheerleader and trying to
help facilitate the vision and having a reactor panel to say what are you
hearing around the table, it was a very interesting beginning.

When I think also about the first Datapalooza where I think the winners of
whatever the contest was. It was a card game, I think, about how many something
in Kansas as compared to Ohio. It was cute, but it was a starter set on using
data. Let’s go with that. And then what I saw the other day just blew my mind.
I have been struggling of how to put it together and it has been rather easy
because of the discussions that we have had here today.

As I said, we started with the mandate on the left. I think in the beginning
the first Datapalooza was all about the apps and the execution. It was I have
an app in search of a data element. If you look at the top, I have tried to put
together how has the approach changed. I have separated it out. An app is just
a tool to do something and the approach is how we think about it. But the thing
that I will talk most about is what defined Datapalooza this time was all the
things that I heard were about the customer, the consumer, the community, and
what their needs were. It was not an app in search of a data element or in
search of someone who wants it. But it was a very thoughtful thing. This is a
problem that needed to be solved and here is how we have solved it. It was
remarkable and just for that incredible.

I will add also that the keynote speakers were not as much about yeah we
have data, but really about national health policy and the future and how data
is an enabler to get to the future state. It was exhilarating.

I thought it actually resonates a lot with what we have been talking about
yesterday and especially this morning to say that I think we have been jumbling
a lot of things and I broke out five things that are distinct and discrete and
we ought to be thinking about each one as we think about how we leverage the
tremendous potential of the data.

It begins first with the data generator. That is HHS. A lot of data on the
website. Over a thousand data sets. I think sometimes we have spent a lot of
time saying is it the right data. But it is data and it is there. It is machine

Then the next step and again this sort of follows the Datapalooza, which is
there are people that can take data and say here is how many doctors’ offices
there are and here is how many grocery stores and here is the rate of obesity
and they can put it together. It is an aggregator and we can put all these
things together and we can display it on a heat map even better. I think that
was in the early evolving stages.

But I think that what is coming about is really data mining and data
integration. There is so much data. When I saw the different presentations of
people, I began to say wow. We do have that. Who knew we had that and who knew
you could use it that way? I think an emerging critical element or a cohort of
people that know how to mine data. We talked about this actually at the 60th
anniversary about what are the emerging roles and the needs for education and
developing of teams so that data mining expertise. That leads to data
integration. That it is not just I am going to put everything on a heat map,
but I am going to integrate data around a particular question.

One of the things that I will just add that I heard today is actually data
validators. I think Leslie made the point. Just because you can aggregate and
put a bunch of data together about a community, it is what is there, but also
what is not there. There is a role really for big data validation.

I began by saying on the left we have the top-down mandates. On the right,
we have had all along who knew we have just discovered with our various
hearings, there are lots going on in the community and folks have been either
collecting their own data or making do with it, trying to answer the real-world
problems as we have heard in our roundtables and hearings.

What has been striking to me on the theme of convergence is the coming
together. And actually, Leah was extraordinarily enlightening and eloquent in
helping us understand that there are lots of people that really have been ahead
of the curve and coming together. But central to this whole thing is what are
we trying to achieve. Who is the customer and what do they want?

That is what struck me the most at this Datapalooza because they had
real-world questions. They had used data. Even if it isn’t exactly timely, it
was still valid. I think sometimes we spend a lot of time in the letters saying
2013 data would be better than ’12 and ’12 would be better than ’11. That is
all true. We heard some timeliness issues like blue button has your claims
data, real time in the moment you can look at that. We are definitely getting
to timeliness. But it all came together for me.

And the final point is engagement and collaboration. Jim, I was saying, when
we first had that meeting upstairs on the reactor panel, there was a very
siloed kind of body language to the room. And what I am going to tell you about
in a couple of examples is not just engagement with the consumers, but
collaboration among agencies that was really remarkable and exciting.

Again, I said the key notes were about the national health agenda and how
this data movement is helping that. But it was not about look at what we can do
with data. Great maturation there.

They had some winners. One was Code-a-Palooza. And this was a program on
LyfeChannel, Smart Health Hero. They were using Medicare data to create an app
to discuss with the physician what are those charges and what does that mean
and what did you do. It is not complicated really. It is taking static elements
and aggregating them, but having a customer, having a conversation. I did not
see this demo, but I just put that out there. That was the kind of thing.

And the important thing is the way they got to this was they had a focus
group with Medicare patients. They said what do you need. That is where it
began. It is useful. It does not have dancing figures jumping around. It just
simply helps people understand what this is.

Let me go down to the other winner. Health Data Liberator. ProPublica was
awarded, a recognition for transformation of health care policy through
data-driven journalism on prescribing practices for Medicare Part D
prescriptions. That was very interesting journalism that pulled together the
data from Part D and began looking at trends and frequent prescribers and high
prescribers. I think this is where you can put something out there and say does
this look right to you, but not make a judgment on it. As Leslie said, you
can’t just say — because we just saw this.

In some of the CMS data, the highest charge expense was with an
ophthalmology group and it looked like one ophthalmologist must be doing hourly
cataract surgery all day when in fact it was whole different story. There is a
great lesson there that you can aggregate the data, but you have to have a
collaborator, a hand off. Does this look right to you? Is there another story
behind the story?

This ProPublica group ended up — actually, led to the OIG and the Senate
and everybody paying closer attention to this. There is a validator who is
going to go in and have a hearing and look into it and find out what is going
on. Again, much more sophisticated than what we have heard.

In the middle, I want to just talk about the panel that Damon arranged
because it represented all aspects of this. The first was Purple Binder. Again,
a simple thing. In Chicago, if you are concerned about domestic violence and
you are trying to figure out where do I go, what do I do, as Leah, you could
Google. He showed when you Google, you get everything from Amazon to Facebook
to whatever, but you do not have what you need in that critical moment when you
are ready to make an action.

What they did was simply take all this information that is available and put
it in a very readily searchable — actually, they used the term Purple Binder
because they had conversed with social workers and case managers who had a
Purple Binder, which is the paper version of what they looked at. It is very
sensible really.

When we think about all the sophistication that we do, simple things are
simple. We should not lose sight of the fact that making stuff available in a
searchable fashion is a tremendous value right there.

The second thing was Stride Health and this I thought was really cool. This
is a group that took CMS claims data and created individual profiles. If you
are a person with this chronic illness and that and you take these medications
and you smoke, your expenses in aggregate are going to be in this zone and you
are going to need coverage. How you think about what your deductible is and
what your co-pay is and what your medication is.

You might think about it this way. If you are a 23 year old who is a jogger
and works in an office, you need something very different. You can find these
apps on the web and see them. This, again, is so timely. It aligns with the
national agenda. We want people to have insurance. We want it to be the right
insurance. We do not want them over paying, under paying. We want them to have
what they need. This is the simplest simple thing. You simply answer questions
about yourself and you get a profile right there of what you can do.

I asked them actually if they had validated it and he said he would speak
with me afterwards and then he was called away. I think it is a fair question.
I am just giving you the narrative. No intention there of anything. But I think
it is a very important question obviously. I think we have to remind ourselves
that after we do this aggregation and integration, we have the ownership of

The third thing. One of my favorites. Pill box. Pictures of pills.
Physicians in the room can think about the PDR and in the middle, there are
these glossy photos of what pills look like. Back before there were so many
generics, it was very helpful. If someone said I have a pink pill, it would be
is it this one. Yes. Perfect. It was a very helpful, very important aid.

I have learned that to get those pictures in the PDR, it costs a lot of
money. It was part of a promotional thing. While it was a helpful aid, it was
not just a free service.

Over time as there are genetics, actually, there is a multiplicity of pink
pills, big ones, little ones, round ones, square ones. You cannot easily say
that. In fact, there is not a compendium that has pictures of all those pills.

These guys worked together with NLM and the FDA. And they uncovered a whole
host of issues getting to data integration. That should probably be on the list
as well because the package inserts are sometimes wrong in the medications.

When you think about it, you can understand how maybe something like that
happen. But it points to the fact that it is only in using the data that you
begin to find the validity and what to watch out for and how to improve it. I
think that was a big aha moment and they had that.

And then it turns out that the VA also wanted to do this. There was an
extensive period of time spent about what are the interagency requirements of
sharing data. I guess I can say this. What they ended up doing was bagging all
the agencies, making a CD, and walking it over to the VA where they took that
and looked at it and they have been able to amplify and expand and enrich what
was already created. This is now a remarkable a resource. I am not quite sure
where it is.

It captured my attention because it is simple. It is sensible. It adds huge
value. And they worked together. And actually, they worked together among
agencies against all odds, which I just found so inspiring.

DR. SUAREZ: Which data were they exchanging? Pill pictures?

DR. CARR: The pictures of the pills and the list of the pills. Nothing
illegal. He said it in public. I do not think it was a big deal. Maybe it was
just that there was paperwork and timeframes and there was a time urgency.
Sometimes simple things are simple.

The next one up. CliniCast uses the NCI SEER data. This one is mind boggling
because the example that he showed was if I am an individual who has lung
cancer, if I am in year one, what is going to happen to me? What is it going to
cost? What is it going to get me for the therapy I get? If I am in year five,
what is going to happen to me? What is it going to cost? I do not remember all
the details, but I remember in year five or whatever, it showed it is going to
cost a lot and it is going to be spent in ICU time. On average, people in year
five with your cancer after this kind of treatment can prolong their life, but
much of that time might be in an ICU.

It reminded me of Paul and the conversations that we have had over the years
about personally. What does it mean for me? To tell me my survival is X number
of years is one thing. To tell me my survival will be on an event in ICU
informs my decision making much more profoundly. This is just incredible.

And then finally, we are going to see this afternoon the health indicators
warehouse. A lot of us struggle with online things and checking about it. I
think the health indicators warehouse has responded to people like me who
struggled through making things simple. It is brilliantly designed, incredibly
simple and intuitive.

I just want to go back to the previous slide for a second because as I am
putting down tools, I am putting indicators down there too and it is something
that Chris just said because sometimes a lot of people are all asking the same
question. A tool is simply to make that indicator so that all the people that
are asking the same thing. My definition of tools’ execution is sort of
eclectic, but it is there.

And then something that Jim Walker said earlier also reminded me that the
EHR — we can do all our studies and meta-analyses and focus groups and CMEs
and all of that and it is very hard to get doctors to do things. At our
institution, we wanted to align transfusion practice. We made an educational
thing. I think seven people looked at it, but we made decision support in CPOE
and we had a 30 percent of improvement and alignment. When we think about the
EHRs of our repository or CPO, I guess, would be the better term that we can
drive practice over night. It was just very stunning when I went to see it.
This is such a great outcome. Let me see who took the test. Nobody. But that is
okay because that is the old way. We do not need that anymore. Those are my

The final thing is there are emerging data releases. FDA is now releasing
all the adverse actions. I can think of a million pieces of misinformation one
could derive from that, but I can also think of a million things that can get
better. When I think about it, I think about the way that it was demonstrated
in some of these presentations that you have all of the folks in the mix. That
it is a collaboration involving the data generators or the FDA and maybe there
is some data aggregation and what are the like things as they put it together,
but really the data miners, the integrators, the validators. But what is the
question? What is the question we are trying to ask? I think that is where it
all comes out.

And also, the CMS payment data and actually the New York Times did a quick
heat map. Where does it cost a lot? Where does it cost less? For me, this is
remarkable. This being my last meeting. And actually, as I put up my thumb
drive and I saw all these old presentations from 2006 and 2007, I have to take
a look at them. We have been struggling and we have been on the edges. We think
we were onto it now, but five years from now you will look at this and say wow.
Remember, we were so on the edges.

But I actually think that the discussion this morning was the best of the
best. It is great to have the multiplicity of the perspectives around the room
and the expertise in a very convergent manner. Just as we said, as Vicky said,
we do not want to hold a hearing where people come and tell us I cannot get
this. We want a hearing of what can we do. That was really demonstrated by the
incredible engagement today in the discussion.

One other thing. It just caught my attention yesterday. We heard about the
NACCHO data. And the comment was there is a ton of really cool stuff here. I
hope there is someone who wants to make sense out of it. I think that is right.
There is someone. We just have to make sure of what is the question we are
asking, who is that sophisticated data miner who can collaborate and engage
with the others and make this happen.

Others were at the Datapalooza. I know Chris and Leah. I open it up to you
to add what your perspectives were.

DR. VAUGHAN: I think probably with my data nerd hat, I was really most
impressed and excited to hear about open FDA. While it is just the beginning,
just the beginning is pretty phenomenal to think of — we have talked a lot
about this macro-level of data that we put forward and this is actually really
patient by patient bringing our intelligence and awareness about what is going
on in terms of post-marketing surveillance. I think it is potentially just
earth shattering, ground shaking, ground breaking awesome and certainly that
would be a great presentation to hear from their team, but just thinking about
the personal stories of patients who have gone through it, family members who
have gone through these terrible drug reactions and knowing that they are
having an isolation and the prospect of being able to change that in a very
fundamental way that is going to change people’s lives and likelihoods is just

DR. CARR: It was really in a way foreshadowed, the whole Celebrex story. It
was Kaiser. But everybody else who had their own big data had that same story
buried in there. It shows you that you can have all of the pharmacy data, but
you need the frontline person who says my patients with this drug seemed to be
having a high degree of cardiac disease. And then you need someone who can dig
down into that and get that story. And the power of that data will be
tremendous. I agree.

DR. FULCHER: I thought that Datapalooza this year was very good. I was
involved in the community track and I really felt that they — it was moderated
by Paul Tarini at the Robert Wood Johnson Foundation, Cheryl Wold. They helped
organize that. I think in previous years, I did not see a lot of liberation of
data for communities. I thought they really showed that up well this year.

As I presented in the very short ten minutes, you know how when you present,
you always say I wish I would have said that and you didn’t. Well, I realized
that I really do not thank the federal agencies that make all this data
available for us to be able to leverage and build on. I know it sounds simple,
but when you are up there and all these things are going on, it is really
thanking HHS, USDA, the Census Bureau, Bureau of Land Management, all these
federal agencies and all the people that are working to make that available.
That is an amazing thing.

That is something in our website. We have meta-data where it goes to each
agency and what they have. I do not think that is good enough. Put it out
there. Thank you. Put it out there that don’t take this for granted. Put it out
there that American Community Survey that they have to go and fight for trying
to continue to do what they do year in and year out. I think it is really
elevating that. That was a big aha that I wish I would have said, but that
would have taken my full ten minutes. But I really enjoyed the Datapalooza.

DR. CARR: I would say hear, hear for HHS. It really was unbelievable how
things were changing. There were several others, but one other that comes to
mind is to say the MACPAC was using claims data because they have to answer a
question about what happened with the expansion of the health insurance. What
happens with the kids? Do we still need CHIP or not? It was a very elegant
piecing together of we have this data and this data and turning it into a
family. If a parent has this and the wage is this, but the children is this
age, it turns out there is a donut hole that you could slip into. We would
never know that in the old way. We would have opinion or whatever and this was
just very data driven. Each presentation whether it was just take data and put
it in a compendium or whether it was get a sophisticated person who could work
together, it was fantastic.

MS. JACKSON: I do not know if anybody else had a chance to go. Kathleen and
I had a chance to go. I have seen the evolution of Datapalooza in the fact that
it seemed like the program itself is a learning system because watching what
they have done with what they have had over the years and they don’t just stop
where they were at one year. They just build on that and took perspective to
the point of shifting mandates. That is what came through for me and the use of
nontraditional data in just invigorating ways and just to have that all
together was exciting.

DR. FRANCIS: This is a question for data about Datapalooza. I wonder if any
of you have any sense of this. When I went a few years ago, the overwhelming
majority of people presenting there were for-profit entities. There was one
little tiny set of sessions where it was public health departments or not
for-profits. Of course, when we talk about our community data users and
toolkits and things like that, we are envisioning public entities or NGOs
mostly as our audience.

I just did a little play around on my iPhone. Most of the ones you have
listed were for profits including Purple Binder because their primary business
model as I read their website is to sell their services to providers. Providers
then know what to tell patients. Now, I do not know whether that is good or
bad. My data-driven question is has the landscape changed or is something we
really need to be aware of that basically the audience for government data is
not the audience we have been looking at, but it is the start up hoping to
figure out how to actually gain some kind of traction in the market. I am
really asking that as a data driven. Has it changed?

DR. FULCHER: That is an excellent question. I gave my sentiments to Paul at
RWJF before accepting to speak there. I gave him the notion that is very much
private sector and industry oriented. He said that is all the more reason you
need to be there. I think Datapalooza and really bringing in more of that
public sector NGO presence I think we should really work hard to do that rather
than spinning off to yet another kind of a forum is to really blend that. I
think it may be nudging on the right path. It is not there. I thought I was
more critical of Datapalooza, but then I went to the HIMSS meeting in Orlando.
You talk about being blown away by that space. I think we have a real
opportunity, but we need to really work on nudging that.

DR. CARR: What I would say goes back to what Bruce said earlier about
generative thinking. I think that we have a lot of great ideas here and then I
heard more than once yesterday was is there funding for it. As we think about a
partnership, a collaboration with a focus on the national health agenda, I
think there is a role for each party. I think that the stimulating — what was
presented. Yes, you are absolutely right. It comes from that.

But I would say it was invisible. The focus wasn’t — it is not what you
came away thinking with. You came away with ideas about this is how we do it.
For me, to understand that there are multiple participants in this and when we
feel frustrated. It is not working. It is because we have not figured out what
all these roles are.

DR. VAUGHAN: I think that it was principally for a private sector type of
audience, but I do not think that is our majority audience. It is not the
majority of the people generating the data. It is not the majority of the
people responsible for implementing the consequences of the data. It is not the
majority of the people analyzing the data. I think it is a wonderful
opportunity and reminder that as we are growing this other space, that we make
sure that we are being inclusive of all of our partners as we move forward
because they bring a lot of richness to the table.

I think Ed Sondik has been a real persistent voice in terms of broadening
the participation at Datapalooza. He was there, which was fun to see him. There
were people there with that view, but I think that they were not necessarily
featured at this point. That is an opportunity to grow.

DR. CARR: And again, I think it gets back to not what is wrong, but what is
right and all the shared learning. I just was so excited.

DR. SUAREZ: Having been attending this since the beginning, it is always
amazing to see how much a person with a computer and a lot of data can do. It
is incredible of the amount of things that can happen. In fact, in many cases,
the innovation comes from people that are thinking really outside of the box.
They bring those to organizations even like Kaiser and other large
organizations that have a lot of data and that sometimes our own data is
entangled inside in ways that we do not necessarily have the ability to step
back and look at it from a different perspective. It is very invigorating.

But since the very beginning, I always have been concerned about the way in
which a specific topic is addressed and that is privacy. At the very beginning,
the very first ones, people — this is all about liberating data. People were
loosely talking about but we do protect the data or but we do de-identify the
data or but we really do not deal with patient identified data. When you
started to dig into more details, you could see how easily things begin to
break. First of all, they did not really know exactly what they were talking
about when they were saying we really de-identify data. We really do not deal
with protective health information.

I have seen some increase in the amount of attention being paid to privacy.
Actually, I could not attend this year, but I was searching. By searching the
word privacy in the agenda, there were only two sessions really that dealt with
privacy across the board.

I am curious about your sense — about the privacy perspective inside the

DR. CARR: I think that in fact the working group on HHS data access and use
that was one of the early things and we ought to have it on the agenda. It is
complicated and the toolkit, same thing. It is complex. Just even yesterday
talking with Joy Pritts, trying to figure out meaningful use and the EHR data.
It is unbelievably complicated. I think I understand why it is not necessarily
a track for the people who pay to come to Datapalooza. But that does not mean
that it is not important. Datapalooza is really a showcase, a leading indicator
of what is going on. But I think the hard work and I think the work group needs
to think about that of how do we manage that. I think that was a very clear

DR. VAUGHAN: One of the people who participated the first time this year was
one of my favorite Bay Area people, Dr. Kalmar. She is exuberantly invested in
quantified self and wears many sensors and compares the data in ways that are
very unique to Rachel. Her observation, it is interesting because there is
actually user meet ups and groups in the Bay Area where people come together
who are developing in that space and often led by Dr. Kalmar. And privacy comes
up a lot. And the balance between this is the data is my product versus it is
about the patient is a tension that was amplified that she made a lot of
observations about. I think we have a long way to go, but I think, again, it is
an opportunity and more reasons to bring the different communities together,
not just HIPAA to be a throw all or catch all basket, but the reasons why this
is private. This is what you can do and these are some examples of solutions
maybe going back to the solutions. We protected privacy. We got the data back
to the patient to their provider and yet we move forward.

One of my favorite hack-a-thons actually was at Kaiser last year. I had used
some of the Kaiser data as a graduate student at Cal, which is great. But the
data that we were allowed to just use in the building I think has so much
richness and I think there is going to be some really great things coming out
of all that. I am looking forward to seeing that grow.

DR. CARR: I think a couple of things. I have Paul and then Raj and then Jim.
I just want to do a time check here. We have time. We are going to wrap it up
after Jim. And then you will introduce the next topic.

DR. WHITE: I will just say two quick things. First, I totally get what you
are saying about feeling the predominance of for-profit companies. That is not
necessarily a bad thing. Right now, everybody thinks that the Weather Channel
is a bad thing, but they rely so heavily on NOAA data, et cetera. But yes. I
get what you are saying.

I do not want to totally be a drum that I feel like I have been pounding on
for the last couple of months. Yesterday Judy Murphy mentioned a robust health
data infrastructure, which is a report that we have been talking about a lot
recently. It was the Jason study. I have been presenting on it a lot recently.
A lot of this feels similar to some of that discussion. Part of the technical
approach that is proposed for people to think about is requiring application —
interfaces for current EHRs to be able to make available in a much more robust
way than it currently is. The machine readable data. To the extent that we all
appreciate that it is not perfectly dealt with.

The discussion about privacy is very key. And the approach that is discussed
in that report is patient privacy bundles. Devon McGraw and I have been talking
back and forth on the issue. We are both at a place where we are saying it is
one thing to let other people plug into these things and be able to have it
shoot back and forth, but really if those APIs put the data in the hands of
individuals, puts your health information in your hands, that represents a
really significant shift in the balance of power, which having the data
confirms that power and what you can and cannot do with that. Once it is in
your hands, whom you make it available to is your business. Now there are all
sorts of issues that go along with that.

It is also one of my firm beliefs that a key role for the government is to
protect people that have a hard time protecting themselves. You see that
through a lot of what the federal government does. I think that having — thank
God there are folks like Joy who work on stuff like this because that is a
pretty key part that the federal government has to play of saying what those
protections are that people should have and should have access to. Thanks.

DR. TANG: I just want to congratulate Justine — in terms of presenting the
life cycle of the Datapalooza. I thought it was very insightful. I also want to
acknowledge that even a great splitter can —

PARTICIPANT: It took me a decade, Paul, but I got it.

DR. TANG: As further acknowledgment, I am going to add a splitting part. One
of the interesting things and you brought it up with the Purple Binder is they
had to — I am going to add data supplementer to your approach because they had
to supplement what was in the federal databases to make it more useful.

Here is the app that I saw as you talked. Maybe we need a feedback loop. The
example that was given was in the — you are looking for social services. One
of the things that you would like to know is what languages are spoken. Could
there be a feedback loop? Because I think the way these databases are put
together is you have these federal contracts. You make them submit some data to
you. But could you turn around and have them submit these disparity variables
like language to you so that it could turn around and come back and feed the

In other words, if there were feedback loops back to the data sources, the
data sources could actually collect different things because it is in their —
as well. That might be an addition to your wonderful lump.

DR. CHANDERRAJ: I have several things. First is the data harmonization. You
are collecting all the data, but the data is coming from different sources and
different people are using different standards for data. The Mars project, the
space project was the first one in the 1999 that started that. So much data
became available. And they found out they were measuring all the SI units
whereas it was all based on US-based units. The data that they were using to
analyze were different. For platelet counts, there are 68 different parameters
for estimating the platelet count. Different people are using different — for
dementia analysis, people were using the same scale in different ways. I think
data harmonization needs to be very focused and definitions for each term
should be displayed out.

And I think the Institute of Medicine convened a recent seminar on PCAST
with the patient-centered research. They tried to bring this together and form
a data harmonization.

The second thing for the data collection is there are different data
elements. They are usually good, but sometimes proxies that you need to attach
to the data elements like meta-data, for instance, if your patient is using
oxygen system. You knew that he has COPD or emphysema. Those proxies can
sometimes be used for data elements.

The third thing is we gather so much data, but we do not know what to do
with this data. Analytics have to be developed so that we can utilize this data
to make meaningful solutions like Vickie was suggesting this morning so that we
can move forward with the data that we already have. We do not need to reinvent
the wheel.

The last thing I had was data integration into science that is already
developed. Data is statistics and statistics can be a lie. We found that out in
so many projects that medicine was involved. For instance, recently vitamins
became a big subject and big large study from the US and Australia and the UK.
They came out and said further statistical analysis of vitamins use has to be
abandoned and no such studies have to be reported. Until we integrate the data
into science and make a meaningful interpretation, I think data gathered by
itself will be not fully served.

DR. CARR: I think where we are in this journey is now discovery on the kinds
of things that you are saying. I think that is where data validation is
something that we have not talked about. It is not the data. It is the
conclusion validation or assumption validation. You are right in terms of the
definitions. Paul has his moment when he discovered how smoking is measured in
the public domain versus in a physician office. That is absolutely true.

But I think that the fact that as we heard yesterday from Jim as new
programs are rolled out, we are thinking at the very start of what is it that
we are going to measure. We get a uniform definition and we have that so that
the collectors are collecting to that definition and we know that data element
is going to drive some action that we will take as we go along. I guess this is
the starting line, not the finish line. I think what has changed is that by
starting to use the data, we are starting to understand where are the
vulnerabilities. I agree. We have to be cautious to not declare victory.

I do go back to the Celebrex study. Somebody had an impression that people
were having heart attacks that were also on this drug. It turned out to be
true. Similarly, I heard at the IOM meeting last year that the Tamoxifen study
of how many years should you take Tamoxifen for breast cancer. If you go to the
data today, you can demonstrate that with incredible precision that that data
in fact was there, but it was a question of what is the question I should be
asking. Now that the FDA data is available, I think there will be more

I think your observation that there is uncertainty and vulnerability is
completely correct, but I think that, again, the key thing about the — the
focus on the national health agenda as well as engagement and collaboration
with the frontline stakeholders is what is going to be the turning point.

DR. WALKER: Very quickly. The theme I heard exemplified really powerfully.
Everybody has a computer. Everybody has access to an ocean of data. What
distinguishes for profits and not for profits is whether or not they are
capable of understanding some audience, some set of users and their needs well
enough that they can turn that computer and that information into answers to
goal relevant questions.

DR. COHEN: Just one other quick comment. This is a fascinating conversation
about the use of data and it really piggybacks on our earlier conversation. I
am reminded. I get asked. I am an epidemiologist data geek by my friends. What
would I say the data say about mammography for women? We can have data that do
not make decisions that are all over the place that you can interpret in many
ways. What is our responsibility as data generators to provide the context and
help people understand? How can people understand what to do given conflicting
data that we can generate with great ease?

DR. MAYS: This was actually a question for Larry. Are we going to talk about
how the work group —

DR. GREEN: In the next two and a half minutes. Yes.

DR. CARR: Larry, did we change what you just said because I thought you
wanted a bigger discussion about the working group?

DR. GREEN: We are going to adjourn on time and let’s make the transition to
what the working group is going to be doing this afternoon, where it is headed.

DR. CARR: So quick high-level lumping journey. When the working group was
established actually just two years ago, it was bringing together these
different groups and trying to understand how we could assist HHS. I think just
as we have a lot of Datapalooza, we have learned a lot in the working group.

As Vickie takes the reins today, I think we have been able to focus on how
they can be most effective unlike this groups who works hours and hours. The
working group is a group of consultants that come from different array of
backgrounds, very rich. When you read the charter, that is the role that they
need to play.

We have not been able to — we now have a plan to configure the meeting
structure so that the consultants can speak in the moment without coming with
too much homework. Damon Davis is going to play the role that Jim plays in this
meeting, setting the stage. These are the things that we have done. This is
what is on the horizon. This is what we are struggling with. We want to
re-energize the group so that we can give immediate, in the moment feedback. We
can take minutes and that can be a document. I am not sure that letters are
going to be the product of this committee as much as the dialogue, the
immediate feedback captured in the minutes, things that HHS then may or may not
want to do. I would say that is a prominent thing.

Today, we have a couple of presentations, SAMHSA, and I think the Health
Indicators Warehouse. We are going to see. Because again, I think sometimes we
jump to the apps before we understand what is the raw data. I have seen the
richness of expertise. I think Vickie is brilliantly positioned as the next
chair because she dwells in this world and she has a lot of understanding about
this. I feel very in a way relieved that you are taking the reins because I
think that is what is needed. I know a lot about meaningful use. I think what
your background is is going to energize the conversation and I think it is
going to be very exciting.

DR. GREEN: Vickie, any retorts?

DR. MAYS: Only that I am a little speechless because I have a lot in terms
of following Justine. I think that where the group is positioned is very
exciting in terms of what we have talked about in terms of next steps, having
Damon involved.

I also think for the way in which we are also including standards and
privacy and each of the groups as that is a topic and area that we want to make
sure. Again, it is not about the apps. I think the direction that we are going
is very exciting. I appreciate the confidence from Justine.

DR. GREEN: I want to speak to the committee. Justine, if we took a page from
Chris Fulcher’s comments earlier this morning about archetypes, you are your
own archetype. We are indebted to you for working with Jim Scanlon and Debbie
and others to invent this group, get it started, rummage around, figure out
your way, get to a product and you are now doing an elegant synthesis and
transition at the end. I just want to say thank you very much.


DR. GREEN: Do we have any public comments? Anyone on the phone? Debbie, can
you work some of your magic and suggest to us what are some of our next steps
are here?

DR. JACKSON: Justine, do you have any final ideas?

DR. CARR: Just thank you.

DR. GREEN: She has work to do.

DR. CARR: Yes, I have work to do. Truly, I said this last night. This has
been over a decade of my life. It has transformed my life. I have learned so
much about content wise as well as working together as a team and keeping our
eye on the mark. Each of you is very special to me. You can invite me back if
you want to hear about meaningful use. I am just processing now how integral
this has been to my life. I just hope that I have made a difference for the
country, for HHS, and for all of you. Thank you for all that you have done for


MR. SCANLON: I remember, Justine. It was ten years ago. You were right. We
were looking for new members. Several agencies said I know this doctor up in
Boston who knows a lot about standards and is interested in data and all of
this. We said all right. That is good. That is how Justine started and eight
years on the full committee and the last two years with the work group. From
the department’s perspective, I really want to thank you. You really pioneered
the work group and got it to where it is functioning. It is a nice hand off
now. I think we will make you an emeritus so we can always call on you again.
But thank you again from HHS.

DR. CARR: I just want to clarify because you may not think I know that much
about standards. The way this came to was we had a program that was an
inpatient oncology program, IL-2, that was being paid at a DRG rate for the
sight of disease when the treatment cost more than that. I was so wedded to
this program that I ended up coming down here with Nick – Kennedy’s guy –
on the way down, I said this is great. We are going to get this fixed. What do
you think my chances are? And they said zero to none. What? I took a day off
from work. I was my most passionate bordering on ravage self, to show them the
data and tell them this cannot happen. In fact, we got it fixed. But I hope it
was not overstated what my knowledge ICD was.

MR. SCANLON: They came from CMS and others.

MS. JACKSON: I saw you primed ready to respond and I had to get your notes
in. It is going to be a very busy summer for the committee. Just so you will
know, those of you who have been here a while, you know that usually we have
gone around the table and the subcommittee chairs at one point, now co-chairs,
were reporting out to each other what you were doing. That has not happened in
the last couple of meetings because everyone has been in the seats and sees
what is going on and is progressing forward. I think that is an amazing step
for just everyone pulling together and being on target.

The two action items that were taken from this meeting we will process next
week. As you know, they go to the Secretary first and then are posted. We will
work through the action item timing so that we can do a nice introduction to
the new secretary with briefing materials just as you will know what we are
about, our 60th anniversary and some of our recent reports. We will coordinate
that so there is a good communication as to who this group is while we are
giving her this amazing material.

Next month is going to be very busy with probably an Executive Subcommittee
call. We will pull that together in preparation for the September meeting. We
had the July 16 Privacy webinar session and that is going to be very hopefully,
technologically astute and pull things together for that material in
preparation for their upcoming meeting.

The action items for fall, and we will talk about this on executive
subcommittee, will probably be two items, one from Standards from their amazing
meeting from June 9. Pulling all of that together. They will have routine calls
in doing that. And then the Privacy material for the stewardship.

The information was about public comment and as Larry already requested,
comment, we are going to incorporate that more and more throughout the meeting
itself and some follow-up items with Jim. Those were the main items that I had
of the ACA Review Committee. We will follow up on that as well.

DR. GREEN: And our most immediate actions are the Working Groups is going to
meet this afternoon and the Populations committee, and everyone is invited to
join them tomorrow. Are you calling that a workshop? A framework workshop.
NCVHS week in Washington continues tomorrow.

If you don’t mind, I would like to run the table, but not to have everyone
say what they are thinking the work plan is, but I would just like to hear
everyone’s reactions to where we are at this point. Anything you want to say
again. Just a sentence or two. A year or so ago I asked you to do this in three
words. Ob, why don’t you start us off?

MS. JACKSON: Off we go.

DR. SOONTHORNSIMA: Tired. Excited. Looking ahead.

DR. SUAREZ: Busy, but excited at the same time about what is coming.

MS. KLOSS: More than three words if you will permit me. I thought that the
benefit of the committee working as a whole was really realized very well at
this meeting. I think it has matured. I just heard more of us pointing to a
need to connect this work with the other work. I thought that was terrific.

It occurred to me that maybe when we talked about a letter to the Secretary
that kind of introduces the committee. I thought it might be great to convey
that sense of congratulations about the impact that the liberation of data is
having in that letter because I think that was a sentiment that probably should
be communicated. Those were my two thoughts.

DR. FRANCIS: I thought this was, like Linda, probably the most participatory
and collaborative meeting in a long time. It was super informative for me. I
guess my take away is there is a lot to do.

DR. WALKER: Five words. Laser information people can use.

DR. GREEN: I thought you were going to say systems, systems, systems,
systems, systems.

DR. NICHOLS: Eager for October.

DR. VAUGHAN: Looking forward.

DR. FULCHER: I am still learning.

DR. CARR: Good luck.


MR. SCANLON: Easy for you to say.

DR. COHEN: I agree.

DR. CHANDERRAJ: Time zones and inclusion of practicing nurses and practicing
physicians in the committee beyond academic.

MS. BOOTHE: Introductory words from my perspective of where I sit and where
population health lives at CDC and maybe even HHS. Happy you are here.

MR. BURKE: Effective committee collaboration.

DR. MAYS: Excited. Very excited, but worried.

DR. STEAD: We are making real progress.

DR. WHITE: Still not Mike Fitzmaurice.

MR. SCANLON: Location, location, location. Three words. Thank you to the
committee again. It just amazes me that when you put smart people together, how
much you get more than the sum of the parts. Two, I thank you for going ahead
taking some of the responsibilities as well. Most productive committee. Again,
I want to thank all the members for their contributions.

DR. GREEN: I want to thank all of you for role modeling civil discourse.
Something the country needs desperately. I thought this was a spectacular
example of people who do not agree, but who care and being able to converse and
speak to one another. It was a lovely thing.

I may have inflicted these on you guys before. Rebecca Costa and her book,
The Watchman’s Rattle, quotes the famous scientist E.O. Wilson as expressing
dismay about the future of the human race. She quotes him as claiming that the
problem with the human race is that we have Paleolithic minds, medieval
institutions, and god-like technology. That quote has been going through my
mind over and over and over again for the last three days.

I have also may have inflicted with you something I have on my wall, which
is the stages of project development. You may have heard these before. You may
use them yourselves. The first stage is wild enthusiasm. The second one is
disillusionment. The third one is search for the guilty. The fourth is
punishment of the innocent. And the final stage is promotion of the

MR. SCANLON: We use that at HHS, too.

DR. GREEN: And when I look at Justine and that decade that she runs by, I
believe that you can see all the stages of project development across that
period of time at various stages. And in our different projects, we are at
different phases.

I want personal credit for adjourning the meeting two minutes ahead of time.
We will be back at one o’clock to commence. Thank you very much.

(Whereupon, the Full Committee adjourned.)