[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
SUBCOMMITTEES ON POPULATION HEALTH
PRIVACY, CONFIDENTIALITY AND SECURITY
THE COMMUNITY AS A LEARNING SYSTEM FOR HEALTH: USING LOCAL
DATA TO IMPROVE COMMUNITY HEALTH
FEBRUARY 8, 2011
St. Regis, Hotel
923 16th and K Streets, NW
Washington, DC 20006
CASET Associates, Ltd.
Fairfax, Virginia 22030
- Call to Order and Welcome/Introductions
- Broad Overview and Public Health-Driven Programs
- Moderator: Don Steinwachs, PhD
- Panel: Marguerite Ro, Mary Maddux-Gonzalez, David Patterson
- Focused Community Health Initiatives
- Moderator: Sallie Milam, JD
- Panel: Stephanie Browning, Janet Meredith, Debbi Main, Tracey Stewart,
Matthew Stefanak, Gwendolyn Flynn
- Health Information Exchanges (HIEs)
- Moderator: Leslie Pickering Francis, JD, PhD
- Panel: Patrick Gordon, Nance Shatzkin, Greg Larkin
- Discussion – Larry Green, MD
P R O C E E D I N G S
Agenda Item: Call to Order and Welcome/Introductions
DR. GREEN: For those of you from California and the West Coast, it is after
6 o’clock so it is time to start. My name is Larry Green. Sallie Milam and I
co-chair the Populations Subcommittee. Marc Overhage and Leslie Francis
co-chair the Privacy and Confidentiality Subcommittee of NCVHS and we all want
to welcome all of you here for what we think is a day of exploration and
conversation and we actually think it stands to be quite a lot of fun. We are
for fun. We are so grateful to so many of you who have traveled a long way and
gone to quite a bit of trouble to interact with us in preparation for today.
There is a famous line in the movie Pretty Woman about where she looks up
and says in case I forget to tell you, I had a really good time. I just want to
tell you I had a really good time today. Again, thank you very much.
We have some introductory business to do. What we would like to do is run
the table here within introductions. I would ask all the committee members to
identify yourself, your institution of reference, and whether or not you have
any conflicts of interest.
DR. STEINWACHS: Don Steinwachs, Johns Hopkins University and no conflicts,
member of the committee.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics,
DR. HORNBROOK: Mark Hornbrook, Kaiser Permanente, no conflicts, member of
MR. LAND: Garland Land, member of the committee, no conflicts.
DR. MIDDLETON: Good morning. Blackford Middleton, Partners Healthcare
System, member of the Full Committee and the Quality and Population
Subcommittee and no conflicts.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, member of the Committee,
DR. BREEN: Good morning. Nancy Breen, National Cancer Institute.
DR. MADDUX-GONZALEZ: Good morning. Mary Maddux-Gonzalez, public health
officer from Sonoma County, California, no conflicts.
DR. RO: Marguerite Ro, public health at Seattle-King County, invited guest.
DR. SONDIK: Ed. Sondik, National Center for Health Statistics.
DR. SUAREZ: I am Walter Suarez with Kaiser Permanente. I am a member of both
subcommittees, the privacy and population. Also I don’t have any conflicts.
DR. CHANDERRAY: Hi, Ron Chanderray from Las Vegas. I have not been oriented
so I don’t what ethics so I don’t know if I can answer the conflict.
DR. CARR: Actually today you are a guest and tomorrow you will be a member.
Justine Carr, Steward Health Care, chair of the Committee, no conflicts.
DR. OVERHAGE: Good morning. Marc Overhage, Regenstrief Institute, member of
the Committee, no conflicts.
MS. JAMISON: Missy Jamison, National Center for Health Statistics, committee
MS. KANAAN: Susan Kanaan, writer for the committee.
MS. GREENBERG: Marjorie Greenberg, National Center for Health Statistics,
CDC, and Executive Secretary to the Committee and I want to welcome all of you
here. We are delighted to have you.
DR. FRANCIS: Leslie Francis. I am from the University of Utah and I co-chair
Privacy, and I am a member of the Full Committee, and I don’t have any
MS. MILAM: Sallie Milam, West Virginia Health Care Authority, members of
both Privacy and Population, no conflicts.
DR. GREEN: I am Larry Green. I am member of the Quality and Population
Subcommittees, from the University of Colorado. I do have a conflict. One of
the presenters this afternoon is Partnership 2040, and I am the chairman of the
local board of directors in Denver.
Is there anyone on the phone, any presenter, staff, or committee members on
the phone that can introduce themselves?
MS. HORLICK: Gail Horlick, CDC Atlanta.
DR. PATTERSON: Dave Patterson, chief of health and demographics at South
Carolina Budget and Control Board, South Carolina HIT —
DR. GREEN: Anyone else? Marjorie, do we need to —
MS. SQUIRE: Marietta Squire, staff to the committee.
DR. MILLMAN: Mike Millman, Office of the Assistant Secretary for Planning
and Evaluation, HHS.
DR. BILHEIMER: Linda Bilheimer, National Center for Health Statistics.
MS. JONES: Katherine Jones, staff to the committee.
DR. COOPER: Leslie Cooper, National Center for Research and Resources.
DR. DOWNING: Good morning. Greg Downing, Office of the Secretary, HHS.
DR. SHAIKH: Good morning. Abdul Shaikh from the National Cancer Institute.
DR. LARKIN: Hi. Greg Larkin, state health commissioner from Indiana.
MS. MEREDITH: Janet Meredith, 2040 Partners for Health, Denver, Colorado.
MS. SHATZKIN: Nancy Shatzkin from the Bronx RHIO in the Bronx, New York.
MS. WELLIK: Mary Wellik, Olmstead County Public Health, Minnesota.
DR. MAIN: Debbie Main, University of Colorado, Denver.
MS. BROWNING: Stephanie Browning, director of public health and human
services, Columbia, Missouri.
MS. FLYNN: Gwendolyn Flynn, Community Health Councils, Los Angeles,
DR. GREEN: Okay, we are about to get started here. Just a couple of words
from me and I am going to pass this over to Sallie.
Today is the day to explore. Today is a day for committee members and staff
to ask questions, but today is a day for our guests to ask questions of each
other. It is not particularly a day for your best soliloquy unless of course
you are a presenter and that is what you came to say. That would be most
welcome. Think of it as a journey that we are starting this morning and that we
are going to take together for the day where you just want to find out
everything you can about what is going on out there.
Our goal that I want to remind us of that we started discussing two or three
months ago, is we want really to bring the voices of communities across this
country inside the beltway today. We want to hear what is going on out there
and finding data, producing data, putting the data together in a way that can
help a particular community. Let’s learn together today.
MS. MILAM: Just sort of building on what Larry said, we will have dedicated
time tomorrow in our subcommittee meetings to chat with one another, to
interject our own comments, but as Larry said, we are really almost looking for
a 95/5 percent split, 95 percent from our invited guests and 5 percent on our
side. With that I will pass it to Dr. Don Steinwachs.
Agenda Item: Broad Overview and Public Health-Driven
Programs, Don Steinwachs, Moderator
DR. STEINWACHS: It is a pleasure to be here with everyone. I wanted to note
that Todd Park has joined us. Do you want to introduce yourself? We asked
everyone else to, but we all know you probably.
MR. PARK: I am Todd Park, the chief technology officer of the US Department
of Health and Human Services and so delighted to be here today.
DR. STEINWACHS: It is great to have you here with us. Well, it is my
pleasure to moderate the first panel. We have four communities represented that
bring really quite a history of working with data around public health purposes
and applications. We are looking forward to learning from all four of you.
I thought I would briefly introduce you. You have introduced yourself, but I
will do it. And then ask each of you to give about a 15-minute presentation. My
biggest job is to try and keep us on time so when you get near the end of the
15 minutes, I may nudge you, and then open it up for questions from the
committee members and people around the table.
We are very fortunate to have with us Dr. Marguerite Ro, who is chief of
assessment, policy, development, and evaluation and the Public Health Authority
of King County, Washington. Most of us think of it as the Seattle area. She
will be talking about their initiatives.
Next we have Ms. Mary Wellik who is the public health director of Olmstead
County in Minnesota. I don’t know whether you are thankful being here. Is it
warmer here or not? Yes. Okay.
And then also another Mary, Dr. Mary Maddux-Gonzalez was the health officer
of Sonoma County which is probably warm and friendly this time of the year. You
probably wish you were back there and talking about what they have been doing
at Healthy Community Networks.
And then joining us by phone, we are very happy to have Dr. David Patterson
who is the chief of health and demographics, South Carolina Budget and Control
Board, Office of Research and Statistics and he is a state HIT coordinator.
David, it is a great pleasure to have you join us and be part of this panel.
Let me ask Marguerite to lead off and talk about the work in Washington.
DR. RO: Thank you. Good morning everybody. A little early for those of us
coming from the West Coast, but thanks very much for the opportunity to talk
As Dr. Steinwachs mentioned, I am the chief of the Assessment Policy
Development and Evaluation section at Public Health Seattle-King County. I will
say that I am relatively new in this position. I actually joined Public Health
Seattle-King County in November. And prior to that I was the deputy director at
a national health advocacy organization, and the reason I bring that up is
because it has given me really interesting perspectives on the use of data.
Coming from representing the Asian American and Pacific Islander community our
definition of community I think was a little bit different than how we define
community sometimes at Public Health Seattle-King County, but we can talk about
that as the presentation goes on.
In King County we serve approximately 1.8, 1.9 million people. We have 39
cities. We cover a geographical region that is about approximately the same
size of Delaware. We are the 10th largest health department in the
When this slide was made, we had 1900 employees and we had a budget of $296
million. That number as you might guess is changing daily in part because our
state legislature is in session trying to grapple with our state deficit.
At Public Health Seattle-King County our goal is to create the conditions
that maximize the number of healthy years lived by each person and to eliminate
health disparities. And the framework that we operate under is what you see in
this Venn diagram around promotion, protection and provision. And within each
of one of those functions we conduct assessment policy development and
Now in Seattle-King County besides being rainy and very green, we have some
interesting statistics. And what you see on the right is you see a map of a
segment of King County with different colors. And one of the things that I
could use that map for is to talk about the inequities in King County. Whether
we look at race ethnicity, whether we look at education, whether we look at
different measures of socioeconomic status or of health access, one of the
things that we would point out to you is that there are considerable inequities
in our region. Whether we are looking at infant mortality where African
American infants are twice as likely to die compared to White infants or
whether we look at risk factors such a tobacco use, we would see these great
inequities that face our population.
Overall when we look at aggregate statistics, we have somewhat of a
contradictory story that is being told and I think this is of the general
population as well as what is happening in our county.
Life expectancy for the most part — our aggregate data shows that life
expectancy is increasing, but with that comes the increase in the number of
unhealthy years lived.
I have three messages to share with the committee today. And the first is
that we need accurate, timely, locally relevant data. And that the reasons for
this data are many, but certainly coming from Public Health Seattle-King County
I would say that our goals of protection, provision, and promotion is as we at
the county have been thinking about that. Our data that we are looking for are
to help us drive those improvements and those improvements that we are seeking
are more increasingly through policy systems and environmental change. And to
do that we in the health department work with a number of different partners so
that our audience is for our data are many and we will come back to those.
But, again, if you think about who our audiences are, they are not only the
policymakers who run our governments, be it the county executive or the mayors,
but also looking at our community-based organizations, our health systems and
health providers in our county and also the foundations and other funding
agencies that help support the services and the work that we do in our county.
This is actually some recent data that we have with the new release of the
2005/2009 ACS data showing the level of poverty under 200 percent of poverty
level, the proportions of poverty. In the darker areas are where the greater
proportion of those living under 200 percent of poverty are located and then
lighter areas are those who are more affluent communities.
This data is shown to you by health planning area. The health planning areas
pretty much represent our city boundaries and this is most of our data that we
are able to disaggregate in terms of granularity of data. Health planning areas
tend to be where we have the most stable statistics.
It is not only important to obviously to see where we are currently, but
also to be able to measure trends. This map shows the change in poverty from
the 2000 Census to the 2005 and 2009 statistics that have just come out.
Now, one of the things that is also a little bit challenging though is you
may know from your community experience though is that the data of the health
planning areas is not sufficient to tell us what we need to know. And that what
we are always looking for is we are always looking for more granular data.
This is the data that has been broken out by census track. And what I would
say about that is then we can be able to begin to really look more closely at
what is happening at the neighborhood level. And as we use our mapping
capability in the health department and with our other departments in King
County government, so the departments of parks and recreation or natural
resources and department of transportation, these become the basis for what we
then overlay our data whether it be community health centers, whether we are
looking at parks and trails, where we are looking at waste facilities and so
forth to begin to define what our environment looks like.
One of the challenges I will just say from my previous position, again,
representing Asian Americans and Pacific Islanders, in Seattle-King County we
have a growing community of Pacific Islanders. They are just representative of
some of the emerging minority ethnic groups that I think many counties in many
localities around the nation are beginning to experience. The numbers are
really small. We often don’t have for communities that are defined not by
geography, but by race, ethnicity or some other measure; the data that are
easily reflected and can tell the story.
As your committees are talking about privacy, I think that is where some of
the issues come up when we run against these small populations who desperately
need the data to be able to demonstrate the disparities that exist or they are
just simply health experiences that they are having, how do we have make that
data available to them?
My second message to you is that we need to develop a standardized set of
community health indicators. And the reasons for this are many. One of these
obviously is to be able to compare how we are doing compared to other
communities across the nation in order to be able to compare ourselves with
what the national data are saying about community health or health access
measures more generally.
And when we think about what these indicators might consist of, one of the
things that we really need to be thinking about is really thinking about is
whether we are looking at downstream factors or whether we are looking at
upstream factors. And this graphic is really just to say that as we think about
health and we think about what are the actual determinants of health, how then
do we select those indicators that we think are most appropriate and that where
we can actually have data available and how can we track that over time.
In King County we have another layer that we consider and that we think
about and that is really from an economic, social, and justice perspective. In
King County we are really fortunate under the leadership previously of Ron
Sims, but has continued under our current leadership at the county Dow
Constantine and with the support of our Director Dr. David Fleming where this
issue of equity is of utmost important in all policies that the county passes.
The ESJI initiative, as we fondly refer to it, is really about saying that
health is in all matters and that health really has to do with where we live,
where we work, where we learn. It is these bodies and areas of equity that we
then need to track our progress towards.
One of the key products that we put out every 3 years through a partnership
between local foundations, the county and the health department is that we out
a report called Communities Count. And Communities Count really — this is data
that is available through the web. It is available through a hard copy report
that we publish every 3 years. It talks about the social and health indicators
in King County. And the data we collect crosses five domains: basic needs and
well being, positive development through life stages, safety and health, the
natural and built-in environment, arts and culture.
The data that feeds into Communities Count comes from the BRFSS, the YBRFSS.
We produce a survey called the Healthy Use Survey which generates additional
data. Together we roll up the available data that we have to support the
Communities Count report. That is just one of several different reports that we
publish both that are web based as well as hard copy.
This is some of the data that we present. What I would reiterate again to
you is that the data that we are using is important for multiple audiences in
part because our funders use it to really dictate where they place their
investments and they measure their progress by it. The community-based
organizations use this data to identify disparities and to make their advocacy
cases, and for King County and our city governments is to measure the
effectiveness of our programmatic work and our policy work.
Moving really quickly, to the third message. We need to align our resources
to achieve our population-based goals through effective community, appropriate
strategies, and interventions. The question is where are we putting our money.
If we know that the programmatic work that the public health work that we want
to support is really operating at the local level, at the community level to
have community tailored, culturally competent interventions then are we putting
our money to create the data and information that we need to measure, track,
and inform those community interventions.
Here is a quick schematic of our budget revenues. I think I am just going to
move really quickly to this graph. And this is to tell you what the status is
of what is happening at the county level. Over the past 3 years we have seen
tremendous reductions in support for public health infrastructure. And I am
really fortunate in that I have a director that really believes in the core
functions of assessment policy development and evaluation. My budget which is
about 1 percent of the overall public health budget has been fairly protected,
but I will say that we are very much impacted by the county and the state
deficits that are happening right now.
If we don’t have the data to be able to track what we can do, we really will
be at a loss and we will be acting blindly. I would love to talk to you more
about how the budget is impacting what we would do, but I will say is that we
are facing cuts in the BRFSS. We are facing cuts in terms of PRAMS and our
ability to over sample and really measure what is happening in our county and
the specific populations that we serve. Thank you very much.
DR. STEINWACHS: Thank you very much. We are now going to turn to Minnesota
and to Olmstead County. Mary Wellik, all yours.
DR. WELLIK: Thank you. Members of the committee, thank you for inviting me
to share my — I don’t have a slide presentation. I am going to basically tell
you a story about what has happened in our community. I have left a few copies
of these two documents on the table just to give you an idea of what is
developed in my community or our use based on the data that we have available.
But let me describe Olmstead County a little bit to you. We are the largest
county in Southeast Minnesota with a population of 145,000, quite a bit smaller
than the community you just heard about. Our industry includes health care
information technology, agriculture, and other sectors. Our largest employer is
Mayo Clinic. And there is growing diversity in our community, about 12 percent
at this point. The diversity includes many cultures from around the world and
really our challenge is that diverse diversity and the significant number of
primary immigrants and refugees in that group. We have a large service sector,
a growing population of disabled persons accessing health care in our
community, and an unusual mix of very sophisticated industry, modest community
resources and high expectations of the health department in the public sector
to do community analysis.
You have asked for my perspective on long-term goals for my community. I
would say a key thing is how we work together. In Minnesota, every community is
served by a local health department. We partner with the state health
department very effectively. And we really have valued the standardized way
that we all evaluate our public health problems through a community health
needs assessment locally. This is a requirement in our Minnesota law.
As a person responsible to lead this process, my goal for community health
improvement would be that that assessment would be coordinated with that of
other community partners like the United Way, the health systems, and others so
that we develop a common set of priorities, interventions, and outcomes which
we would collectively address.
These issues from a public health perspective range across a variety of
areas including maternal and child health, environmental health, healthy
communities, disease prevention and control, emergency preparedness, and health
care quality and access, kind of a Duke’s mixture of a lot of issues. That is a
lot of data to be on top of.
The story. In 2007, several organizations in Olmstead County developed a
common vision for health improvement through the formation of CHAC, Community
Healthcare Access Collaborative. We initially focused on improving health care
access. We compiled our collective community health needs assessment and white
paper to show who in our community received health care, who did not, what
factors affected access, and how social conditions of health are not really
integrated in our community interventions effectively or as effectively as they
It was clear that there was not comprehensive and adequate data to provide
an overall picture. What we did provide was a better assessment than had ever
been developed before. We provided and ultimately moved on to a focus from
health care service coordination to a focus on improving health for everyone in
our community through collaborative action.
A foundational element of our work is what we called a common determination
of issues of significance. How do we across all those sectors agree on what we
are going to work on and how do we demonstrate that this is really an issue of
significance. This entailed a joint effort to design the data to develop that
assessment leading to that plan of action and the metrics that would tell us of
progress barriers and outcomes.
The white paper describing that vision was taken to the leadership of 10 key
community organizations ranging from the local county government to the Mayo
Clinic. The partners were in agreement that without executive and board level
commitment to changing systems, policies, ensuring resources we could not
effectively impact the significance of these issues and sustain the results.
What resulted from that was a commitment on the part of those organizational
leaders to address a few issues and see if we really could make progress in a
The elements of the issues of significance included definition of the extent
and scope of the issue, the population affected, its magnitude, estimated
health care costs, and the fit with institutional priorities, the current
effect of interventions that were using, regulation, proposed metrics,
deliverables and timelines as well as factors. We have no framework for putting
this all together. We are designing it and developing it. Again, remember, we
are not a large community. We are 145,000 people. There isn’t local level data
that we can just put from various places to find these things necessarily.
By collectively describing an issue we became more attuned to the factors
outside of our own areas of intervention that are essential to problem
solution. Population health data is often difficult to compile at a local level
as I said and sometimes proxies need to be used. For instance, we often use
free and reduced lunch recipients as a proxy for low-income students.
You have asked me to talk about the factors that enable and impair
acquisition and use of necessary data. Well, I would say from my perspective,
the first enabler is this collaborative. It is very important that we work
across sectors in a way that has some common underpinning that allows us to
improve and ensure health care in our community.
Collective thinking, planning, and action mean that we don’t stop because
the dataset isn’t there or the framework isn’t there. Attitudes about sharing
data can enable or impair productive data use. Community health assessment is
not precise. It is not comprehensive. It does not hold all the answers.
Researchers, scientists, and others have to develop a level of comfort with
this. And believe me, that is a barrier. I find that they often feel we just
cannot proceed. For example, unmanaged mental health conditions in a community
lead to increased use of detox, jails, social services, and public health
services. It leads to inappropriate use of emergency rooms, ambulance services,
health care, and mental health care. Where is the one dataset where we are
going to put that all together and how are we going to share the information we
have with each other to compile what that really costs both in terms of dollars
and the human toll?
Other barriers to acquisition and use of necessary data. We lack the
resources to design data reports based on what we collectively can provide. We
lack the ability to mine datasets for information that is not set up for a
study purpose. And we lack the training for public health, nonprofit, health
system and other personnel to design a plan for use of data to solve problems
What would I need as a way to be able to accomplish the goals in my
community? First of all I need the will of policymakers; the will at the policy
level to focus on health in a way that we already focus on health care and
health care access. I think Dr. Ro has just really very ably described health
in a community as a much broader perspective than health care or health care
access. In order to use data to solve community problems, communities need an
infrastructure to do that just like health care systems do. Like health care
systems the data is there though it is in kind of a virtual system. We have not
yet invested in determining how to bring it together locally to create the
information that leads to the wisdom that informs our decisions about improving
health status in a whole community.
In local communities we need valid and reliable data resources that reflect
social conditions of health like poverty and livable wage. We need individual
and community progress toward control of health conditions beyond clinical
indicators of health status. For example, can we link diagnosis of diabetes
with measures for individuals like the availability and use of resources that
support health improvement, reliable and healthy food outlets, safe and
affordable physical activity outlets, affordable medications? Can we link that
for individuals to what a patient’s goal and demonstrated ability is to
accomplish them? We I babysit my grandchildren? Can I cook my own food? Can I
navigate in my own home?
Can we link the diagnosis of diabetes with measures for populations like the
number of diabetics in my community and the number following up on treatment
plans effectively and what the day-to-day barriers are to effective control?
Can we link it to specific community neighborhood and environmental
improvements that would include treatment of compliance?
We desperately need training of local public health personnel and
informatics; something that might equal the initiatives that we put across for
health care systems so that we have the expertise that helps us to do these
things. We need development of credentialed public health information systems,
staff education related to transforming data into information that informs
decisions about population health and application of evidence-based practice in
ways that are locally effective.
My single most important message to you is please foster and develop these
effective collaborations. Help us to see that there is a plethora of data out
there. How do we use it? We need it for identification and planning for
intervention related to community health priorities leading us to teach each
other locally how to share the work that we do to improve health.
The most important advance is not a particular kind of data. It is our
training in each of our sectors regarding how to collaborate, how to improve
our understanding of community issues, and how to respond effectively through
then sharing data and information. Through this process we will find those most
important datasets and information systems to collect and analyze information.
Along the way we need the resources, training, and ensuring of progress to make
a broad and lasting impact.
Each of our community organizations have missions that focus on what we hope
to accomplish through actions by our own organizations or agencies. When we
link these missions, we begin to see that our collective work will be much more
effective in reaching our goals. These goals typically are about improving
health, safety and well being of our community. Are you hearing a trend here?
It is what we all want to accomplish.
There is more than synergy here. Through collaboration there is the power to
identify and solve community problems that any one of us is unable to
accomplish. This naturally leads us to considering and seeking the facts in
data that we can compile to assist our inquiry. Collaboration is not acquired
In my community we have several community initiatives including a very
active, yearlong training program called Collaborative Leadership Fellows. This
effort develops specific collaborative skills and fosters partnership. We learn
about it just like you learn anything else and develop any other skill. As a
result, we tackle issues with the intent of developing capacity that leads to
For example, one of our CHAC collaborative issues of significance related to
improving performance for school aged children with asthma, a goal broader than
the agenda of schools, public health, or health care providers and one of vital
interest to affected families. We have not met our goal. It was too ambitious.
But by jointly assessing the problem we found a small percent of providers
actually used electronically available asthma action plans for children. We
found health care providers learned that their patients were not benefiting
from asthma action plans directed to the school setting. And schools identified
as significantly larger number of students who had asthma than we thought were
there. That was the by-product of our study.
As a result, we are changing attitudes. We are working on improving our
information systems and we are reaching the tops of the school health care
provider and public health agencies and working with each other.
We have one important outcome has been the development of a single HIPAA,
RIPA, State of Minnesota release of information form. It allows parents with
one consenting action empower health care providers, public health and schools
to work seamlessly in the community setting. It also led to the groundwork for
inclusion in the Southeast Minnesota Beacon Communities Project enabling
ongoing collaboration around this particular demonstration project with Beacon.
Again, thank you for inviting me to present to you today. I hope my
description and examples of collective power of collaboration, best use of
existing data, and efforts to establish more useful data and information will
help you to see that we in local communities cannot only find solutions to our
public health problems, but with your help we can establish ways in which
effective solutions can be replicated to conserve resources and broaden success
in other communities as well.
I have left my email and website information on the handout that you have.
If you would like to look further into some of the material that we have that
uses local data. Thank you.
DR. STEINWACHS: Mary, thank you very much. Moving now to Sonoma County and
DR. MADDUX-GONZALEZ: Well, thank you very much for this opportunity to
discuss some of our efforts in Sonoma County for community health improvement
and our use of data.
In Sonoma County in 2008 had the opportunity thanks to the political will of
our local board of supervisors to improve community health to convene a group
of 32 community leaders from various sectors including the usual suspects from
health care and health plus business, plus education, unions, community-based
organizations, really a broad based group to look both at our health care
delivery system prompted actually by the potential closure of a hospital, but
fortunately thanks to the support of our board we were able to expand this to
also include community health.
This group is called Health Action and we engage stakeholders in a community
dialogue on health to build a collective understanding of our local health
issues and potential solutions and create a share vision of community health
that was based on the determinants of health. We have developed and implemented
collaborative recommendations based on this whole process.
It was a great opportunity. I have been a health officer for 11 years. It is
the first time I had really access to a wide range of community leaders to
discuss in depth indicators, data about our local health and health care
delivery system. We included social determinants of health. We looked at issues
of our health care delivery system.
And through this extensive analysis or extensive given in a community level
we developed 10 goals which we call are part of our 20/20 vision. Our 20/20
vision is that by 2020 Sonoma County will be the healthiest county in
California. These goals are both social determinants and other types of
determinants of health including, I won’t go through all of them, but high
school graduation, improved economic resources for families, the connectedness
of our citizens to our community, access to healthy foods particularly for
low-income individuals, increased physical activity, reducing substance use,
mental health improvement, injury prevention, and access to health care.
We established for each of the goals and indicators that we are tracking and
will be tracking from now and through 2020 and these are just a few examples:
high school graduation rates, smoking, and obesity. We are also looking at
these indicators for the general population and also for parts of our community
individuals in our community particularly disproportionately impacted such as
here are the higher rates of obesity among our Latino population.
Our priorities for action. We decided we needed to really focus our initial
efforts in three areas because of their potential impact on obesity reduction
and chronic disease reduction. And this is the three areas are access to
healthy foods, increasing physical activity, and linking every resident with a
primary care medical home.
These are a few of our projects that are related to each of these with the
access to healthy food. We have established a food system alliance that brings
together all the producers of food and the distribution of food and
particularly focus on accessing food such as healthy food outlets in our
community: iGROW which are community gardens, iWALK are safe routes to school.
We have a patient-centered medical home learning collaborative that involves
all sectors of our medical community: Kaiser, individual physicians, community
clinics talking about how to improve primary care. And we recently completed a
primary care capacity assessment.
At the same time we were going through this process we also learned of a
very exciting customized, but web-based information system that has been
developed by the Healthy Communities Initiative and it was in use in San
Francisco in Marin County which are very close to Sonoma County. This provides
based on change theory access to community dashboard of indicators, promising
practices, collaboration centers, and the possibility of evaluation and
tracking of indicators.
With this it has been very important. It has really allowed us to have a
living community needs assessment that provides updated data for our community
and this has been very helpful in the health action strategic planning process,
but also in other areas. For example, with our local hospitals actually help
fund this because they are required by California law to do periodic health
assessments and community benefit programs. We also feel this would be very
helpful in the public health accreditation process that is currently underway.
The promising practices have been very valuable. We use many of these in the
development of our strategies for health action and the evaluation and tracking
is both for Healthy People 2020 goal indicators, targets and also for our
locally defined goals through health action.
This is what you would see if you opened up our website for Healthy Sonoma
with the community snapshot and on your left you will see there is a large menu
of available resources. And if you drill down, I am going to show you what you
would see if you drill down further into the community snapshot and look, for
example, at exercise, nutrition and weight you see various indicators. You also
see how Sonoma County is doing relative to the State of California. Drilling
down further if you want to see obesity, you can see the number of individuals
who are obese in our community and the source of that. In California, we use
the California Health Interview Survey Data very frequently. And we can drill
down further and by age, gender, race and ethnicity.
We can also look at how we compare to the rest of California in terms of
obesity and diabetes prevalence and how we are doing in 2010 or 2020 targets.
We can look at the promising practices or evidence-based practices for
addressing obesity in a community and use the 2020 tracking that is now up and
available and how we are doing relative to the 2020 targets. And then as I
mentioned, we also are tracking our specific goals that we have set for health
The website Healthy Sonoma also links to our Health Action website as you
can see up in the right corner also links back to Healthy Sonoma. It links both
to our Health Action and the specific iGROW and iWALK and we soon will have
iCARE which is our primary care website also. That is Healthy Sonoma.
We were very happy with Healthy Sonoma and then we were approached by the
Network of Care by Trilogy and through a very innovative collaboration with the
National Association of Counties and the National Association of City and
County Health Officials to be a pilot for a hybrid website which is called The
Network of Care for Healthy Communities. And it combines what I have just
discussed with you those elements of Healthy Sonoma that was developed by the
Healthy Communities Institute together with the resources that are available
through Network of Care. You will see that on the right-hand side these are the
elements I have just discussed that are available through Healthy Communities
Institute. And in addition the Network of Care provides resources such as
service directories, libraries, personal health records which are great for all
of our local community members. It is really those resources available for
individuals are made available in a common website that combines with the
information available, the data through the Healthy Communities Network. This
really combines the two.
In Sonoma County, we are really an exception. We actually have both
websites, but we more commonly use the Healthy Sonoma so that it is not
confusing for individuals. But a community could have either one or the other.
We found this very effective because it really goes to community health
principal factors. We need the combination both of the work of policymakers and
stakeholders with a top-down policy environment and I am echoing many of the
themes from the two prior presentations. We need environmental policy
interventions to change these very substantive health issues that we are
dealing with. But we also need the opportunity for the community to have
knowledge and become empowered to make the personal health decisions. And only
with everybody having good information can these good decisions be made.
We are very excited about Health Action. It really has allowed us a vehicle
to really convene stakeholders to look at health issues, but it has really
broadened beyond that. We have seen what we need for our community to prosper
both health and beyond health. It has allowed us to — I was so glad Marguerite
provided the upstream chart. We have a group now that is looking at the issues
of our county across multiple areas: economic development, educational
development. And looking at what we can do to invest our money differently so
that we are not building prisons with our scarce resources. Instead we are
investing in upstream investments. It has really been very effective in
convening these community members.
The way we look at it there were certain pieces in place and we brought
together the stakeholders to put additional pieces in place through Health
Action which provides a big tent, a way to frame a process for improving
community health. We have established our initial three priorities. We have
built certain key programs, but there are other key programs within our
community that also are fundamental to this process.
Looking at our community and clinic expansion, multiple areas, and Healthy
by Design redesigning our communities to reduce chronic disease. All of these
come together and together we are all building a healthy community.
In conclusion I did want to — I really appreciated the questions that you
have asked. What do we need as local health departments to accomplish our
community health goals? We need to make health data more accessible to both our
local health departments in our communities. And we found that platforms such
as Healthy Sonoma, Network of Care for Healthy Communities are very effective
in doing this.
But I also think there is a tremendous opportunity and that is we need
technical and financial support to take advantage of this incredible
opportunity that inclusion of public health and electronic health record
meaningful use criteria. This is something that will provide local health
departments with access to data on a level that we have never had before, a
level of granularity. How quickly we can have access to this information is
just we have not had this opportunity before.
There have been billions of dollars invested through the HITECH Act
appropriations. We have benefited locally from some of the regional extensions
that are really helping our local physicians, our community health centers, get
better understanding, access the technology necessary. However, we have not had
that type of support for local health departments to be in a position to
receive this electronic health information to have true health information
exchange and that is something we would greatly appreciate your help with. It
is, as you know, the way that this meaningful use criteria is written
physicians, medical groups and hospitals need to make a good faith effort to
provide data locally to their health departments. If health departments are not
in a position to receive that, that door closes and we lose a tremendous
opportunity. It is an area I really hope that you will look into further.
And just in conclusion and to echo the two prior presenters, we need access
in local communities to data to assess, plan, communicate, very importantly
also to motivate, track, align, and sustain efforts to improve community
health. Thank you very much.
DR. STEINWACHS: Thank you very much, Mary. Now we are moving to South
Carolina. David, we would love to hear from you. David is on the phone joining
DR. PATTERSON: I am here as disembodied voice and I hope that you can hear
me. I have my speaker down to try to control the feedback. First let me say
being from South Carolina I am something of an evangelist about all of these
issues in terms of integrating data, going beyond specific domains, looking at
problems as a whole person, and engaging communities from the community level
all the way up through the state and it is very heartening to hear the comments
from the previous presenters and I think that I am going to echo quite a few of
them at the state level.
Let me begin by telling you a little bit about the organization that I
represent because I think that context may be important to this group. I am
with the South Carolina Budget and Control Board which is a neutral service
agency. I am a neutral entity within a neutral service agency. In fact we like
to call ourselves the data Switzerland, a place where others can contribute
data, have those data integrated and be involved in again dissemination and use
of that data that ensures that their own privacy needs are protected, the needs
of their clients are protected, but the data can be used to fit a certain use
In our case my entity is a staff of nearly 40. About half of them are
statisticians with substantive expertise in particular domains. The remainder
are either administration, IT staff, database administrators and software
developers because I think as you go forward and we have heard this several
times already in the presentations, it is important to provide both the ability
to do deeper analysis, but also to provide tool sets so that communities can
access the data, the public can access the data all in appropriate and
controlled fashion. That is really who I am in my official state government
hat. In addition, as I was introduced, I am also a state HIT coordinator.
My office is also providing the quorum for structure for a state health
information exchange and I will talk a little bit about that because I see an
emerging theme where integration for analytic purposes and essentially
integration to support clinical care are increasingly merging together and I
would like to tell you a little bit about how we do that in South Carolina.
The core of our operations is really around an integrated data system or
statewide enterprise data warehouse and I have provided the group with some
handouts on our scope of services and what we actually look like. We have
developed that over a number of decades. It is built primarily out of
administrative data and it is tremendously comprehensive covering areas from
law enforcement, juvenile justice, public safety, corrections, et cetera
through the social services domain including not only economic services, but
also childcare quality improvement and childcare center licensing as well as
child protective services, foster care systems, et cetera.
Then within the health area of course we have tremendous data holdings
ranging from of course all Medicaid claims in the state, the entire state
employee health insurance system, and all uniform billing hospital data. That
is all payer data submitted as administrative claims from every hospital in the
state. That includes all in-patient encounters, all emergency department
encounters, and all out-patient surgeries. We also have data on health
professions, the entire department of mental health administrative system. From
the health department we have vital records and so on that we integrate. We
provide support for various disease registries as well as specially focused
agencies that provide assistance not only in health, but also women’s support
services like vocational rehab.
And as I said, our function under state law is essentially to act as a
neutral clearinghouse. We don’t own these data. We essentially integrate them
on behalf of all of our clients and customers and they include not only state
agencies, but nonprofit organizations like the South Carolina Primary Care
Association, the South Carolina Office of Rural Health, the Free Medical Clinic
Association, and the Hospital Association. And we do that to support them under
rules that are external to us with the data owners essentially making decisions
about the level and nature of release.
We integrate the data statistically so that we don’t have to create a single
state identifier. I am sure you know South Carolina is very resistant to any
sorts of government intrusion. We have managed to do this statistically and
then pull the identifiers off so that the data can be made to be useful.
The data themselves exist over a tremendous longitudinal breadth typically
decades back to the early and mid-90s in most cases. And they are granular to
below the person level actually down to the person event line item in many of
these cases which makes them a very rich resource not only for guiding programs
and making policy decisions, but also a support for researchers and as you will
see when I get into the HIE portion of this that linking gives you the ability
to build a master patient index and sort of a centerpiece of a federated
architecture that is intended to move data for treatment, quality reporting,
and public health purposes statewide. And again public health agency in South
Carolina has always been one of our major partners.
Now the philosophy in doing this I think really echoes some of the other
presentations. We view all of these things as a whole person problem and in
fact if you look at the documentation that we sent out, there is a circle that
represents our data holdings. We call that the circle of love although
technically sometimes it is also the wheel of pain. But the idea behind that is
essentially to allow various communities, researchers, program and policy
officials a chance to look at a person as a whole person because in many
instances we become sort of siloed even in our thinking around a particular
domain because that is just a function like public health and their data
resources that are traditionally associated with public health when the reality
of human problems are much, much broader than that.
That is the centerpiece of what we have. We use that to support a number of
projects. Again, quite a few of those are listed in the material that I handed
out, but just to give an example of how we can drill in the communities. We
have done everything from supporting child care, quality improvement
initiatives with the department of social services. I believe there may be
slide of that or copies of that available to you. We have always supported the
integration of data among free clinics, community health centers, and rural
health clinics. Designed a guide of panels of providers so that they can work
on sharing common patient populations particularly among the uninsured and
figure out how to deal with those issues in different localities because in
some circumstances in context the relationships may be a little competitive
primarily driven by reimbursement issues. In other cases they may not and so
just giving them a chance to develop their own local collaboration.
Also giving them a chance the providers and physicians from these different
settings a chance to work together around common data to come up with a common
standard of care. One of the objectives in this particular sub-project was the
development of a common standard of care for diabetics and hypertensives in
that population. And, again, that was driven by clinicians.
Other projects. We do quality improvement work for the hospital association,
even providing dashboards for individual hospitals that allow them to look at
where they stand on about 50 different indicators including the AHRQ of quality
indicators and for anonymously providing benchmarks either with hospitals of
similar size, patient mixes, or geographically continuous hospitals. We also
provide a lot of support for public health in enhancing their reporting and
ensuring that their various registries and data collection systems are
Stemming from that warehouse now in the environment under ARRA which I think
has already been brought up. We have engaged in a number of projects that
positioned these data holdings to be essentially the core component of a
statewide health information exchange. In South Carolina, it is approaching the
notion of HIE probably in a somewhat different way than many other states
because we do have a statewide infrastructure. The intention here is to be
statewide in scope from the very beginning. We follow off also the government
and privacy protection that extended from the warehouse. It is a very natural
thing for us to have a public utility with external multi-stakeholder
governance but also represents consumers as well as public health and various
In that service bus the intention is to eventually close the loop between
clinical care, public health reporting and decision making and eventually even
basic research. The idea is that in addition to the data assets that the
warehouse has the ability to link these records can move data that really
resides permanently at its particular location, but move it where it is needed
as needed under a specifically defined set of rules.
One of the first cases that we took on of course and I am happy to say that
we have already achieved this is integration of the various EMRs and facilities
that connect to the statewide health information exchange to the immunization
registry at our public health department which is called the Department of
Health and Environmental Control in South Carolina. And that allows of course
the improvement of the immunization registry, but it also allows us to ensure
that immunization history from the authoritative source is available not only
to local public health offices, but also to providers that are treating
patients. And I think eventually we would like to see some limited integration
of that even with our public school system to eliminate the paper chase over
immunization histories and vaccination histories as children enter schools.
Again, the idea behind everything that we do is integration to serve
multiple needs for multiple use cases all within a framework of trust and
collaboration that has been built up over a number of years. And that kind of
leads to the themes that we were asked to discuss. I am going to reiterate some
things that have already been said, but I am going to reiterate them maybe at a
state level with an enterprise approach.
The first is obviously integrating information. And in order to do that you
really need to build on what you have. In South Carolina we have been very,
very successful moving out from administrative data now to the ability to at
least manage, move, and appropriately use some clinical measures as well that
are collected. But you have to build on what you have.
Now having said that and of course it should be no surprise, but a state
official is going to be an advocate of incrementalism. But you take those steps
with a goal in mind and the goal is really a more holistic view of the very
human problems that all patient citizen clients face. Because if we don’t get a
better picture somehow, I don’t think that we can adequately meet their needs.
To sort of take off from a quote by Judge Learned Hand, move by what light you
have. You need some light to move at all.
The second point is collaboration. Again, I am so heartened to hear this
theme come up repeatedly. And that collaboration has to exist at multiple
levels and eventually engaging patients I think themselves in their own
collaboration around their own care. But this is not just about amassing data.
It is about generating appropriate questions, new questions, in other words,
what can we improve, what are the problems, and how do we manage them. And also
those collaborations naturally extend into issues relating to action. How do
you intervene? What do you actually want to do? I don’t think there is any
single entity anywhere that has a monopoly on that information. I think, again,
collaborative teams from multiple perspectives, working multiple dimensions of
these problems are really likely to be the most effective way to get where we
want to go in terms of solving problems for the people we all serve.
A fourth thing that I want to mention is protect and I want to say that on a
number of dimensions. Protect privacy, protect patient’s privacy. The
collaborations that you build at the end of the day really hinge on trust and
the receptiveness of the patient and citizen community also hinges on trust.
And that is one of the things that we have done all along and why in many ways
my organization exists is because it is neutral and it is committed to
protecting the rights of the data owners, the rights of the citizens
represented in those data sources. That is absolutely key to build a framework
for trust, obey that rules, and that actually supports the whole notion of
collaboration. Trust will continue to expand over time and you will find
yourself able to do more and more useful things with the information that you
I think the last piece that I would mention is the need to close the loop.
Again, as we move from an era where data was integrated in some giant warehouse
and I of course managed one of the large state warehouses. But it is moving now
more towards a federated model where data exists, lives where it lives, and is
moved. But it can be moved according to and in support of a number of use
And when I say close the loop that essentially means the same information
that is used for treatment and patient management can also drive quality
improvement initiatives and can also drive public health reporting. And
ultimately basic research which all of the above help to generate further
improvements in clinical practice which can then be fed back to the clinicians
that actually treat patients. At the risk of sounding like a management walk
what you really want to set up is a continuous improvement feedback loop
between the use of the data for practical purposes, treatment, and so on, and
the use of the data to support various forms of analytics that in turn can be
fed back and improve the treatment itself.
Again, I thank you for allowing me to be a disembodied voice. I look forward
to your questions. If I can be of further assistance to any of you, contact
information is provided on the material that I submitted to the group. Thank
DR. STEINWACHS: David, thank you very much. We are now going to open it up
for questions from members of the committee and from the panel members to each
other, too. Who would like to go first? Please, Leslie.
DR. FRANCIS: Well, I actually have a number of questions and I will just
state them quickly and ask you to weigh in on any one of them. I would say that
this comes — as co-chair of the Privacy and Confidentiality and Security
Subcommittee, my goal here is just to begin to think about what we might do to
help you. Quick notes. First of all we thought about a letter on HIPAA and
FERPA. And I wonder if any of you think that the Minnesota model has solved the
HIPAA/FERPA disconnect problem or whether there could be any help along that
line as information comes from schools and gets integrated.
DR. STEINWACHS: Leslie, I think there are some responses. Why don’t we try
and take them one by one but quickly.
DR. FRANCIS: Sure.
MS. WELLIK: If I could just comment quickly. We technically solve the
problem. Our legal advisors tell us that we have a single consent form that
will work. The barrier typically is to get the users to understand that this is
legitimate and change their practices. I think the Minnesota model — we are in
a demonstration project. It is very slow to get, for instance, schools to work
with us on a project where their primary focus is academics and we are asking
them to look at improving outcomes for asthmatic children so that they can be
better learners. They want to. They don’t have the resources to. One of those
steps is changing their consent model. It is just complicated to get a lot of
different systems to change something like that.
I think the step of seeing if this is a fit, if it is a model that will work
consistently in other settings and then to really take a hard look at how do we
get the word out to organizations and entities that you can now use this one
consent form to start exchanging information because once we understand that we
will take some major leaps in starting to work with each other on data
DR. FRANCIS: More information about that might be really helpful sharing
A second quick one. Any experience with thinking more about drilling down to
small groups and I will just link that to is the Switzerland model the way to
do it. That is a particular model that we just heard from South Carolina.
DR. STEINWACHS: Any comments from the panel or David about does the privacy
concerns of drilling down to very small groups? Is that helped by having the
DR. MADDUX-GONZALEZ: I think it is something we currently do on an
individual case almost. We are very aware of this concern when we are looking
at our local data and presenting it. Going back to this potential for health
information exchange with electronic health records I think the level of data
that we need to do are assessment and to use in planning for place-based
strategies for tracking health improvement really is at an aggregate level that
I think we can put adequate safeguards in place for confidentiality and still
be very effective.
DR. PATTERSON: That has been our experience as well and that is part of why
we are such an advocate of making sure that the data owners each of whom has
their own regulatory and legal context in mind are at the table and insisting
to make those decisions. The technical side of any operation like that can
always achieve the level of aggregation that is specified. The question is how
do you specify it. But typically what we do is work through, again, as kind of
a minimum necessary principle. And it also varies by who the data consumer is
to be quite frank because at different agencies and organizations are providing
the data and can legitimately share at its full granularity with each other.
That does in some case happen. But I think the basic rules are that is the
minimum necessary. All aggregations go up to avoid things like trail
identification at the small group level and there is also great sensitivity of
course to the frequencies in different cell sizes. There is a process to that
and I think that is the important thing.
DR. RO: The only thing that I would add to my colleagues is that we receive
about 300 external data requests per year from our community. And that is where
we get the request for that kind of granularity of data.
I think one of our bigger challenges is helping those users understand the
limitations of that data. To the degree that we spend a good portion of our
technical assistance is explaining the limitations.
DR. STEINWACHS: Let me go to Justine and then Sally.
DR. CARR: Thank you, really, for just very inspiring presentations. I think
it has really achieved everything we had hoped to learn and more.
I have a question about the workforce that is needed for data manipulation.
As we move from static reports, census reports, and we saw some nice examples
today of overlay of geographic data with economic data and so on. Who is it
that does this? And as a corollary to that, is there or have you found a role
for data that is more directional? Let’s say Google inquiries about flu and
things like that. I put that out to any or all of you.
DR. RO: I will start. This is Marguerite. This is for the phone people. This
is Marguerite of Public Health Seattle-King County. I have a staff that is made
up of epidemiologists and social scientists who are supported then. We have a
GIS center out of department of natural resources that we work very closely
with to do some of the GIS mapping. In terms of staffing I think that if I
could have a health economist, a health services researcher on board, I would
be really thrilled. I have just brought on a JD who can do some of the policy
analysis for us.
In terms of looking at staffing needs, one of the things that I would say is
that we are constantly in search of training. As we move into looking at
systems analyses, spatial analyses, we need more support for our staff to be
able to attend that. Not all of my staff are doctoral level staff. When NIH,
for instance, has trainings that are designed primarily for only doctoral level
or academic researchers, it is very difficult then for my staff to be able to
participate. We don’t have extra funds to support that training.
The second piece you asked was about using other datasets. I will say that
in our Healthy Eating Active Living we are looking at actually other datasets
that can help us understand both the policy environments and understand data
that we don’t necessarily collect. For instance, looking at where food
establishments are in all of our communities and we can do that by looking at
marketing surveys and so forth in addition to looking at some of the Google
mapping that you have already identified.
MS. WELLIK: You have heard from a large county. Now, I will talk to you
about what happens in a small county. In Minnesota, we have 53 local health
departments serving every community in Minnesota, but they are all under local
jurisdiction and we operate under an agreed upon state model of community
health needs assessment and try to work particularly on healthy communities
kinds of projects together.
Minnesota is the only state, I believe, that has a statewide health
improvement program which is money that has been distributed in competitive
grants through local health departments to communities. And this grant, again,
covers all of Minnesota in 40 grants. A number of the health departments went
together in order to be able to be large enough geographically to offer this.
One of the things that has been very helpful to us is that the state health
department does training for local public health staff on assessment and
evaluation, and also helps us to find contractors that can do some of the more
in-depth evaluation that we are not staffed to do.
But at a very basic level I participate in the NAACHO informatics workgroup.
There are 3000 local health departments across the United States. The majority
of them are too small to even have an epidemiologist. We need to find ways for
local public health to either partner with state health departments or with
each other in order to use data effectively and have the kind of support to get
our programs to work effectively.
In my department, a 12-million-dollar budget, about 120 staff, I rely very
definitely on agencies in my community health systems. There are two health
systems: Mayo Clinic and Olmstead Medical Center. The United Way, partnerships
with the state health department, partnerships with other entities that bring
to us some of the expertise we need in informatics. We can’t partner with those
folks if we don’t have basic training and informatics ourselves. We have been
very aggressive about that in my department. And I see that that is a gap that
really exists across the country.
One of my staff is participating right now in a review of curriculum in
informatics for local public health departments through the Public Health
Informatics Institute. We are very hopeful that this will be a curriculum that
will start to rule out increased capacity in local health departments across
DR. MADDUX-GONZALEZ: I just want to briefly say that one thing that has been
very helpful to us is through this planning process we really realize that
across sectors of different areas of community we are looking at much of the
same data. We are working more closely, for example, with our human services
partners to look at — we are doing the same GIS mapping because for their
goals they are also looking at issues of poverty. Law enforcement is looking at
that for reducing incarceration rates. We are looking at high school graduation
rates. That is a very helpful combining our resources.
But I do think that the platform such as Healthy Sonoma and Network of Care
for Healthy Communities is particularly helpful for small jurisdictions that
may not have an epidemiologist. Because the data from surveys is kind of
centrally they then populate these individual customized websites, it is an
important vehicle for smaller jurisdictions to have access to data.
DR. PATTERSON: Yes, I was just going to chime in again, same themes,
different levels. My internal staff is primarily masters’ level folks, some
bachelor’s as I mentioned in my presentation. About half to two-thirds of them
have some sort of statistical/quantitative credentials and generally around the
substantive area. We have a couple of in-house epi people and bio staff people,
in-house social scientists, even an economist. And, again, they are primarily
to analyze the data and also quite frankly because they have some subject
matter expertise to provide the kind of connections that my colleagues on the
line mentioned to other external resources.
We do a lot of work with the University of South Carolina, Medical
University of South Carolina, Clemson University, and also with other
universities and even national research shops and so on. And having at least a
minimal level skill set in house I think is very important to be able to
appropriately engage these external resources. It ends up becoming what I like
to call a forced multiplier because you can have a dialogue with them and then
take advantage of the knowledge that they are bringing to the table.
The other thing that we do that may be a little different is we actually
have some in-house computer scientists or software engineers and database
administrators because one of the things that we are finding is we transitioned
over time. When I first began working here of course, most of our work was
around generating ad hoc reports and doing data analysis and also providing
support for larger research projects. And increasingly we are moving towards
actually trying to deploy the data together with meta-data so that the public
can take more advantage of it, small areas can take more advantage of it,
researchers can use some of these analytic tools like, for example, data cubes,
on-line and analytic processing as an aoristic to move them to more advanced
And increasingly we have integrated that with GIS. One of the things that I
have neglected to emphasize in my presentation is we not only have an in-house
GIS analyst and have in-house GIS capability, but our sister sections in ORS
actually do the E911 system for the state and the state geodetic survey. We
have very tight integration with GIS and we have also moved that integration
into our toolsets. Some of the material that you may have there gives you an
example of how you can integrate mapping both with point locations and even
geospatial correlations with tabular data.
And part of the reason that we have emphasized those skill sets as well is
that it really multiplies the uses that you can get out of the data and it
spurs others to take advantage of and/or ask more questions of it. It also
helps to ensure quality and quality improvement quite frankly because the more
educated eyes you have looking at data the more questions are raised about its
meaning, its validity, its reliability even though we have a pretty robust
quality improvement process. With large data systems there are always errors
and questions that come out and again putting the data to work helps to answer
MS. MILAM: Several of you have spoken to collecting administrative and
clinical micro-data at the record level. And you have also spoken to properly
releasing it with the use of various statistical disclosure limitation
techniques such as aggregation or cell size suppression and that work is
I am wondering if each of you uses statisticians or whether some sort of
best practice has evolved or if there is a framework that you could point to
and how you have found its usefulness.
DR. MADDUX-GONZALEZ: In Sonoma County we are roughly a half million
individuals. We are very fortunate. We have two epidemiologists and they are
involved in any analyses that would be released to the public or any data that
we have any concerns about revealing personal identity. It is done pretty much
on an individual case-by-case basis.
With this new opportunity for electronic health record data in addition to
the three areas: immunization records, syndromic surveillance, and lab
reporting that are explicitly identified, but these others of body mass index,
of blood pressure, issues of disparities. Those issues we will really welcome
guidance on how we go about doing that because that will be access to data on a
level that we hopefully if we are able to put in place the appropriate process
and infrastructure for doing that will have data available on a level that we
have not had previously so we would welcome guidance from you on that.
MS. WELLIK: I will use as an example a project we have been doing on very
low-income, single adults, general assistance medical care recipients who lost
all funding for health care a year ago in Minnesota. That has fortunately been
reinstated. But in that year’s time the health department and the Mayo Clinic
started to meet and the other health system and medical center joined us
eventually to determine what it was going to cost us in money and in health to
let this evolve this group of people being uninsured and a large group of
people was people that don’t use the health system well.
The process of trying to find out who these people are, what it costs the
government, what it costs the health systems and what we could do with those
resources to delivery health care differently to this group was almost
insurmountable. The only reason we continued to work together on this is that
we had previous relationships that said if we could break through this barrier
we would start to really find a way to develop a very robust health care home’s
model for people who needed who don’t fit the system well. We had to go into
the state database and draw out information. It took a lot of work. They really
weren’t set up to do that. They had the information, the data, but for our one
county to get that information took quite a bit of time. I don’t know how much
work it took, but it took quite a bit of time.
Then we had to determine if those people were in, for instance, the Mayo
health system patient database. There really was no way to do this other than
to get release of information from a subgroup of that group, go to Mayo, give
that release of information to them, have them query whether they had patient
health records, and then start to put together what does it cost us on the
county side for social and public health services, what does it cost on the
Mayo side for medical services, and then analyze that.
And what we found of course was that the county was dealing with, we knew
this, huge implications for uncontrolled mental health concerns with these
patients. And on the health system side they found that these people with
particularly those with mental health concerns were inappropriately using
health care services in a very expensive way. The fact that we could control
their mental health needs would probably impact both social services, public
health, and then the medical care cost, as well. Very cumbersome.
Actually by the time we have gotten to that point the state decided to
change its policy back. We are actually slower than the state legislature on
this one, but we will continue to work on it because it is an important issue.
Finding a way for us to share information. That is just a small example, but
there are many of those that challenge us.
DR. PATTERSON: Well, the South Carolina response there is that where health
information exchange is concerned we first directed a process. Now of course
the ONC required that, but we were already moving in that direction anyway.
Using multi stakeholder governance to set the use cases and to consider
additional use cases later down the road. Again, emphasizing our theme of
incrementalism we really began with the same kinds of use cases that are
supported by the nationwide health information network and by the CMS final
rules on meaningful use and ONC’s rules on standards which is really first
beginning with the treatment use case, again, allowing patients the opportunity
to opt out on the exchange if need be.
Public health reporting as established under state law. Again, we hope to
extend out from that, but we are really beginning with the most solid ground
that we can. And also, finally, reporting for the facilities, the offices, and
also to the Medicaid agency and CMS in support of meaningful use. And the idea
that we have is that we will begin with for the most fundamental use cases
around which exchange would occur, erect a governance, develop trust in that
governance and then have that governance consider extensions of those use cases
DR. RO: Just quickly, the most common discussion that happens between our
clinical staff and say my staff, the community health assessment, is that push
between public health and medicine or clinical. And the reason I put that out
is because I don’t think the path is clear on how electronic health records and
clinical records will feed in to community health assessment because the
comment that comes up is that our populations are different. Who we are getting
clinical data and the other upstream factors we have talked about, if we talk
about place matters, how do we link the two. Really very difficult.
In terms of, I think, privacy agreements I think we follow the guidelines
that are set for us. I think that more information that you can provide to us
and help us work through that I think is greatly appreciated.
DR. GREEN: I would like to direct a question toward Dr. Park. We are so
grateful for your — Mr. Park. I apologize for the insult.
DR. GREEN: We are so grateful that you are here. My question is what are you
thinking over there and what do you want to ask this first panel that is
focused on the public health infrastructure.
MR. PARK: I thank you for the opportunity. It has been really phenomenal to
hear about what you have been trying to do. It is really heroic what you are
doing trying to advance the ball in public health. I would love to ask a deeper
question and one level deeper about your experiences. Can you describe an
example that you actually used power of data to mobilize action in the
community, a collaboration of folks to do something? Beyond reporting out the
data in the form of surveys, in the form — which I think is a phenomenal
contextual thing to do. Have you used a specific data point to try to issue a
call to arms if you will to try to get people to do something? I am looking for
examples of data as a call to arms.
MS. WELLIK: I will use an example from a few years ago. My county was the
first county in Minnesota to pass a smoke-free ordinance and I was asked by a
citizen group to initiate this action. As you recall back in those days, there
is a tremendous amount of opposition to this kind of thing. My county board,
which is my board of health, asked my advisory board to find out what the
public felt about this and to give them data that supported taking this action.
And so we put together a variety of things.
We did a survey of the community as to whether they support smoke-free
restaurants which is where we started. And that survey which was done by a
local survey research group indicated about 70 percent of our community
supported that. That didn’t fly until we did it two more times. We did it the
second time and another group did it the third time and the results were the
same. Then there got to be media support for the fact that these three things
are coming out the same.
The other thing we did was we shared with the board and our opposition that
what the cost and billions of dollars in Minnesota was to having smoke-filled
environments. And our opposition said that is absolutely ludicrous. There is no
way it cost that much money and I can’t remember the dollar figure. There is no
way it cost that much money in Minnesota as to the health and economic impacts
of smoking. I asked the county board if 50 percent of that cost would be
acceptable. They said maybe this figure is inflated, but do you think health
economists made a mistake in the range of 50 percent or maybe it was 30 percent
or maybe it was 75 percent, but are any of those figures acceptable.
That is what I mean about data not needing to be precise. Part of it was we
put a number on it and we started there and then we went from that point on and
started to demonstrate that we could collect data that demonstrated that this
information was accurate enough that it would mobilize action.
And that is why I said earlier we don’t have to have exact, accurate, and
comprehensive information to take action. We need to start talking about these
issues and put what information we have around them and then decide what it is
that is going to mobilize action particularly if we are taking action around an
issue that has evidence base and research data behind it in other communities.
DR. MADDUX-GONZALEZ: I think one of the areas of data that had the greatest
impact in our community has been on obesity. I applaud the CDC for the slides
that were done in obesity. I used those always so the community not always gets
the same audible response from the public. But then to take that to a local
level really has had an impact and with this health action and in other venues.
But it has led to definite changes and specific efforts to address obesity,
also the awareness that it is disproportionately affecting low-income
individuals. Some strategies such as we have enhanced the number of farmer’s
markets where food stamp vouchers can be used, WIC vouchers can be used,
community gardens in low-income neighborhoods, specific strategies to improve
the design for safer neighborhoods for physical activity.
And we just completed a really exciting process, again, based on the — and
the understanding of obesity through data where we spend a year working with
all of the city and the county planning directors and came up with a workbook
on how to improve community design to address chronic disease by working on
general plans and other strategies.
MR. PARK: Just to follow up on that. How did that chain reaction get
started? What was the data that you showed and who did you show it to and how
did you show it to them to get that beautiful chain reaction going?
DR. MADDUX-GONZALEZ: It has been kind of a multi-year snowball. It is
important because I can’t emphasize enough how powerful those slides were and
how if you can help us get that type of data about health issues. The community
is unaware of these things. Many members of our community, not everybody, but
many members of our community are unaware of this. And to present it in such a
way that everybody — people change the way they look at obesity from the first
slide to the end and it is changed permanently. That is very helpful. It is
basically working with multiple groups in the communities and then this overall
planning process with the 32 community leaders was really helpful.
PARTICIPANT: (off mic)
DR. MADDUX-GONZALEZ: Yes. We do have California Health Interview Survey data
which has been very helpful also to look at and I will put my pitch one more.
And if we can get health information exchange, we will have even more current
and more extensive information.
DR. RO: I think we have some similar stories on healthy eating, active
living, that kind of thing, but I want to turn your attention to our current
battle which is kind of for survival of our clinical services in the state
legislature. Our governor when she came into this new recent budget era wanted
to eliminate maternity support services. Our data demonstrating the impact of
infant mortality, low-birth weight has been really critical. Our community has,
our director has, and our executive’s office has been in constant advocacy mode
to try to save our funding for the maternity support services and our federally
qualified health center dollars from the state.
What we don’t have — we don’t have sufficient data on cost effectiveness.
We don’t have sufficient data on return on investment. This is becoming really,
really critical as we try to pitch our story. We are making progress. But I
will just say this is that intersection then of the clinical data, the data
then on improvement and the data on demographics and the populations and where
that hits where it becomes very real.
DR. PATTERSON: In our case because of the Switzerland model we maintain our
neutrality and very seldom actually other than making data publicly available
on the website do anything ourselves to drive these kinds of changes. Now
having said that, the collaborations that we work with have done quite a bit,
everything from HRSA Flex grants in regional health information exchange
supported by our infrastructure that move to quality certification in diabetes
for a family medicine at our regional hospital and it really serves a
six-county area in an integrated regional health information exchange to
demonstration of medical home projects beginning all the way back with patient
navigator initiatives and now to electronic medical home networks and also
through a CHIPRA quality improvement initiative that our Medicaid agency is
engaged in with the South Carolina chapter of the American Academy of
Pediatrics that is precisely about that kind of feedback loop around quality
and so on.
All of these things generate synergy. They generate change in local areas
and ultimately, I think, move the state towards I think a much more profound
dialogue about how we improve health and the various levels of engagement that
are required to do that from the patient level all the way up through the state
as an actor; and also the extensions of health to other areas. I notice that
there has been a lot of discussion of food and nutrition, for example, and we
have done some work in that area as well trying to bridge those gaps. It is
something that the ORS almost never does on its own because again we try to
remain in the background as a service and support entity and that helps
preserve our neutrality.
DR. STEINWACHS: I have four members with questions. Paul, Marc, Raj, and
Blackford, and about 10 minutes. If we could make the questions brief and then
we will keep the responses as brief as we can too.
DR. TANG: This is sort of a check on a summary of what I have heard if that
is okay. I was hoping to get their validation. One, thanks for the tremendously
informative and very inspiring panel. It is heart wrenching really. We had a
public health hearing at the meaningful use work group and this same kind of
story came up and I am going to talk more about meaningful use response to your
comment. It is just so powerful and thank you for the work that you do on so
Let me summarize the messages I have heard so that we can maybe digest it
and try to give back to you recommendations that would be supportive. One is
that you want access to more granular health data. Probably I heard also that
that is nice but it would be far nicer if it were in some standardized
framework so that you could compare locally and nationally. That is something
that could be set up as a national initiative.
The third is well that would be nice too if you had the local resources to
be able to deal with that, the professional services of data analysis. You
You talked about the meaningful use objective. Now, you do know that at
least it is 25 percent of the meaningful use categories are public and
population health. That was acknowledged in the public health hearing that we
had. The stopping point we have is that the statute only gives us in relation
to reward doc side to give information to you, but it doesn’t give as you
pointed out it doesn’t allow you to receive the information then to do good
And then an important part that David brought up was to provide feedback
back to the individual practitioners in communities so they can take action.
That is another area where I think they might be asking for help in terms of
not only this standardization but direct recommendations that support the
funding of activities that they need to do on a local basis that comes back and
pays us back at the community and national level.
Have I sort of gotten the messages as told by the three and four of you?
DR. MADDUX-GONZALEZ: I just had one thing. In addition we would love the
funding and also the technical assistance. I know funding takes a little
longer, but the technical assistance — it is 2011. That is stage one of
meaningful use. The sooner we can get that technical assistance. Each of us is
really trying to do it working with our individuals where this is not their
main area of expertise. Funding and technical assistance would be great in
getting through this.
MS. WELLIK: I agree. I think you definitely hit on the areas that are
important to us. I would encourage you to work with NACCHO’s informatics
workgroup as they really are looking at this from a standpoint of all local
health departments and are focusing on what we need in order to make meaningful
DR. STEINWACHS: Thank you for the comments and response. Let me turn to
DR. HORNBROOK: As I listen to this, I have a sense of a confluence between
the comparative effectiveness research agenda and this public health agenda
around — informatics. And that is convincing our citizenry that participating
in projects that share your data, that have you as a community involved in
research projects and demonstration projects is something that should be
valued. In comparative effectiveness, how do you convince patients that they
should be willing to be participating in research that compares treatment A to
treatment B when we have no data that suggests there is any difference between
them because there has been nobody that has done that research yet?
And then in the public health arena how do we convince people that they need
to let their data be — we are pursuing privacy to the point of self
destruction. In Oregon we have had three suicides by cops, by veterans. In all
three cases the data were there. When you do the post hoc autopsy you realize
that all the warning signs were there, but nobody shared the data so you could
not bring together the right kind of public health response that would have
We are hearing here public agencies at the basic level having a chance to
convince the residents of their respected communities how important it is to
begin thinking about the fact that privacy is something that has destructive
elements as well as productive elements.
DR. CHANDERRAY: There are several elements but I think I will limit myself
to data collection. Just like there are three sexes, male, female and veterans.
There are three kinds of populations which we need to collect data from:
insured, employed/uninsured, and unemployed/uninsured. Data from insured is
very easy to obtain. We don’t need to go down to the weeds to collect this
data. Insurance companies have this data all the time. They are sending data on
all kinds of drug use, diagnosis, and everything that are available to them. To
burden the physicians who have this sword of Damocles hanging over them in
terms of cost and in terms of liability, I think takes a lot of committed
physicians to provide this data to you. I think you can get the insured
population data from the insurance companies.
The collection of data from unemployed/uninsured is very, very difficult. I
think when I was chairing the committee organization to collect data – to
identify the need based resource allocation in Clark County in Las Vegas, we
collected the data in part by going to the unemployed people, but we collected
data from the emergency rooms where people without insurance were hitting the
And from that survey we found out that biggest employer in Clark County was
Wal-Mart, and we showed that Wal-Mart employed a lot of people but only on a
part-time basis so they can avoid paying insurance and as a result BBC came to
interview me. And Wal-Mart changed their policies. But I think the data
collection is important.
But I think rather than going down to the weeds because there is a big fear
element among the uninsured/unemployed people. The biggest fear is deportation
for illegal immigrants, which is a large part in our community and southern
states. The second thing, a loss of any kind of payment checks from
governmental agencies. These unemployed invent devious ways to collect these
checks and their fear of losing their only source of income will prevent them
from surviving and therefore refuse to participate any kind of surveys. I think
those are the biggest elements that I think are important.
DR. STEINWACHS: Raj, thank you. Blackford.
DR. MIDDLETON: I just wanted to echo everyone’s comments. It is a terrific
set of presentations and thank you very much for coming and sharing your
insights. In a way I look back over the several decades of public health work
and always looked with some jealousy. Compared to how we were doing with
clinical data management, it always seemed public health data were being
gathered at the national scale with national surveys and being used very
effectively to influence public policy and impact the nation’s health to a
degree. Whereas on the clinical side we were in a state of complete disarray
and largely had not automated much of anything until recently.
It strikes me that the tables have somehow turned with the national
attention on health care IT and we failed to keep alive what we all maybe take
for granted is this fundamental connection between these data.
I think about the recent PCAST report which focuses a lot on how to
characterize data with improved tagging to support reuse and better definition
and management, et cetera. I wonder if we can bring the same attention and
public health certainly is mentioned in PCAST, but it strikes me somehow we
have lost the attention on the fundamental Clean Water Act as Clem McDonald
says for data. If we can get the data right at the source systems and maintain
the integrity of the data with all these different types of ideas then I think
we will have a much better shot and the minimal dataset idea has been very
attractive to me for a long time where we might have minimal clinical datasets
from public health, community monitoring, and surveillance and clinical data,
I guess the question is to go to the heart of it: who is going to be
responsible? Who would you recommend be responsible and maybe for the committee
at large to take on this idea of really the PCAST recommendations, these
recommendations and these cries for help at a level? How do we actually add the
meta-data standardization exercise and data standardization overall to the
national agenda on both the clinical and public health side? We have lost
funding in informatics at the CDC. The NCHS has been under extraordinary
pressure and somehow we are only spending 4 percent on prevention of the
national health care dollar. We are missing the boat. Somehow we have to
reorient the policy conversation and identify someone who will be responsible.
DR. STEINWACHS: You can see we want the answers to big questions, Mary.
Thank you for stepping forward.
DR. MADDUZ-GONZALEZ: Can I ask — just comment one thing on that? I think
those are very good points and I am not familiar with the specific study. But I
think that there are real opportunities with going to the source of the data
that I believe will include all of the populations you are talking about
because, for example, in our community approximately 85 percent of Medicaid
recipients are seen by our federal qualified health centers and the majority of
the uninsured and those who will remain uninsured even with health care reform.
To me that is really the source of the data.
One option that occurs to me is something similar to HITECH expanding to
include the public health interface with that because this distinction between
community health and health care delivery is an artificial separation. And this
is a great opportunity to bring the two back together again and look at this in
a more integrated manner.
MS. WELLIK: I fully agree with Mary. I think that it is not impossible. It
is just that we are looking at it in a way that we don’t have to. I talked with
Dr. Overhage before this meeting and mentioned the importance of getting to a
point of having standards for local public health software. And he said no, we
really need to acknowledge the adoption of a standardized way that we are
already operating. We just have not acknowledged that there are a lot of things
that we are doing that are standardized by in large. We just haven’t
acknowledged that that is the way it is. I think an investment in finding out
what that is wouldn’t take a lot of time.
I think this committee might be in a position of assuring that you continue
to work with some of the local public health organizations that can tell you
what we are doing and you could then help us figure out how that might align
with things that are already in place.
DR. RO: I would echo, I think, and I would ask the committee to go back and
look at the recent IOM report on data because I think it does provide some
ideas in thinking about how to move forward in getting this to a standardized
dataset. I just would like to re-echo and push a little bit on understanding
what it means to have good public health and really thinking about the upstream
factors because I think it is really easy to move to health care and health
care access and to really think that it is that idea of food, water safety,
where you work, where you live, where you play and that really has a huge
impact on our health. And we don’t collect all that data just at the health
That collaboration I think that you have heard before is really important. I
would say that the collaboration in informing our citizenry and those who are
in the shadows where data isn’t collected that is where those uses of the
Promotoras, the community health workers, the community-based organizations to
get folks so that there is some trust and relationship between the government
and government is always not looked upon nicely, but essentially us as a
DR. STEINWACHS: David, is there anything you would like to briefly add?
DR. PATTERSON: No, I would echo that. Although I would say that in terms of
the standardization that is already emerging from the ONC, we have done a lot
of work too. They recognize some of the CDC’s standards at least from moving
the data. What I am hearing from my community level health colleagues is that
it does ensure that the appropriate data elements are populated or standardized
for analytic purposes. I would strongly encourage that.
DR. STEINWACHS: I understand, Todd, that you are going to have to leave at
the break and we are going to lose you. We wanted to extend this just a little
bit and ask you if you could give an update on the community data initiative
and I understand there are some new releases coming out and I thought good
context for this.
MR. PARK: Absolutely. Just a couple of notes. I think actually echoes,
reinforces hopefully will be helpful with respect to the major themes we talked
First of all we have actually just released softly, not actually publicly,
but it is actually in the public access, something called the HHS Health
Indicators Warehouse. If you quietly go to healthindicators.gov, you can check
out the new Health Indicators Warehouse that Ed Sondik and the NCHS team and
Bill Heimer(?), Amy Bernstein, Jim Kramer, and others have put out there. We
are actually going to be announcing its existence publicly on February
17th, along with a big public competition to build apps off of it.
It is basically a professional grade data hub. And it is the first time HHS
has put out a public resource that is a consolidated inventory of every single
community health indicator that we really know how to manufacture. It has all
the Healthy People indicators for which there is data and that will get
obviously added to over time as more data gets produced against Healthy People
2020, all the community health indicators, all of the metrics we produce for
the match, county health ranks initiative. Also it includes a couple hundred
brand new Medicare indicators that have been produced and released for the
first time in the public at the hospital referral region level across the
country; metrics of utilization of Medicare services and the service population
by type of service, rates of prevalence to disease, measures of quality and
prevention, avoidable ambulatory care sensitive hospital admissions by
condition, et cetera.
All of this has been made available in the standardized form for whatever
level of granularity we have it for; national, state, hospital for region and
county, for whatever time period we have it for, whatever it is sliceable by
gender, age, et cetera, we have that available. There is standard meta-data
that is also attached to — not in computable form, but meta-data that is
information that has been attached to each indicator.
Also for each indicator for which there are evidence-based interventions, we
have that available as well. And the data is not only available via the website
and in downloadable form, but also via a web services API, and API is of
course, application programming interface. It is essentially a really easy way
for other developers, other application websites, to automatically hook into
the data and pull it out, as opposed to having a human walk up to the website
and hit download and then upload the file into wherever you want it to go.
Although if you want to do that too, you can go to the website and hit a
download button and download it in Excel or CSV and so forth.
But it is one-stop shop. Ed calls it a Costco for data. All the data is
free. It is an effort by us to provide an unprecedented amount of standardized
community health data across multiple dimensions of health, health care, and
determinants of health. A ton of data in there about food and water and
environment and stuff that we know is actually more impactful in the health
care systems in terms of community health status. Everybody else is trying to
consolidate one-stop shop of all of that in a way that is super easy to use. It
is a combination of a huge amount of hard work.
MR. SCANLON: It would have as part of that linked data, linked information,
some of it would have cost effectiveness data, interventions that are described
in various ways some of which would have good cost effectiveness information
and some of which won’t, but presumably prove it. All of them are proven
interventions. Whether it is cost effectiveness that is another issue.
MR. PARK: It is actually a huge reservoir of standardized data along these
lines, that should be very easy for you all to retrieve and mash up into your
own applications and websites and that is, in fact, exactly what it is. It is
not really designed to be a consumer facing resource by itself. It is designed
to be a resource for people like you, that easily take our data, all these
different indicators, interventions, and mash it up and use it to support what
you are doing.
The second thing is I just wanted to mention quickly, is that later this
month we are debuting an even bigger site called healthdata.gov. This is
actually a site that is a new community on the Federal Government’s data.gov
super site. But healthdata.gov is a new site that is going to contain several
things. One, a catalog of every single federal dataset that relates to health,
health care, or determinants of health, that we are making available in
readable form for free — downloadable or via API. It is one place where you
can find all of it. And, again, there is a ton of data in there that is not
literal health data, but that drives health — it is not just from HHS. It is
also HUD and from USDA and from EPA and from VA, et cetera.
Secondly, it will contain a link to a virtual application’s expo, which has
a cross section of really cool applications people have built that leverage the
data or help you actually use the data, and that will get added to constantly.
The third is an online community, where anyone can join and complain about
the data and ask questions about the data, request new data, talk about what
they are doing with the data; focus on specific nuances of particular dataset
that performs a particular fan club that wants to talk about it. That will be
And then finally there will be a catalog with what we call additional data
sites. These are nonfederal data sources at the state or local or private
levels, that are making data available on the same terms that we are so for
free — that we think would be useful. It is, again, meant to be kind of even
larger one-stop shop for all kinds of data that could be helpful, touching
really all aspects of health.
The Health Indicators Warehouse will be one of the tools that is featured on
this uber one one-stop shop, but hopefully that should be an even more robust
larger source of all kinds of use of datasets that you can use in your work.
A couple of things I just wanted to mention just to react to what you were
saying which I thought, again, was so helpful. There are actually a growing
number of applications, many of which are free that have been built to actually
help people like you use this data. We can get into that more offline, but one
is a Palantir which is this really cool, analytic Silicon Valley has put online
for free, a tool called analyzethe.us. And analyzethe.us anyone can get an
account and start using it effectively — Palantir — it enables you to mash up
tons of different datasets that weren’t designed necessarily to live together,
but it enables you to mash them up for the purposes of analysis and gives you a
very high powered analytical tool that you can use to look for patterns of the
data. They have imported a ton of our health data and importing more as we
speak and you can actually import more of your own data into it and do
analysis. You should check that out.
PARTICIPANT: What is it called again?
MR. PARK: It is at analyzethe.us and it is via an outfit called Palantir. We
obviously are a big fan of what the Network of Care for Healthy Communities’
people are doing and others along those lines.
One thing I would actually invite you to do is come as guests of honors to
our June 9th meeting that we are hosting on health data — that is
not the official name, but we are — it is being hosted by the Institute of
Medicine and it will be about 500 people from all walks of life who are really
interested in supply and the use of health data to improve health. We are
really excited about having communities there that are engaged in the hard and
worthy work to actually using the data to move the needle in a tangible way to
make communities healthier. We would love to have you there, A, to talk about
your experience and B, to talk with innovators who have data and who have
technology about what they can do to help you make your cause one that gets a
little bit easier over time in which you make progress.
One key take away for me actually is that this whole notion of actually not
providing data, but providing expertise baked in with data to help address the
potential absence of epidemiologists and statisticians in every health
department in America. That is something that technology can really do
particularly in this area. You can really actually through things like Network
of Care for Healthy Communities and other tools and package the power of
analytics together and produce actionable tools that you could use to help
persuade people to do stuff, to help identify things that you might want to do,
to help share learnings from across the country and mobilize action.
Basically it is a long way of saying we would love to invite you to come and
be our guests of honor and help to inspire others to do cool stuff to help you
bake data analytics expertise knowledge and social mobilization techniques
together and to new kinds of services to help you do your thing especially in a
time when you don’t have a ton of dollars floating around to hire a bunch of
experts and a bunch of community organizers. Technology can really help do
that. It is an open invitation June 9th and Greg Downing who is the
guy who is running June 9th will be sure to send you special VIP
invites. I am just so excited of what they are doing.
DR. STEINWACHS: I wanted to thank you for the summary, but even more
importantly for the energy and the inspiration you bring to this and trying to
marry the Federal Government with the locality, I think, is build this bridge
and you have really provided great leadership for that.
I want to thank the four panelists and I hope they will stay with us to the
extent they can through the day because this is a learning process and I hope
that you found too some of this valuable sharing process. And certainly you
brought lessons to us that are important.
The moderator for the next session said that we could take 10 minutes out of
that session. We are going to reconvene at 11:25, 10 minutes from now, and look
forward to the next panel. Thank you.
Agenda Item: Focused Community Health Initiatives –
Sallie Milam, Moderator
MS. MILAM: We will go ahead and get started. Welcome panel two. We have four
different communities with us today. We have Boone County, Missouri. Stephanie
Browning, who is the administrator for the Columbia Boone County Health
We have Denver, Colorado, 2040 Partners for Health, and we have several
individuals. We have Janet Meredith, the executive director with 2040 Partners.
We have Deb Main, and we also have Tracey Stewart.
We have Mahoning Valley, Ohio, Matthew Stefanak, the health commissioner for
the Mahoning County Health Department. And then we have Gwendolyn Flynn, South
Los Angeles, California.
Do you want to go ahead and get started, Stephanie?
MS. BROWNING: Thank you. One of the things I thought was interesting after
listening to the first panelists, I could almost echo everything I heard out of
something in every one of those first panelists’ presentations. I found myself
shaking my head a lot. I have a feeling the same themes may continue throughout
I am trying to keep my remarks today just very similar to some of the
prompts that we were asked as we came into this. In terms of community health
improvement goals our vision for our department is optimal health: safety and
well being for all. We have a real focus on health equity and health disparity.
Five years ago our city reorganized and added the Office of Human Services
under public health which means that we fund about — we fund multiple social
service agencies in addition to our public health role around youth and
families, economic opportunity, and dependent living, mental health and basic
We also run the human rights commission. We have numerous commissions that
we run in addition to the board of health. What it has done for us over the
last five years has really helped us to try to think bigger and broader outside
of the scope of public health and trying to figure out how we as a community
can move things forward. Our local initiatives of course focused like many
others on overweight, obesity, physical inactivity, food security, chronic
Let me tell you a little story. I came to Columbia from a border community
in Arizona. When I was in Arizona, there were times during the year when
unemployment hovered at somewhere between 25 and 30 percent. There were low
rates of high school graduation. A significant part of our population was
limited English proficiency. We had very high rates of tuberculosis, low
immunization rates, vaccine preventable disease outbreaks that were routine,
and women that had little or no prenatal care.
In 1999 when I became the director of the health department in Columbia, I
honestly thought I had died and gone to public health heaven. Columbia is the
center part of the state, similar sized population to where I was. Right now it
is about 156,000 people. It is the home of Flagship University for the
University of Missouri system. We also have two private colleges. We have six
hospitals. Unemployment in 1999 — remember I left some place that was — 30
percent. It was under two percent. Right now we are under six percent. We felt
the economic pinch so to speak, but maybe not so much as other places in the
country. We have a high level — fairly high levels of people that have a
graduate or bachelor’s degree and beyond.
As you can imagine, we are also a rural/urban mix of a community. Outside
the city limits it is very much farm land. It is not that large of a community.
Everybody still knows each other and it is very diverse.
When I came to Columbia of course I looked at all the data that we had and I
am like wow this place is totally healthy compared to where I have been. And I
would say that a lot of what we use on a regular basis comes from the Missouri
Department of Health and Senior Services and what you see up here is really
from the leadership of Garland and that is just a snapshot of what our state
has been able to do to help us to get good county level data.
Back to my Columbia/Arizona thing. It didn’t take me very long to realize
after living there and making some connections in the community, that while
things looked really terrific on the surface, down at a smaller level at a
neighborhood level the disparities were great. There were lots and lots of
people that did not have what it appeared that our county had when compared to
others. For us it has been a real long process over these last 10 years of
trying to call those issues to light.
Probably within the first couple of years that I was there we applied for a
grant that had to do with diabetes prevention, chronic disease prevention,
heart disease, and basically I think put a very solid application together to
get some funding for that and were pretty much denied that funding because our
data didn’t support our need for that. Instead it went to another part of the
state where the data looked much worse. Yet I can point to populations where I
knew there were problems. I couldn’t get down to the level that a funder was
going to accept.
Routinely I have very supportive city council and county commission and in
fact so supportive that if you look at Missouri probably get the equivalent per
capita of a can of soda from the State of Missouri and it is on a decline. Our
federal funding also is probably under 15 percent. The difference in a state
like Arizona with 15 counties is that when federal funding comes to a local
public health department, you have enough resources to hire staff, have
training, do all kinds of things when federal funding comes down to a state
like Missouri with 114 health departments. You might have enough funding to
have 4 hours of somebody a week devoted to something. Fortunately our city and
county have been wonderful at stepping up and providing those resources for us
and they make up of about 75 to 76 percent of our funding.
But times are tough and the city and the county sales tax revenues are down.
Property taxes are down. And when we hit in the city, the 100,000 population
mark, a funny thing happen. Our crime rates appeared to kind of go up a little
bit which is apparently quite common. And growing up in Los Angeles I still
think we are a very safe place to live, but long-time Columbia people do not.
Now the challenge on our leaders is we have to find more police officers and
everything else is they are trying to scale back, fall back from everything
else. It is imperative at the local level that we tell the story about how
social determinants, public health, all of those feed into the things that
people are so worried about with crime and the economy and education and that
is sort of what we are attempting to do.
I am just going to go through just a couple of slides. I just took some
shots of — there were some questions asked about what kinds of data do we use
now. Like I said most of our data that we use is county level data from the
state. Our hospital discharge data, I believe, and emergency department data we
can get down to zip code level in some cases, but in our case a zip code
doesn’t mean a lot like, for example, I live in 65203. My health department is
also physically housed there. It is part of the central part of Columbia. It is
an area that is lower economic citizens. And then the very far south of me is
some of the highest and the nicest subdivisions in the whole community. 65203
doesn’t tell me much.
This is a project that we have been doing for a number of years. It is
getting ready. It will be updated again for 2010 this month. We ranked
together. We work with OSEDA, the Office of Social and Economic Data Analysis,
at the University of Missouri, they provide the support for it. We bring public
school personnel, social service agencies, FQHC, public health, mental health,
just a number of partners out there in the community together, and we have a
set of benchmarks that we look at every year and we just try to keep the focus.
We do a big press release of it every year and get a lot of good media
coverage. Those are like little things that are going on where people are
working together trying to collaborate.
Our most recent effort and that is what I will talk the most about is we
have joined in as part of a community issues management system. And it is
through the CARES program at the University of Missouri which is the Center for
Applied Research and Environmental Systems. And basically it is a way — it
started out with a small number of United Ways that got together with the
University of Missouri looking at how they can pull together various datasets
into a way to better frame issues, bring people together around issues, and
have a one-stop place where people can work together.
Our original partner in that was Charleston, the Trident United Way in South
Carolina, probably one of the original Lehigh Valley health systems up in
Lehigh Valley, Pennsylvania. But it has really grown in the past few years.
That is just a little map that shows where the partners are.
There are too many things in here to just show you different snapshots. We
have issue notebooks. Let me tell you how a person would use it or how we are
using it in our department. We are using it for emergency planning purposes.
That is one means for it. We are able to take map all our resources, assets, et
cetera. Bring our partners in to that process and be able to geographically
look at where in an emergency we would need to put those resources. We have,
like I said, multiple datasets in there so we can pull up census tracks,
We have excellent city and county GIS layers. One of the things that we plan
to do is upload using CIM all of our transit maps where every bus stop is. We
have tools in CIM that will allow us to say how many people within X miles and
map all of those of this bus stop. We have used our environmental health data
to put in where all of our restaurants, fast foods, grocery stores, convenient
stores, farmers market, you name it. We have all of those in labels.
What we are really trying to do is not control this data. Our goal is to try
to bring people together, teach them how to use this tool, allow them to form
their own work groups. To form a work group they actually run through me as the
administrator of that. We set them up as a work group. They can invite people
to that group. They can use any of the 500 datasets. They can create their own
custom maps. They can create their own custom reports. We have it broken down.
They can do it by legislative districts if they want to go talk to a legislator
and they want to be able to show some specific examples of what is going on in
that legislative district. They can. We have an upload tool.
I will give you an example. Tucson, the United Way there was working on a
food security thing. They had a neighborhood that had a high number of
immigrant people in the community that where really food was an issue. They had
a couple industrious students go on and go and map where all the fruit trees
and sources of food were and they actually uploaded that data, marry that data
with their poverty data with all their other community assets and resources and
they are really working at the neighborhood level on how to bring people
together to solve problems and support each other. Our goal and this has been a
little different. We kind of brought this in, tried to train people, but then
tried to turn them loose.
I am going to show you just a couple of little — we have a group like the
others, working on food safety and security. And one of the groups is
interested in access to quality food, neighborhood grocery stores, et cetera,
which we don’t have. What they have done is they have taken a very small,
central core part of Columbia. They have mapped it by the census tracks and the
poverty levels in those census tracks. The green indicate where the community
gardens on. They are working to work with the city right now to identify places
where they have vacant lots, where they are tearing down homes that are
substandard housing that have been dilapidated and left untouched to be able to
turn those into more community gardens and getting that project moving along.
The same group has worked with the transportation department and there is a
whole new movement to try to expand our transportation in the City of Columbia,
our transit, so that more people in these neighborhoods can be served by
grocery stores. Farmers markets are getting ready down in the housing authority
to open up a community garden and farmers market, where it is going to be like
a micro-business going on within that area.
They have been mapping how far the schools and fast-food restaurants are.
You name it. We are letting people try to use this data. And our goal really
has been for us is how can we take data — something other than the county
level, that health data. This is wonderful for neighborhood development,
community development, but then when they pull up the health data, it is just
one big blob. It shows Boone County as x-rayed and they can’t see anything.
They are not sure how they are showing outcomes. Some of the things that we
hoped to get, birth and death records from the state, we know we can get that.
The problem is we just don’t have the resources. We left positions open. And
the state is worse off than we are I can tell you that.
WIC participant data. We have just gone to electronic WIC system, record
system. We should be able for our own internal planning purposes to get shape
files to use with CIM, keep that data private within our own system, and we
will be able to learn better how we can target our WIC population.
Our FQHC is right next door. We have a good relationship with them. They
have gone to an electronic medical record system. We have done the same. We
hope to be able to do that.
Again, I am not going to echo everybody else’s theme, we need something more
granular than county level. Again, the same theme, shrinking resources at the
local level. And to me the charge is how do we maximize what scarce dollars we
have and direct it to the right place in the community because we can no longer
afford to serve every — we are always going to serve everyone, but not to the
degree that we need to make change. We need to get down to a smaller level.
Lack of dedicated staff, lack of training and expertise, the same thing. And
of course always privacy and confidentiality concerns.
One thing I would suggest is as I was looking at the Center’s website and
all that, and listening earlier to the health indicators, one of my staff said
wouldn’t it be great if there was an easy way for us to know what is new out
there. We don’t have the time all day just to keep exploring and one thing
leads you to another to another. Wouldn’t it be great if there was just some
kind of automatic, something that pushed out every week and said here is new
data sets, here is new resource to go checkout so that we didn’t spend all of
our time searching.
Again, more investment in the public health infrastructure and anything that
supports collaboration and community engagement is definitely something we need
to keep moving forward.
I will stop for now since we have a big group and a shorter amount of time
and take questions later.
MS. MILAM: Thank you so much, Stephanie. We will now turn to Colorado, our
three-part panel. And I will turn it over to Janet.
MS. MEREDITH: Thank you. As we pull this up, I thought it would be good to
introduce — there are three of us here. I am Janet Meredith, the executive
director of 2040 Partners for Health.
DR. MAIN: I am Debbi Main. I am from the University of Colorado Denver.
MS. STEWART: And I am Tracey Stewart with 2040 Partners for Health.
MS. MEREDITH: And the reason that the three of us are here is because this
is how we work. We will talk you through our slide presentation here and answer
MS. MILAM: And just for your planning since there are three of you, my watch
indicates that your 15 minutes will end at noon and I will give you a count
down from 5.
MS. MEREDITH: All right and we will try to behave. First of all, to give you
a sense of who we are and where we operate, we are in the Northeast part of
Denver and we are a geographic basis of five neighborhoods that are about
100,000 people I believe at this point, which is about 20 percent of the city
and county of Denver. The genesis for our program originally was because of the
creation of the new Stapleton Development. You may remember there used to be an
airport and as that moved out this created a huge urban development opportunity
that brought very middle class or even upper middle class neighborhood into the
middle of four other neighborhoods with very diverse issues and disparities.
That is the short form.
The way that we work — first of all we are a tiny organization. We have two
full-time equivalents. That is three staff, two of whom are half time. We work
collaboratively with the university and with other researchers outside the
University of Colorado and with the community and we really see it this way.
What 2040 does is provide an infrastructure and a way to convene people. We are
an ongoing organization that allows for a long time view and can bring
different components of health issues. I can answer more questions about that
if that is helpful to you along the way.
Our overall goals are these: to improve community health in the areas that
are of community priority. We work in research on a community-based
participatory research model. Community priorities are what drive the work that
We want to reduce health disparities, again, over the long term. 2040
references the year 2040, and provide a long term infrastructure for community
health improvement: data source, funding, evaluation, programs, community
We want to achieve measurable results in community health over that long
term and work with our partners because obviously we are a very small
organization ourselves. We rely very heavily on partners both community and
health providers and any kind of community infrastructure to help us achieve
In thinking about the biggest challenges that we face in acquiring and using
data, there is a challenge in having ongoing funding. We are not talking about
one off projects that will allow us to accomplish those goals. They have to be
connected. We need some kind of a long-term funding opportunity to collect,
evaluate, and use the data in the community in order to make sure that we are
actually improving health.
In order to keep the community engaged in CBPR we absolutely have to
implement programs that will improve the community’s health. The community is
not just interested in being another research site. They want to see their
health improve. We have to be able to implement programs in order to accomplish
And like everyone else we wrestle with data sharing concerns and logistics,
meeting the needs of all of the partners, making sure that we stay out of any
commercial use of that data, and that becomes something that we wrestle with
The strengths that we have in terms of acquiring using data are that we have
very strong community partners and connections. And the trust that is built is
really the reason that we can work with the community and you will hear more
about that. One of our staff is a community liaison. What she does is she
already has very strong connections in the community, but she maintains those
connections and trust but of course both Debbi Main and Tracey will talk to you
about how we work with community.
What we need in order to accomplish our goals in the community once more
neighborhood level data. We will talk about how it is the neighborhood level
data that mobilizes the community that feels like this data actually pertains.
We recognize this. We understand this and we see that we have an issue. And the
resources to make a difference to the community and to the community’s health
over the long term using data and then having that data available to be able to
mobilize programs to improve health.
DR. MAIN: Okay, I am going to actually talk about — what I am thinking
about is it is a use case for sort of local level/neighborhood level data so
think about — I am going to tell a story in the next few slides to just show a
use case of what happens when you begin to work closely with community and
collect local level/neighborhood level health data. This project actually
started that is called Taking Neighborhood Health to Heart and it is a
community-based initiative for collecting the kinds of data that matter to the
community. This is actually a picture of during one of our council meetings and
we actually meet frequently with this cast of characters. In fact, Tracey, I
think you may be in the picture.
Let me talk a little bit about the beginning of the project and again I
think the distinction is really important that this began as a research
project. We are not out of a local health department. Our goal was just to
begin to work with community and show that you can make a difference and
collect meaningful data.
It started as an NIH-funded project and the first goal and I think this is a
really important point is that we began by developing partnerships with
community and again Janet showed you the five neighborhoods. That was our
number one goal. Instead of just going and collecting data we farmed what I
think is now a very trusted partnership.
The second goal once we really worked together is we began to collect what
we considered to be really critical health data driven by the health priorities
of the community. That was our second goal is to really dig in within the
neighborhood level and collect data that matter. Our focus were those areas
although we collected kind of a broad set of data and I will show you a picture
And then the final piece and again I think this is another really important
point is it is not just collecting data. It is really working together with
community to make sense of data so they are actually believable, relevant, and
used and I think those are key pieces. And then it just so happens that the
research focus when we originally wrote part of the grant proposal was to look
at the impact of built-in social environments in health disparities. The other
magical piece of this is really the needs and the interests of community and
academic researchers really fit well together.
Janet already you or showed you a map of our five neighborhoods and I just
want to quickly go through this because Janet mentioned the fact that these are
really diverse neighborhoods and I think when you think about a use case in
five neighborhoods, it is the real diversity that is of interest and that again
just a few key things and Janet mentioned, for example, Stapleton is at the
bottom. And if you look at in terms of diversity, less so, but you look even
within two neighborhoods say Northeast Park Hill and Park Hill, very different
distributions socioeconomically on a number of measures. Again, East Montclair,
really interesting neighborhood in and of itself, as is Northwest Aurora, the
only neighborhood that actually is outside of Denver, which is a whole
different set of issues when you are dealing with different counties. But you
can see the point that really we are thinking about or we are working in very
diverse neighborhoods, yet we are coming together and we work together on our
common interests and priorities.
I don’t expect you to memorize this slide, but I think the point of the
slide is that we are as a community and again community is involved in
identifying measures, helping us think through stratifying samples for
household surveys, a number of things. Community is really all over this with
academic partners. A number of different levels of measurement whether it is at
the individual household level where we collect data, whether it is doing very
careful audits of neighborhood environments around sidewalks, lighting,
graffiti, bars on windows, et cetera. And then actually doing a really thorough
job, thanks to community interest on collecting price, quality, availability of
fruits and vegetables in every store within our footprint, again, and a number
of other things, rec centers. We use primary data collection as well as
secondary GIS-related data, census data, et cetera to paint the picture that I
will show you in the next slide.
Now that I saw the Sonoma slide I am going to really push our technology
even — next time you see this you are going to see trees popping out and it is
going to be 3-D. I love to have a goal. That is the next goal is to really make
this pop out. I think the interesting piece about our data collection — you
see our data, again, household community, et cetera.
And then the other key piece of this is that we have collected a lot of very
important, interesting information but the community is very meaningful
involved in how we share data, how we disseminate data. We do things in the
newspapers. We have very well written and well designed handouts for community
meetings, et cetera. Community is involved in all of this and I think that is
the key piece of this.
And then finally this is based on our experience and moving forward. These
are what we see as the key pieces of what an ongoing, very dynamic neighborhood
level data collection system should entail. I will stop there and now it is
Tracey Stewart’s turn.
MS. STEWART: My name is Tracey Stewart and I am from East Montclair, one of
the five communities. I got involved in this project by word of mouth. I was
recruited by what we call trusted sources. I got involved because I happen to
be a community organizer on my own right and because one of the people who got
me involved is a community leader.
The biggest part of the piece of it for me was to be able to extend what I
do for a living into a new area which was health. I had always talked about
health as a health care policy advocate, but never looked at overall health
including my own. This was really important for me because it was going to make
a difference in my personal life.
A couple of things that came out of it was I learned how to eat better. I am
a big food advocate now and a part of that was realizing through our studies
that we didn’t have any real sources. It wasn’t all my fault. The next piece
about it was really about public safety issues. The majority of folks living in
East Montclair area are women heads of households and so we are very concerned
about public safety not only for ourselves getting to work, but for children.
The difference that it made is that you start to realize that people really
are out there and they really are active and there is sort of this myth about
people sitting around on their sofas doing nothing and not caring. People
really do care. It is just you have to give them something to rally around. And
what was really good about this research project is that we were involved from
I am just going to jump ahead and talk about what it meant to be involved
from the beginning. What it meant was privacy issues around data collection
were mitigated because we were involved in the design of the survey and also
the ability to take ownership as a community member of the data that was
collected made a big difference. I know people keep talking about
confidentiality today, but I guarantee you that one of the best ways to
mitigate that is to have folks designed in the data collection themselves and
they will freely give it up. That is what Facebook is all about.
What really mattered was confirming for us what we already knew and also
shedding a light on things that we didn’t know. We didn’t know we had an asthma
problem; therefore, we didn’t know we had an environmental problem around one
of the railroad yards. And then also allowing us and I mean us meaning
community members to engage in solving our own problems and getting involved in
the policy part of that working with government entities and rallying for our
own solutions. All of those things made a really big difference and continue to
make a difference. This is going to be the fourth year. Those 60 people that
you saw in that picture are all volunteers for 4 years.
MS. BROWNING: We had three quick wrap-up statements that we point down. I
think each one of us contributed one. I will just say them quickly. Data
provided the standard that allows the community to all get together and to be
on the same page. We have a very diverse group and this allows the group to
focus and set priorities and otherwise it would be tricky to do that.
The communities want health information that actually is specifically
related to them. They recognize themselves in this data and that helps it be
used and have impact. With this community data we are together developing a
model that will allow us to work on the long-term health of our five
MS. MILAM: Thank you so very much, Denver. Let’s move on to Ohio. Matthew
MR. STEFANAK: Good afternoon. Thanks so much for inviting me to join you
today. It is very timely this discussion for us. We are in the midst of a
community health assessment and planning process for the Mahoning Valley region
of Northeastern Ohio, three counties: Trumbull, Mahoning, Columbiana; the
Youngstown, Warren metropolitan area. We are finding that the community health
assessment and planning is a lot like passing laws or making sausage. It is not
a surgically precise process.
What we think we will come out with is a plan that is useful for the key
stakeholders in our community primarily hospital systems who are brought to the
table to participate in this and fund this by the Affordable Care Act
requirements for their tax exempt status and for us and the local health
departments. A community health assessment and plan is two of the three legs of
the stool for voluntary accreditation. My health department was 1 of the 18
health departments that underwent the beta test of accreditation standards this
last year and we are positioning ourselves to go live with accreditation
sometime later this year. We are hustling to get through this.
Your committee staff asked us to answer five questions I believe and begin
with about a 3-minute story about how we have used community health data. I
thought that I would share with you that we have had great success in using
Ohio Family Health Survey data to benchmark and measure improvement in the
enrollment in Ohio’s Children’s Health Insurance Program, the CHIP program,
about 10 years. Beginning in the last decade we had some cash available to do
some grassroots retail marketing of the CHIP program to families who may
benefit from it, be eligible.
The Ohio Family Health Survey data that became available to us early in the
part of the decade showed that we had made some pretty impressive progressive
towards the goal of universal enrollment or insurance coverage for kids in our
community. Subsequent Ohio Family Health Survey showed that we continued to
make progress. In fact, this quote from our local newspaper, The Vindicator,
points out the Mahoning County has the lowest rate of uninsured kids among
Ohio’s metropolitan counties despite having the highest rate of kids living in
families below poverty. We are doing something right and we wouldn’t have known
that our local coalition efforts were bearing fruit unless we had access to
this kind of county-level data.
This is partially federally funded I believe. This will have Family Health
Survey data. It is conducted by the Ohio State University. We are not sure if
it is going to be available in 2010 on a county level and hopeful at least at
least it will be available on a metropolitan area so that we can continue to
measure our progress towards getting kids enrolled in this wonderful health
These are questions beginning with the first one. What are your long-term
goals for community health improvements given the necessary resources and
conditions? This is hot off the press. These are the priority areas and they
are formulated in the form of questions that are community health assessment
and planning group selected this last week. They have just begun to identify
some measures or metrics for those different priority areas in our plan. These
are categories of measures. I think the next step in the map process, the
NACCHO map process that we are using to plan is to select the right mix of
leading and lagging and process and outcome indicators for each of those areas
that you see there and then begin to develop some strategies that are part of a
logic model that flows from each of those priorities.
The second question was what data are you using or do you hope to use to
accomplish your goals. There is a lot of data. I suppose this is kind of off
the cuff, but Census data is very important and not only for some of the
community health assessment planning we are doing, but for measuring
penetrance, I guess, of outreach efforts to get kids screened for lead
poisoning, for example, in zip codes. We rely on that to find a denominator for
our screening rates in those areas.
SMART BRFSS is very important. The Mahoning or the Youngstown-Warren
metropolitan area is one of the metropolitan areas that are over sampled. And
in fact we even have county level BRFSS estimates for many indicators that are
important to us. I am sure it will be part of our community health plan.
The county health rankings really turned out to be a foundation I might call
it of the assessment indicators that we will be tracking in future years
through the community health plan that is under development right now.
The Ohio Family Health Survey, I have already shown you an example of how
useful that was for coalitions, benchmarking and measuring their progress.
We collect data locally in our local health department as you might expect.
Communicable and environmental disease reports and in particular child lead
poisoning is an example that I gave you are important.
Hospital discharge data from our hospital system partners is turning out to
be an important source of access to care. I think one of the things that we
will be tracking through our new community health plan is the numbers of people
who present to local emergency departments with ambulatory care, sensitive
conditions. Hospitals have been able to bring ED and patient discharge data
into the mix.
School surveys are something we have done for a long time. School-based BMI
measurement is big in a lot of places, mandated in a lot of states, not in
Ohio. It is optional. I serve 14 school districts in my county and one of them
has embraced annual BMI measurement as part of its overall coordinated school
health approach. We have a lot of data on those kids; longitudinal data on kids
in the Springfield school district that serves as a lagging indicator I think
of school and community-based efforts to slow the rise in childhood obesity.
County-level BRFSS. Early on when I became the health officer, I realized
that this a very important to supplement BRFSS data that is available to us
from the state with locally collected data and answer questions that people had
that BRFSS couldn’t collect primarily in the environmental health area, healthy
homes area, radon testing, exposure to second-hand smoke for kids in homes.
Firearm safety storage. That is a tough question, but we asked that. I
certainly would hang up if somebody called me and said where are you storing
your guns, but we got data on that. Over the years we have done three county
level BRFSS. We would like to do it again, but we have run into some problems
with the data analysis and the survey.
MS. MILAM: You have about 5 minutes left.
MR. STEFANAK: What do you particularly need in the way of data and other
resources to be able to accomplish your goals for your community? This sounds
like a broken record, more granular data on a community, my town or even my
neighborhood, certainly my village or township. Warren and Youngstown may as
well be as far apart as San Francisco and New York in my community.
County health rankings — I am sorry. Commitment to staying the course with
chosen community health measures. If we are going to use the county health
rankings as a report card on our progress towards schools in our community
health plan, they are going to have to keep it up. Keep collecting that data
and reporting it ideally every year, at least every 2 years.
I alluded to this earlier. Access to academic partners or expertise in
population-based survey data collection and analysis. I am still sitting on the
2005 county BRFSS dataset. I need to find somebody who can properly adjust for
us so it is represented of our population. I have not been able to find folks
in the state health department or in any of our academic centers in Ohio who
are familiar with the use of that software that is used by BRFSS to do those
What is the single most important message you want the committee to hear
about your community as a learning system for health? We are finding that
people — this always happen in a community health assessment. The community
volunteers who get involved and they want to develop measures that are unique
to their community like how many community garden sites have been tested for
lead in the soil. It is not available off the shelf, is it? The struggle is to
help them get the answers that they want to questions like that but still
accept what you might call a pre-fixed menu of measures like the county health
profiles that is readily available and that is available for other communities
so we can compare our progress and our status with other communities.
And then finally too many indicators can lead to indicatoritis. There was a
speaker from NCHS at APHA last fall who coined that term and I have taken it to
heart. There is a lot of data. With a part-time health educator,
biostatistician/epidemiologist who do these community health report cards and
assessments we have to be parsimonious in what we select.
The final question was what issues of privacy or confidentiality have you
faced in the use of local data. And I think this is a very good story. Child
lead poisoning is a reportable condition in Ohio and it is reportable to the
health departments. We know who the kids are with EBLs, elevated blood levels,
and where they live. And we also know what houses have toxic levels of lead,
so-called toxic houses that have poisoned these children. Many of our partner
advocates felt that we should share that information with the community so that
families don’t move into these toxic houses that are unremediated unknowingly.
We struggle with how to present that data. Initially we just published the
list of rental properties and their owner’s names that have poisoned kids. Of
course the landlords were not terribly satisfied with that. They wanted us to
not treat them preferentially. After some debate having an ethics committee in
my health department we decided that it was appropriately to publish the names
and addresses of anyone who owned a home that had poisoned a lead child. I am
done. Thank you.
MS. MILAM: Thank you. All right, Gwendolyn, take us home.
MS. FLYNN: Yes, and I am the one standing between you and lunch, aren’t I? I
will try to buzz along here. While my PowerPoint is getting loaded up, I want
to thank you for giving us the opportunity to be here and to share with you
from our perspective the data collection and how it is used and especially to
Dr. Green for making the recommendation that we join you here.
Let me start, I guess, by just talking about the initiative itself. We are
the African Americans Building a Legacy of Health, basically a collaborative.
We are a consortium of organizations that have come together to promote
nutrition and physical activity as a way of addressing health disparities. And
we are part of a national REACH initiative. Some of you may be aware of REAACH
US initiative that comes out of the Department of Health and Human Services and
the Center for Disease Control and Prevention. We are 1 of 40 communities
across the country that is involved in addressing the goal of elimination of
health disparities and the different racial and ethnic populations.
Community Health Councils REACH initiative is the African Americans Building
a Legacy of Health. We are working to reduce cardiovascular disease and
diabetes health disparities specifically and our population is the African
American communities of South Los Angeles.
Just to give you a little idea about Community Health Councils, we are a
nonprofit, community-based organization that promotes health and promotes
access to quality health care for disenfranchised populations. We are not
affiliated with a research academic organization or we are not part of the
Department of Public Health.
To give you a bit of context about the South Los Angeles and the population
that we are servicing and are providing interventions for I am just going to
give you a smidgeon of some of the geographic and population characteristics of
that area. We are part of South Los Angeles. We are nestled within the City of
Los Angeles. South Los Angeles makes up approximately 100 square miles in the
area. It is a population of about 1.3 million people. In terms of population
size we are about the size of Montana, the State of Montana. You can imagine
how dense the population is. South Los Angeles is comprised of about 64 percent
Latino, 30 percent African American and you can see the rest there. But South
Los Angeles has the highest concentration of African Americans is why we
decided to focus there.
This is what we look like on the map, the yellow area, is primarily 13 or 14
different — I guess it is more than that that is showing there, but we have
shrunken it to 13 zip codes that we are focusing on right now and it is all is
comprised in that yellow area except the little corner on the right there that
was North Long Beach. We decided that our resources would be better spent in
the South Los Angeles region. And the red area is the area that we use to
compare any of the research studies that we do. It is an area in West LA that
has better health outcomes.
In terms of our social economic, the economic make up of South Los Angeles,
it has two different extremes. It goes from very — let’s say middle class,
$75,000 to $95,000 annual household earnings to very poor, $20,000 a year. The
average household income would be about $48,000 I guess.
Now you have a bit of a context about the population and the region. I will
give you a little sampling of some of the data that we do collect. It basically
falls into the categories of our health outcomes and this is an example here
showing the adult health outcomes for geographic areas comparing South Los
Angeles with West LA and with LA County overall. And you can see that generally
where cardiovascular disease and diabetes and obesity are concerned we are
about the sickest in the county.
We also look at our community resources, the infrastructure of the
community. And this is an example of some of the data that we collect regarding
that. And this is looking at the food retail outlets comparing South Los
Angeles with West Los Angeles. And you can see that where full-service grocery
stores are concerned we are lacking. There is 6 percent in our area compared to
18 percent in West LA, again, where they have better health outcomes.
In terms of the sources that we use for our data it crosses a wide variety
of sources, our primary data sources, a lot of which has been involved in
collecting data on the resource environment and we have had to develop those
tools to collect that data because we didn’t find much of it available just in
the public arena. We created some of these assessments and surveys and you will
see them listed there, the healthy food assessment and on down.
Earlier in our iteration of the project we have been doing this for about 10
years. In the early days we were involved in collecting data on individuals and
we developed these physical activity participant surveys or wellness
We also collect data from secondary sources. The California Health Interview
Survey or we call it CHIS and the BRFSS which we look at from a regional as
well as a national level. And then the REACH US has its own survey that they
administer as well. We rely heavily though on the LA County Department of
Public Health, their Office of Health Assessment and Epidemiology.
Basically in terms of how this data has helped us and how we are using it,
it has helped us in this way. We operate from what we call our Community Health
Councils Model for Social Change, and that model is a systematic way of what we
use to address these health disparities and it is the identification of a
catalyst and in this case it is the disparities in health related to
cardiovascular disease and diabetes that becomes a catalyst.
And we convene a coalition and that coalition is comprised of a very
broad-based cross section of folks representing multiple sectors and multiple
disciplines in the community. We have faith-based organizations involved. We
have the university academics that are involved in our collaborative. We have
city officials representing city planning especially now in the work that we
are doing with community redevelopment agency. We have public health obviously
involved in our collaborative and those folks comprise the coalition and the
consortium that I mentioned earlier to come together and get some agreement on
exactly what the issue is we are going to study and try to address and get some
agreement on our values that we will use to address it.
And then we do the research which becomes very key to the work that we are
doing and using the data sources that I mentioned earlier. The research that we
use basically informs our policy recommendations. We come together to decide
now that we have some agreement on what the issue is and how we go about this
and we collect the research on it. Now we want to develop some recommendations
on how we might go about addressing this in the way of policy development. The
research becomes very key in that process.
Once the policy is developed, we create policy briefs as an example here.
Our latest one was the food desert to food oasis promoting grocery store
development in South Los Angeles. We then try to advocate for the
recommendations to be implemented at the various levels of government or
wherever institutions. And then we monitor and evaluate that.
Some examples of what we have been able to do using that research is the
creation of zoning and land use strategies to improve health ultimately. Here
is a listing of some of those ordinances: the Fast Food Interim Control
Ordinance which was developed to really control the growth of fast food
restaurants in our oversaturated South Los Angeles area. We are working to
influence community plan updates that are happening. And then some other
initiatives that we are working on for the future is the creation of Fresh and
Healthy Food Enterprise Zones and the list goes on there.
Some of the challenges that we are facing with the use of data is finding
available data by race and ethnicity. Having health outcome data by race will
help to track the impact of the interventions or the policy recommendations on
health disparities. There seems to be a lack of coordination of data collection
time. Some data is collected annually. Some is collected in 2-year cycles and
that data is usually — sometimes it is aged. I was excited to hear what Todd
had to share in terms of this one-stop shopping in data seems to be available
to us a little bit sooner. And finding data on the resource environment is
another challenge that we faced and I think this is something that has been
mentioned as well.
And what we would like to have are stats that would measure the resource
environment, for instance, being able to ask the question about how long does
it take you to get to a grocery store, or where is your nearest grocery store.
That would be one of the questions that I would liked to see asked in some of
the surveys that would help us to be able to track the changes in the
infrastructure over time and ultimately the change in health outcomes.
The timeliness — I think somebody mentioned that. Having greater
flexibility. I guess that speaks to the level of data information at the —
people talked about the granular level and that is basically what that is
And then funding which I don’t think this committee has any responsibility
or influence over. But we are always strapped for funds. Everybody is. And some
of the funding that we receive is only for 3 to 5-year intervals and it usually
takes a lot longer when you are talking about making policy change and
certainly trying to change health outcomes that are based in tradition and in
individual, cultural influences. It takes a bit more time than 3 to 5 years.
Some funders are getting at and are looking at longer terms which we really do
That pretty much is it. And I guess I would like to leave you with these
important messages and I think these are some of the take away messages I would
like you to have is how important data is to development of policy. It does
give the credibility that is needed especially when we are talking — we are
talking to different audiences. We are talking to community members,
policymakers and funders and having data really does become a tool for that
policy advocacy to make change at the institution and community levels.
The other message I would leave is the importance of including community
members that are impacted by these different diseases, these disease conditions
that they should be involved in the design of the survey and give input into
that, and the importance of having race and ethnic data that addresses health
disparities so that we can be able to track the changes over time by the
populations in which we are studying.
That is my time, I think. I look forward to answering any questions that you
might have if I possibly can.
MS. MILAM: Terrific. I want to first thank the panel. I know I have learned
a lot and I can see that a lot of us have taken notes and have a lot of
questions. What we have before us is a half an hour to hear more from this
panel. I would like to suggest that our committee members ask one brief
question so we can get through as many people as possible to quarter ‘til
and then we will mix it up a bit.
I understand that a couple of our folks had some questions from the last
panel and didn’t get a chance to ask them. Is Dr. Sondik still with us and do
you have a question?
DR. SONDIK: It is a hot topic, but I am telling you I feel like I am back in
New Hampshire where I was over the weekend.
This is really very eye opening I must say. I think the juxtaposition of the
two panels is very insightful and I congratulate whoever put the panels
together. I think it is really terrific.
The area that I wanted to ask about is not really on getting more granular
data because actually what you have shown is that you can get more granular
data. It costs money and it comes from different sources, the money that is
from different sources. And especially between the two panels we have this
tremendous range of resources — of sources I should say where the information.
But what seems to come out in all of it is how to use the data, the training of
people. I just wanted to get your reactions to what you see whether there is a
training issue. Matt Stefanak, in fact, mentioned that to me is what I heard
you say. You have BRFSS data and you can’t analyze it without somebody having
the requisite training.
But I wondered if you could just give your reaction to training needs and do
you think there is a training need on the part of staff or resources at your
location or whether you think that things are actually okay.
MS. BROWNING: I would just say in an agency our size with 63 people
everything from public health nursing services, environmental health, animal
control I will tell you that our resources are not there with the level of
training. We would need to have people be able to really do a good job at
looking at data, being able to interpret it, seeing if it is relevant, how it
applies, explain it. They are smart enough to do that part of it, but when you
get to the point where somebody starts asking very specific questions, you are
working with neighborhood groups. They start asking questions. A lot of our
folks are going to hold off and look to one person in the department who I
would say has that level of expertise. And when you are so stretched so thin I
think more boots on the ground with a better understanding is going to make a
big difference in our being able to be out in the community. One person can’t
be everything to everybody in a department.
And then also with the community itself when we are trying to teach — we
are trying to give people a tool to be able to act on their own self interest,
take charge of areas within their neighborhood. A lot of those folks have no
training at all. It is definitely challenging.
DR. MAIN: From the Colorado side in terms of our project I think the issue
is not the training of people per se, but I think it is relative to funding in
the sense that there are a lot of data requests that come fairly quickly and in
the ideal world we would be nimble enough to be able to deal with that, tackle
it, get data back out as quickly as possible. We have the people but it is sort
of being in the back of the line. We are not quite at the front of the line in
terms of being able to always get what we need in a timely way so that it is
I think with community that is one of our big commitments. We want to be
able to get the data back out fairly quickly otherwise it just feels like any
other activity where as academics we sit on our data and don’t get it out, and
I think our goal as a community is to get it out there and use it. For us it is
a little different. I will let Janet expand.
MS. MEREDITH: Just to add to that, we rely really heavily on the
researchers. For instance, Debbi’s group at the university has not only analyze
the data that they developed but they have analyzed all kinds of related data
down to our neighborhood level, census data, all kinds of things in order to
inform their study.
As an organization, that is one study. We are working on other issues where
we may or may not have access to those kinds of academic resources and over the
long term we need a way to be able to combine. There may be some new kinds of
data that are available that we think would make a really incredible overlay to
inform our communities and yet in house we don’t yet have the resources to do
We also have access to incredible assistance. We are getting great interest
from medical students and students in school of public health helping to
analyze and pull together data for us, but our ability to coordinate that data
so that it is useful in all a variety of ways is really limited. The resource
to even have somebody in house, that is really trained in how all of this data
works so that we can connect it would be phenomenal.
MS. MILAM: Ohio or California?
MR. STEFANAK: I think we are in pretty good shape as far as having a
workforce with the right skills to do most of the work we need done with
community health datasets. A year ago that wouldn’t have been the case. I have
lost 15 percent of my workforce over the last 3 years and a lot of the folks
who were quantitative and literate, got deployed to more direct service or
communicable disease investigations, but we are on the rebound now. We are able
to carve out time for people to work with some of the home-grown data that I
I think the county-level BRFSS example I gave is a special case and it
probably doesn’t make sense for — I can’t make a business case for getting
somebody trained in acquiring the software to do that kind of data adjustment
for a periodic survey like that that we do every 5 years. I need to go to a
person somewhere in the academic community or in the state health department
and we just don’t have that right now. I don’t know if that is a special case
in Ohio or if it is common place to other states where local health departments
try to do the same thing.
MS. FLYNN: So Community Health Councils is lucky enough to have a research
data manager on staff that was hired to work primarily on the project that I
talked with you about, but has been enlisted to work across the agency and I
didn’t mention the other two policy areas that we work on at CHC. The policy
area that I was focusing on has now five members on its staff and one of them
is the research data manager that is not nearly enough.
We are trying to build our capacity to become expert in the GIS mapping
process and that person is now being trained on that as a person to train the
trainer and we will be training other staff members. And we are hoping to build
the capability of making GIS mapping available to community organizations,
other community-based organizations within our area so that again it would give
folks a little more involvement at the community level for developing the GIS
mapping, but we could use that for other data collections, but we would need
staffing to dedicate it to that purpose. Funding would help support that and
once we got that and another staff person and then the training would follow.
MS. MILAM: Thank you, Gwendolyn. Walter, did you have a question from the
prior panel? Were you one of those folks?
DR. SUAREZ: No, I think I am going to reserve for the HIE one next time.
DR. FRANCIS: This may not be an accurate observation, but I didn’t hear a
lot of talk about clinical records in any of your discussions. And I just
wonder whether it is — Stephanie used the technical term blob or whether it is
actually the case that when you have consulted with community groups and asked
them to help formulate the questions that you are interested in that just
hasn’t come up as much or whether there have been questions about privacy or
something else. My observation may be inaccurate, but the kinds of datasets
were not HIE data that I heard people talking about.
MS. BROWNING: I would agree with that. When you are working with these
community groups, we are not hearing them ask questions about health
information, exchange-related data. We hear things about — there is a lot of
people in our neighborhood that don’t have a health care provider, things like
that, but the things that we are hearing more about are many of the income
about transportation, about access to fresh food. We are not hearing those
MS. MILAM: Anybody else?
MS. STEWART: I am just going to say in our communities it is a given. You
have a conversation about who has what illness, but that is relatively
irrelevant when people start talking about how they got there because they are
not so much interested in — they will give you their diagnosis very easily,
but that is like the tip of the iceberg. The real conversation is how I got
there and what do I do now that I am there? When you talked to people about
their personal motivation for being involved, that is when you will hear about
their self reporting on their clinical diagnosis.
DR. GREEN: Could you comment about how you sampled once you decide to do
primary data collection?
MR. STEFANAK: I outsource that. We have a data research center at Kent State
University which is College of Public Health. The sampling methodology is
something the contractor handles for us.
DR. MAIN: We basically — thanks to using Census data we got down to
basically randomly doing a stratified random sample at the block level. And
then within blocks we randomize. We actually had community recruiters on the
recruitment phase for our household data collection and then we had a — it was
a BRFSS-like call center collecting the follow-up data and they literally
followed a sampling plan up the side of the street. You get to a house, which
direction do you go, things of that sorts. It was actually pretty
sophisticated. They actually even were trained on random table of numbers. When
they came upon an apartment building with five floors, where do you start? We
really did fairly systematic data collection and then collected data around the
household sample for the survey so audit, sort of neighborhood data around the
MS. MEREDITH: And added just a detail with our — we followed up with a
kid’s obesity study and it took almost 9000 household contacts to get 250
completed surveys because of age group, because of people not being home for a
whole variety of reasons. It is a lot of work.
MS. FLYNN: And for us we work with the university, I mentioned USC, and have
called on their expertise who sometimes call on their master students to help
us with some of the sampling. We also talked with and enlist the individuals
who are part of our coalitions. I should say that our coalition members in
addition to the stakeholders I mentioned we also are comprised of folks who are
directly impacted by these conditions. They have been part of the process from
the very beginning, the identification of the issue all the way through to the
advocacy for the policy change and that includes involved in the community
research that goes on and have been involved in the sampling as well.
DR. LAND: There are basically three different types of data presentations or
access to data. One is community indicators, standard public health indicators
that somebody — low birth weight or whatever, which is fairly inexpensive to
produce and generally available.
The second type I would classify as access to data files that may be purged
of identifiers but you have the individual records. It requires somebody with
analytical capability to go in and analyze the data.
And the third type would be — we just have been talking about of actually
going out and collecting primary data. Of course there are incremental costs
associated with accessing each of those types of data.
From your experience which has been the most valuable to you in terms of
public change, in terms of community intervention to just have the first level
which is fairly inexpensive. Does that drive change and is sufficient for your
needs or is it the second level or do you really have to have a third level?
MS. MILAM: Let’s start with California and work backwards.
MS. FLYNN: I think I see the value in all of those levels. We are basically
involved in trying to transform the infrastructure of a community and having to
identify a policy champion in an elected official and provide them with the
data that they need in order to build the case that they could use for their
colleagues to get through implementing a policy I think requires all of those
levels of data to be collected, the primary, the secondary especially looking
at health outcomes, looking at the resource environment. I think it requires
all of that.
MS. MILAM: Anybody feel differently?
MR. STEFANAK: I just think the second and third levels of data have been
more potent for me in affecting change in the community. Disease reports, child
lead poisoning cases, and the example I gave. It was only when we mapped that
by city council award and not only the cases but the homes with remediated lead
hazards that we finally got the city government and Youngstown off the dime to
hire a special prosecutor to take on these landlords.
The third level, where you go out and collect data. I can think of several
examples where that has been very effective over the years for us. Think of the
so-called tobacco compliance checks that health departments have done for many
years where we send kids, 15 and 16 year olds out to try to buy tobacco. We
take a sample of tobacco vendors. That was very eye opening for people in the
community to see how accessible tobacco is. When you publish the names of those
tobacco vendors and there is a debate about that too because you don’t want the
kids to go that is where you can buy cigarettes. We use shame a lot in
Youngstown as you can see. Those kinds of tools are very potent for advocates
MS. STEWART: I definitely want to concur with Los Angeles. I am an advocate
by profession. I don’t know where eyewitness testimony falls in. But that level
of story collection and qualitative analysis makes a big difference in Colorado
especially if you can pull together constituents that are saying something
contrary to what their legislation representative might believe.
MS. BROWNING: I would echo what you are saying. And the story telling part
of it is if you can use the data to frame the issue, to create a picture that
people can understand and you can have somebody that gives you that personal
story, it is amazing what you can do in the policy environment.
MS. MILAM: Sure, go ahead.
MS: MEREDITH: Can I add to that? I think that it is great that the
qualitative piece is coming up here because we find it particularly with
certain issues and I would say that mental health is a great example of that.
Quantitative data are never going to tell the story and they don’t tell you can
drill down into what really happens to people whether they have insurance or
don’t have insurance, what happens in their lives, what happens to their lives.
You are not going to understand the real importance of some of these health
issues in communities.
MS. MILAM: Other questions from the committee? This has just been fantastic.
We want to thank you so much for joining us and hope that you will continue to
be with us through the day. It is 10 ‘til 1 now. We will break for lunch
and we will return and resume promptly at 2. Thank you.
(Whereupon, at 12:50 pm, recessed for lunch.)
A F T E R N O O N S E S S I O N
AGENDA ITEM: Health Information Exchanges – Leslie
Pickering Francis, Moderator
DR. FRANCIS: My name is Leslie Francis and I would like to welcome all of
you back here to the third session of The Community as a Learning System for
Health: Using Local Data to Improve Community Health. This session which will
go for an hour and a half, although I recognize we are already 5 minutes late,
we will put into our break in discussion — this session will go for an hour
and a half. It is on health information exchanges and we have three people
here: Patrick Gordon from Grand Junction, Colorado. Patrick, are you on the
MR. GORDON: Yes, I am.
DR. FRANCIS: Great. Nance Shatzkin from the Bronx, New York RHIO, and Greg
Larkin from the Indianapolis, Indiana IHIE. I think we will just go in that
order and get started. I think the idea is about 15 minutes for each. Patrick,
if you start to run a little long, I will let you know verbally.
MR. GORDON: I just wanted to be clear; can you guys view the PowerPoint
presentation that I forwarded?
DR. FRANCIS: Yes.
MR. GORDON: Okay, great. Thank you all very much. It is just an honor to
join you today and to talk about this very important issue. We are very honored
actually to be a part of a national program sponsored by the Office of the
National Coordinator in which communities are aggressively pursuing the
development of health information exchange not for its own end, not as an end
in itself but rather actually to realize the triple aim, rather to realize the
value of the investment in the health information technology that is underway
in federal programs.
Within the Beacon Program there are several different models being tested
and developed, more traditional, integrated delivery systems that have a
corporate organizational focus. The other end of the spectrum where we have
community-based programs really partnerships and collaborations among multiple
independent and sometimes competing organizations to advance the idea of health
DR. FRANCIS: Was that slide one? Should we have moved the slide up?
MR. GORDON: Go ahead and move to the first slide. While that is well and
good — while there is a heavy technological focus to our project ultimately
this is about improving outcomes. This is about reducing costs, changing the
paradigm of the current health system, which I would characterize as basically
a system in most cases where decision making if it occurs on a community basis
is driven by administrative data, claims data, eligibility data, datasets with
severe limitations in terms of reflecting the actual needs of patients and the
actual health status of the population. It is driven by a top-down managerial
approach, again, where interventions are done at best on a broad population
level and sometimes not executed in a way that creates actual value or health
impacts for patients. And last but not least of course we have a system that
pays for volume, not for outcome.
Beacon at least in our case and I think in most cases is about moving away
from that paradigm to one in which clinical data aggregated through health
information exchange drives more of the decision making and in fact really
plays a leading role in that decision making around resource allegation and
intervention development at a community level. But more importantly it empowers
individual care teams and patients to improve health themselves.
Again, our ultimate vision is the model that you see in front of you here.
Many of you will recognize this. This is the chronic care model or now referred
to as the care model developed by Ed Wagner almost about 12 years now. But it
really is an excellent template for how a community can organize itself around
delivery system designs and decision port systems in clinical data to support,
inform, activated patients and prepared pro-active practice —
Primarily in primary care settings and in our particular project there is a
heavy focus on primary care development of medical — and so on, but beyond
that also we recognize that medical columns can only be successful in receptive
medical neighborhoods. Connection with the community is never lost.
This next slide — you will have to excuse the art work. It is my own. But a
lot of times when I begin talking about this stuff, it sort of occurs to folks
as an alphabet soup of federal programs, federal initiatives, different
concepts and ideas that are being advanced around the theme of accountable care
across the country and in practice and trying to actually implement and
integrate all of these various concepts, programs, and services. Oftentimes it
just occurs like a soup particularly to primary care providers who are already
in many cases overworked at the limit of their capacity and not necessarily
In our Beacon project we have been tempted to as clearly and cohesively as
possible integrate all four domains of a successful community-based health
system integration effort in one place. Of course there is a technology
component and that really includes both the adoption of electronic health
record systems that are actually suited for meaningful use that can store and
report code clinical data and other data about patients. But beyond that then
of course development of the health information exchange which is first and
foremost an ongoing set of interoperability services connecting various and
diverse record platforms up across physician practices, hospital practices,
public health, community and mental health centers and beyond.
And then of course aggregating that data in a way that can be used to
support powerful new analytics that really are — I think we are just at the
cusp of being able to unlock new technologies that will have a very powerful
impact on population health management if we can bring this together correctly
that is necessary but not sufficient.
Of course in practice settings we need to promote to keep the team-based
concepts. We need to find new efficiencies. We need to enable physicians to
practice at the top of their license and we need to grapple with the shortage
of primary care. Finding new efficiencies and new work flows that support the
technology and yet prove quality is fundamental to our work.
Within our program we have adopted a very robust practice transformation
program working across up to 75 diverse primary care sites over the course of
the project. I will get into that a little bit later.
We have a quality improvement focus and also a payment focus. I won’t
belabor those, but you cannot advance this work unless you have both of those
focuses within the program.
This next slide really details all of the various services that we are
offering through this. Again, I won’t belabor each of these points. Every
single element on this slide is reflected in our Beacon program and our
practice transformation teams are actually bringing this together in process
through a learning collaborative approach.
We also recognize that incentives are fundamental to success and we mean
financial incentives in our case. In our collaboration in Western Colorado we
are fortunate to have a not-for-profit, independent health plan participating
with us in this work and that health plan has put up three-quarters of a
million dollars to support incentives for practices that undertake the Beacon
curriculum that achieve process improvements and meet certain milestones and
then most importantly meet measured outcomes over the course of our project.
Again, we are working across a very vast geography in a very diverse
community, actually a collection of communities. One of the major themes in our
project is that we are really trying to leverage local leadership and empower
local leaders to own the project and to act as autonomously as possible while
again promoting the overall objectives of our program which are aligned with
the triple aim. Again, in a future slide here I will show you how diverse our
geography is but we thought of a way to make it clear that this is not a grand
junction thing. Grand junction itself as a community has received recognition
for work in this area over years, but really trying to replicate and expand
some of the concepts that play in that community to the surrounding areas.
Collection of rural and frontier counties in Western Colorado. This is a
vast geography — several New England states, quite frankly. It is only 300,000
people. Lots of space and lots of space in between people in this geography.
But the health information exchange is operative across this entire territory
which not coincidentally reflects the Dartmouth Atlas’ current measurement
region for hospital referrals. We thought that was important from the
standpoint of having a credible distant set of Medicare baseline data for the
purpose of our work.
This next slide is just an example of how we are trying to engage our
participating practices on the concept of team-based care and the fact that
team-based care can promote measured improvement and quality. It is sort of an
aggressive and creative lack of a better word marketing campaign to recruit
participants and reinforce the concepts that we think are valuable.
Just to give you a look and feel for our project from the standpoint of
primary care participants anyway. We have a total of 94 practice sites in this
region. As I mentioned earlier, we intend to recruit up to 75 of them by the
end of the program. We are already a third of the way there. Participants thus
far have been early adopters. They are the ones that typically lead the charge
on most changes like this. We are already making progress into sort of the next
rung of participants as well. It is quite a lot of diversity at least from a
primary care standpoint. And of course just being a rural area in many regards
in underserved and low-income area we have rural health centers and federally
qualified health centers participating I projects as well.
We are trying to be as flexible as possible. There is not a big bang
approach to this. That wouldn’t work if we wanted to engage as many different
participants as we intend to engage over the course of the project. We have
basically adopted a quarterly cohort system where groups of 5 to 10 practices
will join in cohorts when they are ready and begin the practice transformation
and quality improvement programs. However, we conduct region wide learning
collaboratives along the way. Wherever a cohort is in process we bring them
together for a learning collaborative and really try to promote the value of
team-based learning and even friendly competition.
Last but not least and I know we are short on time so I will not get into
detail here. Our measures for this project are for the most part reflective of
the meaningful use measures that were published in July of last year. There are
issues quite frankly with the specifications for many of those methodologies
and practice. There are issues in trying to make them work on a community basis
and the way that we are attempting to do. However, we think it is important the
extent possible to align our efforts with the incentives and the policy
framework that is out there and really quite frankly work through some of these
methodological issues so that again we can really achieve the vision of
meaningful use which in shorthand is the vision of the triple aim when we look
at the measures: cost, quality, population health are all reflective in this
I realize I am pushing time and I will stop talking and be available for
questions if you have any at this point or later in the program.
DR. FRANCIS: Thank you very much and we will indeed be having questions in a
little bit. Our next speaker is Nance Shatzkin from the Bronx.
MS. SHATZKIN: While I am waiting for the slides to get up, I will just start
taking advantage of every minute here to tell you a little bit about the Bronx
RHIO. New York State as many of you may know really fostered the development of
health information exchange within the state by making grants available, the
first of those in 2006. These are the HEAL grants and the Bronx came together
early with great interest in applying for one of those grants. The Bronx is an
interesting health care community of the 1.4 million people who live there.
Eighty-five percent of the care for those people is delivered within the
Another interesting factor is that 85 percent of the physicians in the
community are employed by one of the major health care facilities in the
bureau. It was most of these facilities that came together to form a
collaborative or a not-for-profit organization called the Bronx RHIO to begin
to work towards health information exchange at a time when I have to say nobody
really understood exactly what it is was.
Now I was able to write a successful — this is the nature of the people who
came together to create the Bronx RHIO. I realize when we talk about the Bronx,
we think we are dealing with a really sick community and then you come to a
meeting like this and everybody puts their statistics up. But I will mention
that in a recent study completed by the University of Wisconsin Population
Health Institute and the Robert Wood Foundation a county-by-county health study
of the State of New York sadly the Bronx ranked last among 62 counties and
primarily that included worst in the state for socioeconomic factors that
affect health such as unemployment, poverty, street crime and low education. We
are dealing with a fairly sick population.
The member of the Bronx RHIO include four major hospital centers, one of
those is the James J. Peters Veterans Administration Hospital and there are
then in addition to that several nursing homes, home care agencies, community
health centers, most of them federally qualified health centers. There may be
one I can think of offhand perhaps two FQHCs in the bureau that are not
currently members of the RHIO. I don’t think it will take too long before they
will join, but the vast preponderance of health care providers are members of
the health information exchange. It doesn’t mean that they are actively using
it yet or sharing their data, but they are at the board room table. They are at
the committee room table engaged in the discussion.
I think we are one of the few health information exchanges that has actually
gone to our department of health and said here are your user credentials. Use
the data. I have to admit to a certain disappointment that they are not using
it as much as I would have thought they would be, but I think that will come in
The prison in the Bronx has just finished installing its own EMR, electronic
medical record, and I think that has distracted them a little bit over the last
Currently within the Bronx we are able to share a considerable amount of
data, demographics, all lab results including pathology and microbiology. I
think I highlight that because this is the second generation of a RHIO in the
Bronx. We went live with our first system in 2008 with our second platform in
2010. I have been working in IT for a long time and it is my experience that
second generation systems are always better. We couldn’t do the microbiology
and pathology in generation one of the Bronx RHIO. We are very pleased to be
doing it now. We share radiology reports, cardiology reports, and some basic
data about each encounter.
While allergy and diagnosis information is also available, we find the
quality and inconsistency of that data from site to site makes it difficult to
raise it to the patient level which is the level at which you would think about
those data elements. Allergies are obviously not encounter specific. But
because of challenges and the way that data is being recorded, managed, the
quality of the data in some cases, we actually have chosen not to elevate it to
the patient level while we try to get the community to take a more aggressive
approach towards making it cleaner and better data.
We are sharing in-patient medications now. Some of that comes from the
interest of our members, home care agencies and long-term care settings which
are really concerned with having the full inpatient medication history from an
And we are adding data as aggressively as we can that includes discharge
summaries which are critically important. It is hard to share them
electronically when they are still being written on paper even in hospitals
that say that they have information systems, but that is the reality.
I think a lot of the drive for the data that we will be adding to the RHIO’s
exchange will be driven by what helps most in care coordination and in
transitions of care. I think that these are two of the areas where we see that
the health information exchange can have a powerful impact and so I think that
will drive a lot of the data. I for one, however, am also very interested in
getting EKGs in because I have had enough emergency room doctors tell me about
the admissions they could avoid if they only had a baseline EKG to compare to.
This problem though is when we talk about the things that get in our way. There
are lots of things that get in our way for getting that data.
The Bronx RHIO offers two major services through its health information
exchange. One of them is the web-based portal. Come to it and the data is there
for you is well organized and is well presented as we are able to manage. We
call this our Virtual Health Record. It is difficult to get people to step out
of what they do all the time to go log in somewhere else and look at something
online. I talk about muscle memory. I think it is very difficult to get
clinicians to change their muscle memory.
We also offer another service, a messaging service, a subscription service
now commonly being called subscribe and notify, but what we offer is that
clinicians can subscribe to patients of interest and if that patient is
registered to an emergency department admitted or discharged in an emergency
department or in an inpatient setting, we will push the message to that
provider. And depending upon their system savvy we will push it either to their
own EMR in the form of a message that their EMR can accept transforming it as
we may need to transform it for them to be able to accept it or we will give
them a mailbox where they can receive those messages from us. I have to say
this is one of the most effective offerings in terms of the uptake of the
providers in the community wanting to be able to use rather than me trying to
cajole them. They see the benefit of this quite quickly.
The home care agency immediately acknowledges the benefit of not sending the
nurse to an empty house as well as the ability to follow up on a patient’s
visit to a provider and be able to make sure that the discharge instructions
and follow-up care are being followed.
In addition to that I think perhaps and especially having been here today
what may well become the most important thing we offer is an analytics
database. This database which we are just beginning to work with and which I
think will really come into its own in the second half of 2011 will offer us
data marts of all this rich clinical information that we are collecting. We
will be able to make that available to all of our members as well as be able to
do our own analytics with it. And when you think about the power of being able
to layer on top of that evidence-based medicine tools that are now able to look
at a comprehensive chart on the patient and deliver evidence-based medicine
alerts either again into clinician’s mail boxes or into a portal from the RHIO
and began to see the opportunity to really affect the care.
The specific example of one of the things we are doing in the Bronx that I
wanted to share with you today has to do with the Care Transition Program.
Against this background we have a very sick, largely with chronic illness and
multiple chronic illnesses population in the Bronx. We know that there is a
high volume of potentially preventable admissions. We know there is a high
volume of readmissions and ED visits.
In response to this three of the major institutions and two of the important
health care plans in the community came together with the idea of tackling just
one issue. Readmissions within 60 days or patients from medical units
discharged home. And to see whether or not they could have an impact on that
transition of care from the hospital to home in a way that would improve the
performance and result in less readmissions.
When they came to us at the RHIO, it was just perfect. You couldn’t have
asked for a more perfect marriage between something that somebody wanted to
accomplish and a set of tools that you were building and wanted someone to use.
We were very excited to be able to join with them in an effort to build the
Bronx collaborative Care Transition Program.
What are they taking advantage of? They have built a program based on again
on evidence-based research. I think a lot of people are very familiar with
Coleman’s Project RED, but there are others around the country that have been
working on this issue and that have found that there can be a difference made
if the patient goes home properly understanding what is going on and if there
is enough follow-up support to make sure that there is access for follow-up
care, et cetera.
What we offer to this team is the ability to use the virtual health record
as a way of starting their own record. So the patient has been admitted. They
don’t need to reenter any of that data. They pick that patient up for their
study and they can immediately bring that data into the care transition chart.
And the care transition chart is actually built on a tool that the RHIO also
makes available. It is essentially like an EMR light.
And we have developed for them a number of specialized encounter forms that
allow them to do the initial patient interview to understand the care giving
situation. Where do they get their meds filled? Are they going to be able to go
home or are they going to go somewhere else, a sister, and a daughter before
they actually go back to their own home? It allows them to track every contact
that they have on behalf of that patient to do a pre-discharge interview and
then to be able to track their follow-up care as well.
And a key component of this supported by the system is the ability to take
those discharge instructions and move them into a patient discharge booklet
that will make those instructions readily understandable even to a low level of
education so that when you need to take your meds is very clear and what your
follow-up visit should be is very clear. If you need any follow-up tests or
follow-up appointments, they will be able to make sure that that happens, all
the reminders generated through the system.
I want to use my last 3 minutes to talk a little bit about, as long as you
asked, what some of the problems are that we face really in trying to do this
out there in the trenches. I don’t think we can underestimate patient identity
issues. I will tell you that we have a full-time team that works. We are the
generation of multiple medical record numbers within facilities. The ability to
create all of your Kenneth Chen’s and know that one Kenneth Chen is the same or
different from another one when there are errors in Social Security number or
no Social Security number.
The challenges within the systems that are implemented, the legacy
environment of our health care institutions which means that there are no
outbound interfaces or the OB system is not tied in to the rest of the clinical
information system, so that data doesn’t get out the door. Or the fact that the
Medicaid data in New York State, is still not available to us because state
regulations will not allow the RHIOs to touch it. There are a number of issues.
Now I wanted to be able to focus on the positive as well as the negative.
And the one positive I am able to come up with in the Bronx is a high
percentage of employed physicians does help us and does mean that as we make
inroads we can get an organization to move. We are not having to do it
individual by individual, but it was really hard for me to come with a positive
concept. I obviously like hard work. I love this work.
When you look at what it takes to accomplish what we are trying to
accomplish what we are trying to accomplish here, I have to say there was a
whole new for me today which is how do we energize the community we work in so
that the community is also working because it is so hard to energize the
physicians and the administrators that I am ready for a new group people to
work with. And from what I heard today they are out there.
But I think changing muscle memory is really hard and we need the help of
leadership to make that happen. We also need for state and federal guidelines
to align incentives to a push for standards and to really be paying attention
to the data collection process all the way along the line. And as many of us
have mentioned having a good analytics team there is no substitute for having a
good analytics team.
But there is one more issue I really need to comment on even at the risk of
going over my 15 minutes and that is the issue of patient consent. It was
somebody sitting along this line who said early this morning perhaps we do this
to our detriment. We worry about patient consent to our detriment and I am in
your camp sir. It is my experience that we have instead of removing costs we
have added a cost, the cost of consent collection is a huge impediment to the
sharing of data in the Bronx. We operate under the guidelines of the State of
New York. That means that a patient must give consent to access the data to
every provider or because they were afraid of every provider they said a
provider really means a facility. Every provider facility the patient must sign
a consent for those providers to have access.
The cost to those facilities to collect those consents and we have tried to
make it easier for them by saying put it in a user defined field in your
system, get it into your HL7 out and then that will update the Bronx RHIO so we
will know what the consent value is. But that is an ongoing cost.
And as we talk around the nation about making it more sliceable and diceable
and people should be able to go online and decide which doctor gets to see what
data, I fear. I see that as 1984 four times over in terms of holding us back
because in my experience actually a lot of these patients really don’t — they
almost expect that the data is already being exchanged and they want clinicians
and health care organizations to take care of their health in the most
efficient way possible. They do not want to have to make those decisions
I really see that especially before we have the tools to manage consent in
the way some would like to see it managed. We have in fact added a huge
impediment to the sharing of data.
And I can’t help but mention that two of the major platforms for health
information exchange are now owned by insurance companies. They have the data.
What will they use that data for?
I just have one more slide. I am over, but I will just say I think that the
Bronx is an interesting environment in which to be doing this work because it
is the kind of island that it is with so much of the care being delivered
because it is such a dull background against which there can be some shining
light. And I think that we are proud in the Bronx in spite of the impediments
that we face that we have begun to create some reproducible models that others
will be able to share and build upon. Thanks for much.
DR. FRANCIS: Thank you for hiding your light under a bushel. Our next
speaker is Greg Larkin from Indianapolis.
DR. LARKIN: I will try to add at least some energy to the presentation. I am
Greg Larkin. I am here to talk about Quality Health First, just a little bit of
background because this morning’s meeting was particularly interesting to me
because for several years I managed a corporation. I am a physician. I manage a
corporation’s health plan and we struggled for data to make some decisions.
Secondary to that I was in private practice and I always wished I had more data
to make decisions. And then I also worked for the Indiana Health Information
Exchange’s chief medical officer and started understanding how data can
expertly be gathered. Then most recently I am the state health commissioner for
Indiana. I am overseeing the public health care system which is in dire need
for active and good data. This morning’s discussions were very interesting to
I want to talk now more about the Indiana Health Information Exchange which
is the nation’s largest nonprofit health information exchange with over 80
hospitals and nearly 20,000 physicians. This is a quick schematic of the many
hospitals that are exchanging data with the IHIE as we call it and maintain a
rather large and dynamic database. And this just a schematic of the number and
the distribution of physicians that are using the clinical messaging service or
they call it docs for docs throughout Indiana. And as you well can see across
the state line because of care programs patients do cross lines.
The Indiana Health Information Exchange, its core database and its — Dr.
Overhage is here so he may know something about this. I don’t know. I feel like
an uncle describing the nephew and the dad is here in the room. The data
management, the resources for the database are broad and continue to expand,
not only physicians’ offices and outpatients and labs and public health records
and payers and as well as pharmaceutical benefit managers. It is a very robust
database and applying the secret sauce that Regenstrief and IHIE need to know.
They can generate a multiple use product that are, A, appropriately
confidential and private, but secondarily and more important a value to all the
end users be it hospitals, physician offices, labs, public health payers, and
researchers which leads me to the Quality Health First Program.
QHF was a program that started with actually in Indiana large employers and
therefore payers as well as research of Regenstrief and others were discussing
how can we better differentiate providers who render better care as well as
health patient care. Quality Health First is a program of multi-payer,
value-based reimbursement program. I prefer to call it a pay for performance on
steroids because it is really a program that is focused on patient care and
secondarily on engaging and rewarding physicians.
It does help physicians identify and prioritize health care screenings for
important health care goals that have been defined by community physicians. It
does enhance reimbursement for physicians because the payers, the insurance use
IHIE as their incentive program — excuse me, use Quality Health First.
It focuses on an array of chronic diseases for which there has been good
clinical data that intervention in screening make a difference. And in Indiana
our primary insurers are Anthem. Anthem is the large one and then there is
United Care, Unified Group Services. We also have Medicare and Medicaid data.
Within Quality Health First Program there are over 1300 physicians and these
are primary care physicians defined as internists, family docs, pediatricians
and OB/GYN and it covers over 900,000 patients.
This gives you a schematic of the growth of Quality Health First when it
really first started in 2009. And as we contracted more and more providers to
engage the Quality Health First Program obviously the number of patients also
increased considerably. This is a growth line of the providers. And the
questionnaire I completed prior to this presentation there are many complicated
issues with any program like this, but walking into a physician’s office saying
we have some measures and you should trust us. That really wasn’t an easy sell.
And over time the physicians have — in fact, I have been recently told a
rather prestigious reward in Central Indiana where the community nominates
medical programs for recognition. Quality Health First has been nominated with
physician’s support. It tells you we have come a long way.
Whether you look at structure, process or outcomes the Quality Health First
Program and the pay for performance design really covers them all very all. The
physicians basically can use Quality Health First as a high-quality registry
even if they don’t have EMRs because we have the data from all of their
sources. We have a high percentage of diabetics are having their H1Cs measured
every year. We can look at the claims data to get that. We have the clinical
data to also then identify not only those who get screened, but for those whose
control is reflected with the outcome of the study.
There is a couple of slides here that are really excited and exciting to me
because I have been away from Quality Health First a little bit over a year and
this is what we aspired to hope we can demonstrate. What these slides show that
the providers participating in Quality Health First over a period of a little
more than a year using the Quality Health First Program which we believe made a
more attentive to know which patients they had, know which ones need screening
and knew that their incentive pay was somewhat based on improved control. This
shows how the control panned out which we feel is an impressive trend for a
short period of time of the measures’ improvement.
There are other ones too and I won’t go through each one of them. Some from
significant improvement, some maybe no improvement, but overall we feel the
trend is impressive. Some of this is just complexity of getting all the data in
the system and making sure it is accurate, for instance, well-child visits it
took a lot of formula tweaking to make sure we were being fair to the physician
and to the families and give them credit for well-child visits.
This isn’t a small program by any stretch. It covers a large amount of
patients as noted here over 5 million unique patients statewide. And Indiana,
by the way, we all feel we are unique. A little humor there – as well as all
the cancer screenings.
The other interesting thing the program has created is a provider portal.
This program unlike most other programs allows physicians to reconcile their
data. If our data is incomplete or IHIE’s data is incomplete, the physician can
go in and correct that data. There is a host of rules that have to be applied.
They can also search on patient. They can drill down to any measure down to the
patient level. The way you buy physician credibility is you allow them to help
to find the measures and you help them prove what is valid and what is not.
There is a patient care view. When the physician is notified within the
Quality Health First Program there are 27 measures, 10 of them have incentive
dollars behind them. The physician can drill down to all of his diabetics, for
instance, in the system who have had A1Cs, who have not, which ones need
cervical cancer screening, mammography screening, and they can all do this
online now. Initially we did it on paper which was a temporary fix for a
solution which we knew would be electronic.
Many available reports. The slide here is too small, but it will tell you
that the physician can go online again and do the analytics within his practice
or within his group practice.
An important report the physicians receive when they participate in Quality
Health First is this report where if the physician has enough patients in his
or her practice to quality for the incented measure then the metric is graded.
And this particular slide has a lot of information, but what it does tell the
individual physician not only which number of patients — how many patients he
or she has in the practice that qualify for the measure, how many are outside
of the performance metric and helps them then identify how to call in those
But it also does something again which is very physician acceptable. Instead
of just giving the physician a score saying well, doctor your score mammography
is 60 percent and the community’s score is 75 percent, the physician’s pushback
has always been you don’t allow for those noncompliant patients and believe me
I have more compliant patients than anyone else in the city because I happen to
have 20 percent plan A and 40 percent plan B. And we all know that 20 percent
plan A are noncompliant or whatever that might be.
The adjusted peer score compares the individual physicians’ practice on that
particular measure. It is the adjusted peer score that takes the same plan
percentage mix that physician has and overlays it on the community as if
everyone in the community had the same plan mix. No longer can a physician say
that somehow I am being unfairly compared because everyone else has 100 percent
commercial and I have 100 percent Medicaid or Medicare or whatever the case
might be because the peer score is adjusted to the same mix and you need that
The Quality Health First Program dovetails well with the Beacon Community
Program. It has a very powerful online reporting tool. Now going into a
cardiology specialty what I talked about previously the 27 measures are
primarily ambulatory primary care from cancer screenings to immunizations. It
is clearly a statewide growth. The payers have insisted on that because they
did not need a two-county solution. They want a state solution because their
customers were statewide.
It is demonstrating affordable and cost savings in a meaningful way by the
demonstration of not only the improved metrics, but also if you look at docs
for docs the other sustainable and reduced cost. And it is sustainable because
it is growing as our slide showed that we had more and more payers, more and
more providers are being enrolled. And the vision of reaching the critical mass
seems pretty much coming much so to a reality. I think the main demonstration I
have an opportunity to see now my role in public health as well as my previous
role. This is one of the most expansive and ingrained solutions we have seen. I
am pleased to talk about it further as anyone else in this room might know
something about it. Thank you.
DR. FRANCIS: Thank you very much. It is now time for us to begin to engage
in discussion. I haven’t been scanning for hands, but I know Walter reserved
his — please go.
DR. SUAREZ: First of all thank you very much for the testimony. I originally
wanted to ask this question to all the panels, but then I decided to reserve
the question to the last one. I think clearly one of the most important aspects
of all of what we are doing with the adoption of EHRs and all these initiatives
at the national level as well as the pursue of health information exchanges
certainly involves public health significantly and to try to create a much more
dynamic exchange with public health bidirectional as well as real time almost.
But the biggest concern I have had all the time and I have expressed this a
number of times is that while we are investing over $20 billion in pursuing
that option of electronic health records across the country, we are forgetting
completely about the infrastructure in public health, the health information
infrastructure in public health. And we are at risk of creating still and I am
more afraid so now than ever what I call a digital divide between clinical care
and between all the systems across the country that are adopting electronic
health records, adopting very relatively advanced and sometimes variant ones —
like SNOMED and transitioning to ICD-10 for a lot of the condition reporting
and adopting very complex transaction standards like HL7 and others whereas in
the public health side and I am concerned again that the infrastructure is not
there today and with all these advancements it will be even less likely to be
capable of supporting all these exchanges.
From your perspective in the Beacon community side in sort of the advanced
communities adopting and implementing these approaches, what is your sense
about that digital divide and that risk? Is that something that we should be
really very concerned about and should be pursuing some mechanisms to help the
public health community improve their public health information infrastructure
to be able to communicate electronically with the rest of the world?
DR. LARKIN: I would be glad to take at least a crack at that. I think you
are absolutely right. With my most recent career change as a health
commissioner seeing the infrastructure within Indiana which is I think typical
of almost every state, there really isn’t much of an information
infrastructure. One of our goals in the State of Indiana has been identifying
all of the silos within the state department of health that maintains important
population data. Each of these silos has been created by a singular grant for
whatever that reason might be and all departments of health run on grants.
And for instance we have 150 different grants that collect information one
purpose or another. And we had our own little problem within the department of
health, but we are now actively working with IHIE and other state HIOs to try
to start directing some of that information into the HIOs but when you get down
into the local health departments and if you are a home rule state, which
Indiana is. We have 92 counties; therefore, we have 92 local health departments
each under its own governance structure, each under its own priority and
budgetary issues and each well underfunded. The infrastructure just simply
A good example of that is many local health departments are important to the
community for providing immunizations. When you provide immunizations but you
don’t have a good infrastructure, you end up giving probably redundant
immunizations because people don’t know what they have gotten. Or in Indiana
had a rule that in high school this last year students had to verify complete
immunizations. We had schools have to shut down because the records were so
incomplete. And some private providers provide immunizations and some don’t.
Our big lift for us now is to take the state registry which is pretty cool and
try to plug that in to the HIOs. There is a lot of potential digital divide as
you phrased it which I think very appropriate.
MR. GORDON: This is Patrick. I would just quickly echo that. In our case in
Western Colorado, we are home rule and subject to county-level limitations. Our
local public health agencies are participating in health information exchange,
and at least able to through virtual health record, identify redundant
utilization and had an impact on nematological vaccines.
But the digital divide you referenced is real. The only concept that I have
for bridging it is that, again, in making services available on a web services
basis and in doing much of what was articulated through the IHIE presentation
is really the only way to do that. We are not going to build that from the
ground up with the researches that are available in local public health.
MS. SHATZKIN: I think the thing I wanted to add was I think in maybe echoing
what has just been said is that rather than again building redundancy I think
that what we need to be looking at is how our health departments take advantage
of the infrastructure that the health information exchanges are themselves
building. In New York City we work very closely with our department of health,
not surprising since Farzad Mostashari came down here to try to make it a
national agenda. He left it as a New York City agenda. And the department of
health in New York City has actually spearheaded the implementation of EMRs and
is now spearheading a lot of the additional development. When they see the
potential of our analytics database, they just say are we going to get access.
DR. LARKIN: But the caveat there that for every department of health in New
York City or Indianapolis there are hundreds out there that hardly have three
people working there and they are under budgeted.
DR. SUAREZ: We heard earlier where there are 3000 local public health
agencies, city and county public health agencies. There are 50 at least state
agencies. And within each of those and with the work of entities of the Public
Health Data Standards Consortium we have been able to identify some of this
silo systems that exist across public health agencies for each program whether
it is disease surveillance or vital records or any other component. There are
silo kind of systems, multiplicity and duplicity of — the same environment.
And then more importantly it is really this ability to now connect and exchange
data using these advanced standards.
I am again very concerned about that situation and that kind of disparity
that is being created and sort of divide in the digital environment.
MS. SHATZKIN: I think one of the things you can be sure of is that you don’t
really want to have to build that at the level of every county in this country.
Clearly maybe it needs to be made available at the level of the counties, but
it doesn’t want to be built at the level of the counties.
DR. CHANDERRAY: My question is a common theme but we address both to — and
Dr. Larkin. The physician participation — I noted that — that in your area
you have a homogenized population of physicians where they are particularly
employed physicians. In communities where physicians in disparate practices and
they are competing with each entities then they hit the emergency room. A
patient belonging to one physician is assigned to any of the physicians without
the knowledge of the prior physician. It is a competitive issue. We cannot
force the hospital to comply all the physicians taking over the care saying
that your particular physician you should go back to that physician. That is a
major concern in the diverse populations of physicians.
The second thing, Dr. Larkin, how did you get the physicians to participate
in such large numbers? Was it physician reimbursement an attractive tool or
what did you do?
DR. LARKIN: You would like to say we come in and have a tool to help you
manage your practice better and they would say gladly. It wasn’t exactly the
case only because for every reason we could think of. Physicians in the past
feel they have been disenfranchised. To your answer it had a lot to do with the
insurance carrier and saying that we have an incentive program and we are going
to use Quality Health First. If you want to participate in an incentive program
then you are going to have to sign up for Quality Health First and that was the
initial spark. But once the program demonstrated its worth then went beyond
that. But all being said in Quality Health First, it is a pay-for-performance
program. It is an incentive program. The clinical messaging which is a
different model altogether, which includes 19,000 physicians that is no cost to
the physicians and just helps them expedite lab studies and reports.
DR. FRANCIS: Patrick, I will make sure to get to you on that one too in a
MS. SHATZKIN: One of the interesting things that comes out of that — it is
interesting to me the difference when you have one or two major health plans in
your community as opposed to New York City where I don’t think I can count
them, but it certainly is in the double if not triple digits the number of
health plans that operate within the New York City environment and so the
ability to get one major player like that to jump on board and give you
standing automatically just doesn’t exist.
The issue of competition still exists even within the Bronx. I think that
certainly — the question of to whom does a patient belong is a challenging
question. One of the notes I made to myself about the Indiana presentation is
how do you know to whom the patient belongs because one of the challenges that
we face in building a diabetes registry is to whom does the patient belong. I
don’t get good primary care physician information in messages that are coming
from the providers. It is not reliable even if it comes from the health plans,
but the health plans aren’t really readily providing it yet and when you have
so many health plans, that is a tough thing to do. I think that remains a
The interest in being able to alert clinicians because they have said their
interested that is the way we do our registration alerts. A home care agency is
obviously interested in everybody that is on their census. A patient-centered
medical home is interested in the patients it believes are in its medical home.
And if it subscribes on those patients then we will pass them registration
DR. FRANCIS: Patrick, did you want to jump in?
MR. GORDON: I don’t have much to add. Those points are all very well stated.
The problem of attribution is fundamental. The necessity of payer participation
is clear and particularly in the environment you just described where you have
double digit payer sources. The only thing that occurs to me is the necessity
of public policy.
I remain hopeful that Medicare, the center for innovation can drive a lot of
this particularly the creation of methodologies that other payers can adopt
around episodes of care. I remain hopeful that they can resolve a patient
attribution, but there is a very long way to go. And I don’t think it is a
coincidence that in communities like Indiana where you have had some structure,
some history of large payer collaboration or in our own community where there
are relatively few payers that these efforts are a little bit further advanced.
DR. LARKIN: In Indiana patient attribution is done through the INPC or
Regenstrief and that is dictated by the provider and it is determined by how
many times an individual physician has cared for an individual patient. That
attribution has turned out to be a very powerful tool. The success rate is well
into the mid-90s. Doctors can say this patient doesn’t belong to me, but that
doesn’t happen very often because it is a powerful patient attribution tool.
MR. GORDON: It is the attribution for the purpose of payment in more global
payment methodology is the biggest challenge.
DR. FRANCIS: Ed Sondik.
DR. SONDIK: Three terrific presentations. I may have missed something though
with all the energy. Actually I am serious. You may have said this. It has to
do in a way with the digital divide, but it is the divide between the public
health agencies whatever they are that we heard about to begin with and the
RHIO. It is not clear to me exactly what the relationship is and a simple
question. Does the local public health agency have access to the information in
MR. GORDON: In our case, in Colorado, yes.
DR. SONDIK: That is complete access?
MR. GORDON: Services that are available now.
DR. SONDIK: So they could go in and look for breast cancer screenings, for
MR. GORDON: Yes, in the virtual health record that is maintained in the HIE,
quality of which improves is the number of participants in the HIE improve and
the ability to maintain a high degree of accuracy and patient identification
that is sustained. That is one of the immediate values that they have and they
can access at low or no cost and the services fulfill the function of avoiding
duplicate utilization of vaccines primarily.
DR. SONDIK: That is a great laboratory it seems to me to learn how to
improve community health given a broader set of records than the typical or the
non-RHIO supported public health agency would have.
MS. SHATZKIN: Except that in both Colorado and the Bronx that is available
on a patient-by-patient basis. It is not yet available as an analytics tool on
population on accumulated data. It is am I following. That is why we have made
it available to prison health and disease surveillance although if anybody else
in the department of health in New York City asked would they too could have
access. But the point is you are tracking these patients and looking to see if
they have had their follow-up labs because you are doing disease surveillance.
We can give you an easier way to look up John Smith, Nancy Jones than have you
going around to each of those hospitals to get. But we don’t yet have the
I talk about those being available later this year. I think Colorado is
probably looking — Western Colorado is probably looking at very much the same
timetable since we work on the same technology platform. But I do believe those
should be available to the department of health.
DR. SONDIK: The only barrier there is technology?
MS. SHATZKIN: Resources and time.
DR. SONDIK: Or governance.
DR. LARKIN: At least in our model in Indiana, the other barrier would be
that patient data is not provided unless you can verify that this is a provider
of the patient. We would not be able in our local county health department go
in and say Mary Smith, tell us everything you know about her unless the local
county health department had a physician who had a patient-doctor relationship.
You can get aggregated data perhaps. And different health departments provide
different levels of service understandably, but access to this type of data has
to be first verified with the patient-doctor relationship.
MS. SHATZKIN: That is in Indiana. In New York City we would say that the
public health law gives the department of health access to the data and that
therefore consent is not required. It is very complex.
DR. FRANCIS: Next question from Larry and then Sallie.
DR. GREEN: I am struck by all three of your presentations about how —
although you are young in terms of being RHIO you are all standing in
historical developments as I understand it and I am wondering if you could
each, all three of you, indicate anything you care to about what were the
antecedent conditions that positioned you to be doing what you are doing now if
any and secondly, how long did it take.
MR. GORDON: That is a very interesting and difficult question. I think the
role of the payer, a not for profit, but independent and community-based HMO
getting back to the ‘70s which has always focused on quality improvement
and always focused on align financial incentives has a lot to do with it. That
has sustained a sort of this amorphic culture of collaboration that made it
possible for the community to recognize health information exchanges the next
most important thing 5 or 7 years ago. It is very difficult to pin down, but
again I think you can’t divorce the development of some of this stuff from
payment and financing. I think that is a very important antecedent for this
DR. LARKIN: I would think in Indiana we had — it would be my estimate an
unusual alignment of planets. First of all the Regenstrief Institute has been
there quite a long while and well known for health informatics. Secondly, you
had the major payers in Indiana. I mean employers as well coalesced around the
need to improve the exchange and care of patients. I think those two things.
And then you had an unusual, I think, cooperation of four competing major
hospital systems to agree to not to compete in this arena. And then you just
had a lot of heavy lifting. I think the state was fortunate in that. Now that
being said the best I can tell it has been a long, laborious process that is
ongoing and others in the room can speak to that.
MS. SHATZKIN: I would say that in the Bronx the department of social
medicine at Montefiore Medical Center was certainly an inspiring source. The
statements or the actions of the State of New York Department of Health to
actually make funds available so that people who believes this was something of
interest and import could have the resources to try to start to experiment with
it. And, again, the amazing reality of competing health centers coming together
and saying we may compete in other ways but this is an infrastructure that we
should build that all of us can stand on. I think that that was a lucky day
somehow that that happened.
I think that the fact that it has become a national movement has definitely
helps keep people’s shoulders to the grindstone because you begin to believe
that it can really happen. The influx of dollars on a federal level certainly
makes it more doable, the building of standards on a federal level does. Some
of us wonder whether NHIN Direct will put the rest of us out of business or
whether there will actually be a movement that private companies ought to do
all of this and not a public/private partnership. That certainly does keep some
of us awake at night.
There is still a long way to go. There is no doubt about it. I have 5 years
in. Marc has many more I think. I have five already and I feel like a newbie. I
think there is still another five before we will really be able to look to
these things in ubiquitous within our health care environment as things that
people just take for granted that the data will be there when they need it.
MS. MILAM: I was wondering if each of you could speak from your own
experience in terms of your community stakeholders, the participation, and
whether they impacted or not your levels of trust and willingness to
MS. SHATZKIN: Whether our members affect our willingness to participate or
MS. MILAM: Their own or others in the community. How involved has your
community been I guess in governance and policymaking types of decisions?
MS. SHATZKIN: Well, governance and policymaking is very much done at the
board level and we have one member, one vote regardless of size. The FQHC has
as big a vote as Montefiore Medical Center in the decisions of the board. And I
am quite impressed by the level of participation at both the board of
director’s level, a technology committee, a health information management
committee, a clinical committee, patient rights, and member responsibilities
committee that in fact there is a considerable level of participation from the
organizations in the Bronx. Of course that is one again where we have a highly
centralized and facility-based model.
But I think that while they struggle to participate because their own
agendas are very demanding and I find it frequently the reminder is necessary
that it is important for them to align the strategic objectives of their
organizations and the strategic objective of this collaborative they have built
so that they are marching towards the same goal and they are relying on this
organization that they have built to deliver things to them. They do forget
that sometimes I think. It becomes just another thing they are doing on the
side. In general, I think the participation has really been quite remarkable.
DR. LARKIN: The IHIE Board of Directors is well representative. Obviously
the major hospital systems are on the board, but also the school of medicine,
the physician representatives, state medical associations, and other groups.
The measures that are in Quality Health First have actually been determined by
a committee of community physicians so that helped the buy in. I think that
type of broad partnership has been important.
One quick point I want to make which I didn’t make earlier. When we talked
about reporting our physicians’ metrics, for instance, on mammography and
Anthem, for instance, will pay that incentive, the physician’s community wide
score is what is reported out. That is, it is just not the physician’s score
for Anthem patients. All the payers have to agree that they pay against a
community score. This discourages physicians for whatever reason to treat a
patient differently depending on the payer.
DR. FRANCIS: Patrick, did you have something you wanted to add and then
Sallie said she had a quick follow up.
MR. GORDON: Again, in our case the governance has been key. We have a
community board as well, that benefits from multiple highly organized IPA
models for the physicians and we have competing hospitals. And, again, there
has been commitment or consensus that we would not compete on infrastructure,
not build duplicative infrastructures. That is working now. The next level I
think is going to be more difficult when public payers in particular begin
coming down with more global or accountable care oriented methodologies. Our
structures are going to have to become stronger with additional teeth I think
quite frankly in order to support those types of global payment methods.
MS. MILAM: Just a quick follow-up. I am wondering if there are any consumers
on the different boards or on the boards’ committees.
MS. SHATZKIN: We have a consumer representative on the board of directors,
not as many consumers involved in the multiple subcommittees, but I am going
out to find those community organizations when I get back to the Bronx.
MR. GORDON: We are working on that. We have consumers at an advisory board
level, but not at the actual board. And we have small business engaged at the
DR. LARKIN: We have a consumer advisory committee.
DR. FRANCIS: Justine.
DR. CARR: Thanks and thanks to everyone for really great presentations. Just
a question for Dr. Larkin. The measureable improvement, does it surprise you
that it looks the way it does because the work is so great and so important and
yet I wonder if these measures are telling the whole story because some are a
little bit better, some are the same, and some are not as good. What is that
DR. LARKIN: I think, I am not surprised because what these measures are
doing are actually causing physicians and physicians’ offices to change
processes and that takes a while, for instance, how they handle screening for
mammography, how they alert patients. The trend — I am concerned a short
period of time that you are looking at I think is very impressive, but not at
all discouraging me.
DR. HORNBROOK: This is Mark Hornbrook. I am wondering whether there is
something about the structure of the health care systems in which our speakers
are embedded. I am struck by the high degree of advance in Bronx. There is
something about the nature of public health, health care systems that allow
this integration of information that I would expect to happen inside of an HMO.
I hear something that is like Kaiser, but it is inside the whole health care
system in the Bronx and I am wondering what is going on in the Bronx that is
different than in Portland, Oregon.
MS. SHATZKIN: Maybe I have painted too rosy a picture. I think I mentioned
that I can’t get some of those discharge summaries because they are still being
hand written. I think that what has allowed us to advance in the Bronx is the
fact that it is such facility-based medicine, the fact that it is not a lot of
individual doctor’s offices but rather is FQHC’s large ambulatory care centers
that are associated with inpatient facilities that it is so facility based. I
really do think that that has allowed us to achieve a lot more in a shorter
time than if we were dealing with a geography like some of those we have seen
maps of that where if you then laid the doctors out, you would find that you
had a lot of private practices and small group practices and we have in fact
very few of those.
DR. FRANCIS: Greg or Patrick, do you want to weigh in on that? Raj.
DR. CHANDERRAY: In the implementation of this, Dr. Larkin specifically, has
there been any concern with physicians about model practice issues being raised
regarding the publication of the metrics of the performance of individual
DR. LARKIN: No. I mean I am sure but that has not been a common concern.
Although I understand soon the metrics will be public. They have not been made
public yet and that was a conscious decision because we wanted the credibility
of the program to be accepted by the physicians before we went public, but as
far as I know that has not been raised as a large concern because all the data
you can drill down to so it is there.
MS. SHATZKIN: I will tell you that it is the first question from someone in
the room anytime I speak to a group of physicians. And that the notion that
their charts are no longer only within their own institution’s walls but rather
are now available across the bureau is absolutely an issue that concerns them
I think the other aspect of it that concerns them is what if I didn’t look.
I am liable because I didn’t look and there might have been data there that I
didn’t see. I think both sides of that liability issue is in fact a real issue
and one that we have to address.
DR. FRANCIS: Patrick, did you want to comment on that?
MR. GORDON: No. Again, those are both very well stated.
DR. FRANCIS: Mark had a quick follow up and then Garland has a question and
then we will wrap this session up.
DR. HORNBROOK: Just a comment. The epic is building in the view function
inside the system so that with the right permissions the patient can give web
access to their chart as long as their physician or the system has that web
link inside their EMR home base.
DR. LAND: I wondered if the presenters could give us some examples of how
public health has requested and used the data in which they have.
DR. LARKIN: I mentioned earlier of course is the immunization registry which
is actually owned by public health, but to integrate the school systems. We are
still struggling with trying to get the school systems engaged in public health
because they only really want to focus on education.
The other pieces that we look forward to but have yet to demonstrate are
other elements of public health include testing children for lab and
susceptible populations. Physicians typically don’t do that, but by integrating
the data we can be alerting the primary care physicians that a patient should
be referred for lead testing as just one example. But there are a lot of those
types of programs. There would be inner linkage between WIC and a primary care
physician for nutrition alerts. That is all what we would like to have, but we
have not done that.
MS. SHATZKIN: I have to say that I am sort of disappointed still that we are
not doing more with our department of public health. The prison does use it a
little bit. The disease surveillance people do not yet although I think there
should be an incredible day-to-day efficiency for them, get them off the
subways. But I think that in New York City anyway they are much more engaged in
the process of getting EMRs out and I think that that has been where the major
interaction with the HIEs has taken place is that part of the primary care
information project in New York City which is busy there and less in those
niche areas within the department that are doing actual analytics. They are
still collecting their HbA1cs however they are collecting them, not from us.
MR. GORDON: In our case, again, there is a lot of effort to integrate
immunization registry and very similar to Indiana in terms of the integration
of EPSDT services. If programs are built within the primary care practices that
would benefit from that information — primary care which is accountable to
timely screening of kids. It is just again, getting the interfaces complete,
getting the data delivered through the HIEs to those practice sites is the key
issue at this point.
DR. LARKIN: It depends on how broad and the net of public health is. With
Indiana we have a FES system that we actually use Regenstrief to alert us to
emergency room diagnosis presentations that might be a public health threat if
the trends are unusual or high. That is a public health application.
DR. FRANCIS: Ed, you had a question.
DR. SONDIK: I just want to be clear for Dr. Larkin. I am looking at page 12
and 13 which have the charts on measurable improvements. What patients, what
population are we actually looking at when we see these figures say for breast
DR. LARKIN: That improvement is looking at the providers who are in Quality
Health First, for instance. But all the providers who in Quality Health First
at the first measure who had patients who qualify for mammography screening who
go it and taking that same snapshot later and they are still in Quality Health
First, the providers and looking at their patients now and see what percentage
really go the screening.
DR. SONDIK: Regardless of plan.
DR. LARKIN: Regardless of plan.
DR. SONDIK: In Quality Health First.
DR. LARKIN: They had to be Medicare, Medicaid, and Anthem. They were all
DR. SONDIK: When this kind of data is available, I would think —
DR. LARKIN: It is almost a community score.
DR. SONDIK: Exactly. This is the kind of thing I think that people in a
public health office would be point a phrase dying to get. This is giving them
information that they can’t get. The only other way they can get it is through
a survey. But even reporting from a facility won’t necessarily give them the
data in this form.
DR. LARKIN: Correct because the data is rolled up not only just from
DR. FRANCIS: I want to thank everybody and say that we get to build on the
richness that we have already had from all three of the panels because the next
step is going to be after a 15-minute break to invite everyone who has been a
participant to come to the table for a last discussion. Those of you who were
on in the morning if you are still here, we hope you will retake a seat at the
table as well as those who were just here. We will start at 3:45.
DR. GREEN: Well, I noticed something during the break. It was that Don
Steinwachs and I, sitting next to each other, had a hard time hearing each
other sitting next to each other because of the noise level in the room. That
is an objective measure that we are doing something. It is some sort of
Do we have anyone on the phone? In terms of housekeeping, we recognize that
some of you have other things to do in your life like catching airplanes soon
or who knows what and it will be just fine whenever anyone needs to leave, go
right ahead. When you need to go — we are going to assume you didn’t get mad
and leave. We are going to assume that you need to — you don’t have to ask
But to the members of the committee I would like to ask you to start if you
haven’t already a list of the themes you are hearing and the opportunities you
are spotting. We are going to have to distill a lot into that today. If you
have already got that going, please keep that going. If not, would you please
start now? And we will start working off of that quite soon.
To those of you who presented if you would not mind, I would like to give
you the floor now. If you would just do like we have been doing all day long,
put your hand up if you would like to speak or have a question. But I would
particularly like to invite you for the first part of this session to ask each
other any questions you would like to ask each other. And then we committee
members we are going to eavesdrop. We are going to be voyeurs should you have
any such questions like that.
MS. JACKSON: In the spirit of communication also we have been informed that
we have had speakers listening in on the web. They have access to write in
comments and questions and to participate that way. So far nothing has come
through, but if anyone is out on the web and would like to participate with any
commentary please feel free to do so.
DR. GREEN: I assure you that the people at this end of the table will start
talking if you don’t.
MS. MEREDITH: I have a question about the health information systems that
you have been presenting. I would urge you to have consumers on your boards. We
were talking a little bit about our community and with no disrespect to anybody
in the room there is a significant community distrust of insurance companies
and particularly when it comes to behavioral health and mental health, there
are a lot of people who opt out even using their insurance because they don’t
want their insurance companies to know. I really wonder about how effective the
information system can be when there are pockets or areas where people will say
by no means can you use my information or where they will be very upset if they
live in an area where it is done without permission.
It is a concern, I think, not only related to mental health which I think is
a very particularly extreme area, but in so many ways that people feel that
because of the profit motive, sorry to throw the elephant into the middle of
the room, that the insurance will be using ways that isn’t good for the
consumer. I am just interested in your perspective on that.
MS. SHATZKIN: If I can just stick my two cents in. I think what is important
is to differentiate between the use of that information by insurance plans and
the use of that information by providers. And certainly my motivation is to
make sure that that data and mental health included is available to providers.
I think the gap we have created between medical health and mental health is a
real problem and that there needs to be greater integration of that data when
any provider is trying to help a patient, but I think that you raised an
important point about the difference that that is between making the data
available to providers and making it available to insurance plans.
DR. GREEN: Greg, do you have anything to say about that?
DR. LARKIN: If you think patients don’t trust insurance companies, try
doctors. Our program, Quality Health First program, we spend a good deal of
effort because our first participant was Anthem and since that was a major
insurer and Indiana therefore covered most providers. We had to make the
message very clear that were an independent agency for which Anthem has
contracted our independent services because if they thought we were something
different, they thought in fact we were working on behalf of Anthem. Instead of
Anthem using our services we would have had a much heavier lift. It is nothing
against Anthem. And I want to say right in front from our perspective they have
been a very good partner. But you are right. You have to really work at
building the trust and the independency of an HIO is crucial I think.
DR. CHANDERRAY: I think the new health care law passed wherein no prior
illnesses would preclude a patient from getting insurance. The federal mandate
can be made so that none of the information especially the mental health
initiative can be used for raising a person’s insurance and discriminating
against employment issues and made plausible. Then I think most patients would
come forward and they are both willing to have the physicians have this
information — that the safety precaution. Most of the patients would never
want to divulge that information to the providers or anybody.
DR. RO: I think these are really important issues around privacy and so
forth, but I also want to maybe also move the conversation also and extend it
to I think — I think it was Dr. — question about what were the examples of
public health — the data from the RHIOs. And I want to say that I think that
this is an area where the committee could really do some thinking about what
that bridge that we need. What I want to encourage you to think about is really
thinking about — certainly from a public health agency we are looking at
population-based strategies and interventions. The work that we are doing and
our focus is moving is really into policy systems and environmental changes.
While you have the clinical data of individuals, I need to be able to list
that up and figure out what is my strategy because my strategy can’t be that I
am going out to find that individual go solve the current clinical needs. That
is not the parts that we take from our prevention standpoint. Nance challenged
me in the early conversation quickly before then that currently on the Recovery
Act — we have what is called — grant. And we are soon to have the Community
Transformation Grant coming down the line. Hopefully those grants will still
continue to support the kind of population-based, policy-based change that we
think are really important to do those upstream factors.
What are the various measures at the different levels that we are going to
need to have to be able to benchmark progress? And this goes back to that point
about having some standard indicators or a way to think about how we are sure
that we are making progress from a public health standpoint. I don’t
necessarily have the answer. There is a great amount of work that is going
right now into developing evaluations around policy — but cost to see only one
part of it, that that can’t be the only part.
DR. SUAREZ: This is certainly not the answer, but I think in many respects
all that we have been talking about here is really about improving population
health management in many respects. I think all of the participants in all of
these initiatives whether it is in HIE and the provider entities and the payers
and the public health agencies all are looking at improving not just the
individual’s health but the population health. I think there are great
opportunities to share and to exchange information about population health
management. I think most health plans, for example, have in one way or another
very strong population health management programs. Our own system in Kaiser has
very strong and advanced I would say in population health management programs.
As you pointed out, the focus of work of public health is not necessarily
clinical care, the individual clinical care, but more improvement of the
population health and finding ways to facilitate and encourage population
I think that the opportunities that I see is really facilitating the sharing
of these experiences that are happening in every health line, in every provider
system, integrated health system that does have a population health management
program. And sharing that with public health, sharing the public health
experience with respect to population health, and finding common ways of
ultimately improving the health of the population. I encourage us to think
about ways to facilitate that exchange of population health management tools
MS. WELLIK: We were talking at lunch about the first panel. I have to thank
you for this wonderful forum today. The only suggestion I would have if you do
this again is that you go to lunch with all of us that were on the panels
because we talked about a lot of the issues that you are asking us to reiterate
today. We will make that different the next time.
One of the things we talked about and Marguerite totally sized this up in
the best way was that there are a lot of good things happening and we could
probably let them grow like top seed and hope that we learn from each other and
create forums for that to happen. Or we could take the approach that we really
need to look at those 3000 local health departments, think about what they need
at various levels, fund them adequately, and get them started in the context of
a long-range plan to improve their use of data and information and apply it to
population health and then provide the ongoing technical assistance so that the
interventions happen and not just the training upfront. We have just done that.
In the past 10 years starting at 9/11 we developed a tremendous fear about
what could happen around a public health emergency or bioterrorism. The Federal
Government financed every local health department in the United States and gave
the money to the states with a caveat that the states had to sit down with
local health departments and figure out how they were going to split that money
up in order to meet the objectives that were laid out on that plan. And every
single health department, every single community in the United States today has
a public health emergency preparedness plan that functions out of the local
health department and with the community. I think we have a lot of money to
save if we use that kind of an approach as opposed to continually hoping that
health departments will be able to interpret what we are doing with health care
systems in a way that we can tag along and do a good job on population health.
DR. STEINWACHS: One of the things I picked up on was I don’t remember who
said that a focus on the transitions and handoffs. You may have said that. I
guess it was two interesting challenges to me. One was from a public health
point of view what measures do we have that says hand offs are going well.
There is coordination of care or not. And I know that some of the quality
metrics organizations are trying to figure out how do you measure coordination,
integration, and make it reality.
And then it made me think a little bit about and this may be a question more
to the RHIOs — we focus a lot on the medical information and the public health
information being useful to health care providers and to public health
agencies. I was wondering do you see a vision out there where the consumer in
the community could go to the RHIO and log on and say I want you to feed me
information on certain kinds of things or that we had decision support for the
family and the consumer that drives out of this. Because it seemed to me that
when you are looking for coordination, a lot of that starts with what is going
on in the community, but people having to try and deal with it.
DR. HORNBROOK: As I was listening today and again thinking about the Bronx
— experiment — emphasize my environment we have one medical records system
that was set up this huge wall because they were concerned about the members
leaving if they thought their medical record was sitting in a — they wanted to
make sure it was very, very secure. And of course now we are beginning to learn
all the business reasons why you need to send data out in order to provide
high-quality care as well as be compliant with government regulations. All the
data moves back and forth.
I am still trying to get at the issue that in order to provide high-quality
care I need to convince patients that they need to give consent for somebody to
look at their charts. In order to get — supposedly you are traveling in Europe
and somebody could get on the web and look at your chart. And if you are
unconscious in a car accident, shouldn’t you have some way of letting the
providers in the emergency system in Europe see your medical records. I think
so. Or if you are in some horrible kind of disease status. The same way — in
the United States.
The view to access is coming in more and more is the EMRs is going to make
people more aware. We have the educational problem and challenge of how to
convince different segments of society that this isn’t big government trying to
take away their privacy. It is trying to give them the best quality care that
they deserve and their family members even if it is my great grandmother who is
very frail and needs care and I am 3000 miles away. We need to talk as a
committee on this educational notion of how we talk to everybody whose medical
record is private yet at the same time it needs to be shared.
DR. FRANCIS: This is actually — I am going to send it to Debbi with the
observation that you all made a fantastic point as we were talking over there
how we were understanding communities differently. Maybe you want to elaborate
DR. MAIN: In fact that would be another lunch conversation. And in fact
before the conference in fact we were having dinner last night with Larry. One
of the take homes for me was that we all really talk about the work. We use the
community in very different ways. The question was asked earlier sort of how do
you involved you communities and often I get it. I have worked in all sorts of
areas where at times the community is providers, the times the community is
your constituents in a larger kind of higher level of grassroots levels or
actually grass top levels as they say.
But in our case I think one of the missing — what I see as one of the
missing pieces is the community at the levels that we were talking about where
we are really talking about neighborhoods and even Dr. Suarez mentioned the
digital divide between public health and clinical care. And I think the missing
piece but the common denominator is the person or the patient, however, you
want to frame it. I would really like to push this group to think about the
other divide and to spend as much time thinking about what that would look
Really based on your question about how do you think about involving people
with HIEs or I would argue the other way around. How do we make it more
bidirectional where it is not me or somebody else who lives in my neighborhood
going up and grabbing but also spending? I think it is really to push it a
little further to think about the divide in three ways.
And I don’t know if that actually answered the question but I think that is
part of our point is we really need to think about all the communities.
MS. GREENBERG: I can’t help but comment that the three ways reminds me of
our diagram of back to our information for health. I don’t know if any of you
are familiar with it, but the community, the provider and the person.
I want to thank all of you so much for coming. Sometimes we worry that we
will throw a party and no one will show. Certainly in February we worry about
that. Wiser minds might have told us we will wait and spend a little more time
thinking about this and have it in the spring. We felt this urgency as we saw
what was going on in the bigger environments to really get these community
voices here and I know you are humbled about representing many other
communities as well. It has been a really exciting day. As the executive
secretary, I am absolutely thrilled that the Committee was able to do this, but
of course none of these things happened without a lot of people.
I do have a question or two but I just wanted at this point before everyone
scatters, to recognize the members and the staff and consultants, et cetera, in
addition to all of you, who obviously made this possible who helped to bring
this all together. I know all of you have talked to Susan Kanaan, who actually
has been working with us for 20 years, but this was a perfect assignment for
her because she is really very much of a community-oriented person. Also Missy
Jamison — I think many of you talked with her. Missy is on my staff and I
wanted to thank her. I think many of you talked to Jeannine. I don’t know where
she is and helped make the arrangements and may have talked with Marietta and
of course our subcommittee members and chairs. I wanted to thank everybody
before everyone scattered, how worthwhile I think this is.
Of course this is not the last you have heard from us. We always tell
members that once you are on the committee, it is forever. I am serving the
role as mother of the committee. I am particularly pleased to a lot of what you
were talking about was this in this American Journal of Preventive Health
special supplements that just came out and Strong Medicine for a Healthier
America and the introduction was written by two former members of the National
Academy: Risa Mourey and David Williams. We have had a long time interest in
this. They talk about how there is more to health and health care where we
live, work, learn, and play and affect our health more and what happens in the
physician’s office or in the hospital and of course that is what we were
hearing all day.
I always think of myself — I spent most of my career in the government. I
think of myself as rather energetic, but I must say in comparison to your
community people, I don’t feel all that energetic — since I am not in a
position to roll up my sleeves and actually do things the way you are, what I
try to do is — I think there is a role in the Federal Government to identify
best practices as we have been talking about strategies, et cetera, needs that
we can promote.
Two particular things I had a question about because — two things that I
heard a lot about all of you talking about. One was of course what I just read:
the environment and very broadly based. I was saying to one of my colleagues 5
years ago you would never go to a meeting on health information or even public
health and hear anything about grocery stores. That has become a major issue
and I guess with the obesity, et cetera.
I think that many — a few of you talked about doing environmental scans or
environmental audits. I am somewhat familiar with this group and another
related area and that is the disability area where communities have actually
done audit of the accessibility of their physician’s offices, the community
resources, et cetera. I think that all of you have done something in that area.
And what I wondered is is there a — I know there has been some actually some
work that has been published in disability and accessibility area. Are there
standards — that is what I can do in the Federal Government, try to bring
standards and best practices and that is kind of what my main focus is. Are
there templates? Are there standards? Is there literature or are you all kind
of having to make it up as you go along?
Now each community is unique, but on the other hand each community as
someone pointed out I think has all these environmental factors at one level or
another. I wondered — that was my first question and my second one is
obviously collaboration, getting people to collaborate. Susan identified some
themes before the meeting and one of them was strategies for building and
Clearly that is critical. Again, is that something you can learn? Are there
venues for people to learn how to do this better and to learn from people who
are doing it well? Or, again, are you having to make it up as you go along?
In those two areas which don’t really have much to do with health care
though — obviously the coalitions include the health care system — and just
welcome some feedback.
DR. MAIN: I will give you a couple of examples. I talked earlier about
Minnesota’s statewide health improvement programship that is a program being
delivered now through all of Minnesota with state funding. And the way that
that was developed was after several of us did Steps to a Healthier US pilot
project. Minnesota developed a program based on evidence-based models that we
could select from. We had interventions, a menu of interventions we could
choose from that would improve healthy eating and improve physical activity and
reduce tobacco use in our communities. This is being done at the system policy
and environmental levels. For instance, we have sample policies, sample
ordinances, things that we can standardize across the state.
An example I would use would be complete streets. In my county there are
five cities and four of them have adopted this complete streets strategy of
saying that any future development or redevelopment will include streets that
not only accommodate cars but also bicycles and walkers. It is easy to pass
that. The next step is holding them accountable to deliver on it. In Rochester,
we have had success with all but one project, but that project backfired 250
percent. We have learned some things along the way. Yes, there are standardized
ways, for instance, across Minnesota that were doing that.
And then just briefly about collaboration, there are definitely ways people
can learn to collaborate. I mentioned our Collaborative Leadership Fellows
program that we have locally. This is based on the national model of the study
of collaboration by the Turning Point Initiative in the 2000’s. We found that
there was a leadership gap in local public health and as we looked at that, the
particular area we felt that we are very weak in and that we were responsible
for was collaborative leadership development in our communities. I served on a
national group that developed a framework for teaching collaborative leadership
based on the research that was out there. University of Colorado was
particularly helpful in some of the work we did there.
There are specific skills that people can learn and be taught and can
practice with each other. It is also important then to network with people who
have done that so that we get past the idea that collaboration is just sort of
intuitive for some people, but it isn’t real. It is definitely real and with
that kind of training can help people get there.
PARTICIPANT: Other responses to Marguerite’s questions?
DR. MADDUX-GONZALEZ: As you can tell sort from what — there were
similarities in many of our presentations. There is a sharing of some of these
policy interventions, environmental interventions, and some strategies to help
us with that, but there is a lot more that can be done.
One thing during lunch that I was sharing, I think California, there is
actually, we have the California nutrition network GIS map dealer. That has
allowed us to look in our county to see how close people are in walking
distance from a grocery store, poverty level relative to grocery stores and the
source of information is Dun and Bradstreet for that. We need to be looking at
other sources for information that don’t traditionally think of as health
DR. RO: A quick response on that one. Some of the challenges that, to a
standard, standard measures. That is actually a really great one which is like
walking distance to a grocery store. That works as long as there is not a
freeway between you and the grocery store. That is like a really real example.
Have we come up with the right set of measures for obesity? I would argue that
we are not there yet when we look at environmental measures. And understand
when we talk about collaboration — location of freeways and walk ability of
streets is not a public health agency. It is not a community-based
organization’s purview. That is the department of urban planning or whatever
that department is.
What we are being forced to do now is figure out how do we work with these
partners that we haven’t otherwise worked with that have authority over some
part of the environment, and match what their priorities are. Maybe it is safe
streets. Maybe it is ensuring that there are fewer pot holes. Maybe it is
fastest commute from this major business to somewhere else. It is matching
those priorities. I think we are looking forward. That is where we are headed.
MS. STEWART: I do not know what part of the Federal Government you are with,
but HUD, EPA, and the US Department of Transportation, are working on all these
templates under livable communities and just produced a series of templates or
ideas and guidelines about what a livable community is, what the best pieces
are, this last fall. On a state level, Colorado is also trying to include
within that a sense of — when you turn in your comprehensive plans for urban
development and community development, that it includes a comprehensive health
care plan. That looks at overall issues like walk ability, like being able to
get your kids to school without them walking across the street, those kinds of
What is interesting about how they are doing it is the level of conversation
that they are having around rule entities as well as metropolitan entities,
makes a really big difference. That has had a big effect on the Department of
Transportation who has previously never considered that piece of it as being a
part of their criteria. I don’t know where your building is, but I would
definitely have a conversation with them.
I know that Health and Human Services and the Department of Education are
trying to do the same kind of thing in terms of making sure that the livable
communities also include pretty decent schools since we know that education,
the serious social determinants for health outcomes.
MS. FLYNN: I guess I would just underscore what was said previously
especially by Marguerite, and now by Tracey. When we started working in this
area more than 10 years ago, there were no standard set in terms of these
audits from the environment and we had to create the documents that I referred
to earlier and kind of out of the whole cloth, but in those 10 years the
conversation has evolved to we are having these conversations now. That is
really exciting. Some entities are beginning to look at that as Tracey was
But we, in the meantime, have been looking at ways to influence communities’
general plans, which I think everybody has to have no matter where you are in
the United States. And we learned that in LA, maybe in the State of California,
of the seven elements that are required that are mandated to be addressed in
city planning, health is not one of them. We are working to try to incorporate
health as an element of the general plan that would create a set of
standardized criteria by which the city planning and the community
redevelopment agencies will have to operate. That is something that we can
maybe all work on in our various locations.
In terms of collaboration, I think that there is an opportunity for there to
be a mechanism set up where people can be trained. There are certain skill sets
and certain other requirements that I think would be included in having an
effective collaborative and we are now learning that it is necessary to have
everybody at the table, the city planners at the table along with community
stakeholders who are impacted directly by these priority areas.
It is something that we have developed. Community Health Councils has what
we call model for social change that talks about — and I think there is the
community — what is it — community-based participatory research, that
everybody is using in their — in that you are asked to build a coalition that
is representative of the community across sectors and we have been doing that
before there was a name given to it. There is this model out there that can be
taught and replicated in other areas.
DR. GREEN: Other responses to Marguerite?
MS. SHATZKIN: I just wanted to add a couple comments to things that were
mentioned earlier. One of them is I would hope that across the 50 states the
Department of Health is engaged in whatever health information exchange
activities and regional extension center activities there are going on within
that state and to the extent that this committee have some influence on making
that happen. I would urge you to do everything you can and to make sure that
they are meaningfully engaged. I think I made more than one comment about
really having to — even in New York City where we have a pretty active
department of health having to pull them in to get them to really be engaged
and involved and look for their opportunities both to speak their mind for what
they need and to take advantage of what exists. I just would put that plug out
When it comes to engaging patients, I think it is important to note that
personal health records while not mentioned particularly today are absolutely
growing on the horizon, brightening on the horizon. I think that there are a
lot of challenges still. We talk about them in terms of those that are tethered
to specific electronic medical records and those that are independent of any
medical record, but I don’t think there is any doubt that most of those
involved in health information exchange are actively looking for ways to bring
the personal health record into the mix and make it possible both to push data
out to patients so that they have ready access. In the HIV community I worked
with people who say anything to save the patients another stick.
I think there is an effort to push data out but also to move towards the
point where you can be bringing data in so that the patient can be providing
information about over-the-counter medications, about concerns that they might
have access to various tools that they could use for assessing their own
health, et cetera.
I would say that this is very nescient. I don’t think it is very well
developed. I think it really does have a ways to go, but I think it is only by
paying attention to it that will really make it happen more.
I have two more. One is the whole issue of educating our entire population
which is certainly a difficult thing for any of us to imagine accomplishing.
But I do think that departments of health have a role in this issue of getting
patients to understand that health information exchange is a good thing, not a
bad thing. That their data should never have been locked up inside the doctor’s
filing cabinet. It should always have been available for the other providers
who are engaged in their care and that they should not be afraid of health
information exchange, but rather should see it as a coming of age and an
improvement and I think that the departments of health just like they do
education around smoking. In New York City we are bombarded with the most
horrendously terrifying advertisements. Easy for me. I never smoked. And you
hope that they are having some impact.
If just a little piece of that budget could go towards educating patients
about the benefits of health information exchange, maybe that is another real
thing that the department of health could be engaged in. I can tell you in New
York State they promised us that they were going to do this and they haven’t
done any of it and it has been a big disappointment to all the health
information exchanges that there had not been a statewide education campaign
that reinforces what we are trying to do.
And then I definitely believe in the personal stories. I look for the
powerful patient stories that I can find to be able to use them as the
demonstration patients that I then take to clinicians to show them the power of
what a health information exchange can offer. And I think it is tremendously
important to understand that there is a the patient who gets the extra stick,
who have the unnecessary radiology exam, who gets a medication they shouldn’t
get because somebody didn’t know about another medication, all of those are
At the same time I have to say we don’t build the systems patient by
patient. We build the systems at a much larger level by taking ourselves out of
the case by case and looking at what we can do over a larger scale. We have to
remember that we have to do both.
DR. LAND: I was struck but probably not surprised by the answer to my
question of examples of how data was being used at the public health level. I
understand the reasons for that I think. There are many reasons, not the least
of which is the lack of infrastructure in state health departments and local
health departments. And that is probably not going to change very quickly
especially in the environment that we are in right now where a cutting back on
resources. That infrastructure in public health is in dire straits and getting
worse by the day.
All of our previous discussions which we had of great glorious days that we
are going to be solving the information, public health needs by RHIOs and so
forth. It is just not going to happen folks. It may happen for our
grandchildren. In the next 5, 10 years, let’s be realistic folks. We are a long
ways off. We have the most developed systems in the nation that talked to us
and said they have very few examples.
What do we do about that? A couple of things. I think one is — I think
somebody mentioned the fact that actually public health has data that the
information systems don’t have and maybe that is the place to — if we can
build bridges that go in the other direction that might be as valuable to
anything else because there are several — especially every public health
agency has data on lead, has data on newborn hearing, newborn screening,
immunizations, and so forth.
If we could develop bridges for the infrastructures that already exist and
being developed in helping the clinical side, maybe that is a more productive
area in the future.
But then I am not willing to give up on the idea that there isn’t good data
out there in the clinical arena that public health should be used, but I don’t
think it is a matter of actually transmitting the data files because the
infrastructure isn’t there. If there is a lot to be learned about — one of the
questions that public health need to be asking that could be answered by these
information systems that it isn’t down at the individual level. It is at the
aggregate level if we knew how to frame the questions so that everybody is
asking the same questions of their RHIOs and information exchange. Again, the
reality of public health is that we are in a programmatic funding situation. We
have programs for chronic disease and for diabetes and for cancer and so forth.
That is not going to change.
Taken that into reality if the diabetes people or the cancer people or
whoever come together and say this is what would really be helpful from a
public health perspective if we had this type of information not at the
individual level, but at an aggregate level and then create a standard around
that that they all of the health departments could go to the information
exchange that said can you provide us at least these types of indicators and we
can show you how we are going to use that information at a community level
because most public health agencies don’t have the time and the energy to be
thinking that through by themselves. The questions are basically the same from
one state to the next. And if at a national level we could be pushing TDC and
other public health infrastructure in that direction to provide that type of
standardization of what are the data that could be provided back. It would help
us. I think it is doable where some of this other may not be doable in 15, 20
DR. MADDUX-GONZALEZ: I am glad you went before me because I was going to say
much of what you are saying right now in that the RHIOs and the nonprofit
health information exchanges are really interesting. At one point we are going
to do a RHIO some time ago, but at this point frankly we would not be able to
afford doing that at the community.
In those communities where the infrastructure doesn’t exist how do we
proceed and even if they do I am not sure that is a substitute for
bidirectional, real time exchange of data with local health departments and
providers, that exchange. I think it is critical. I know in Sonoma County where
our providers are Kaiser, our community health centers are very interested in
looking at the aggregate data to look at the health of our community whether it
be hemoglobin A1Cs or BMIs but looking at aggregate data for the community and
the helping that inform the public health strategies and then tracking the
effectiveness of those. I think that there are some relatively low-cost
strategies that will allow us to have that health information exchange. And any
help you can provide in accessing and better understanding those opportunities
and then the standardization of some of these indicators would be excellent.
DR. HORNBROOK: I am doing a project right now about sharing data between
Kaiser and the Sierra registry in Milwaukee, comorbidity data. So expanding
cancer reporting through comorbidity. I am very impressed at the fact of how
much more valuable the cancer data becomes when you know somebody has listed
comorbidities. And in fact one of the things I am just very interested in
knowing this, but if you look at the month before cancer diagnosis in the
cancer population, average days in the hospital are 4 or 5 days. There is the
day of diagnosis. You look one month behind that the average number of days —
one stay or multiple stays — is 4 plus days for women and 5 days for men. And
men are consistently higher comorbidity than women whether you look at
diagnoses or prior use.
There is a whole set of issues of linking — abstract data, EMR data, claims
data in the cancer registries which almost all these informatics pathways we
are talking about. And yet when you look at the funding for cancer registries,
it is very squeaky like any public health. I don’t know what the answer is
other than the fact that we are missing a whole bunch of cancer advocate
telling the rest of the world that they are being killed by bad data. Where is
the TV movie of the week? Where is the story of somebody testifying in front of
Congress that their grandmother died of cancer because somebody didn’t know
that they had diabetes? Clearly diabetes is one of the most important
comorbidities of cancer patients.
Just talking from my own research agenda here you could probably pick any
disease area, any problem area and of course you start going backwards in time
and you look at hypertension, hyperlipidemia, elevated blood sugars, your
metabolic syndrome, and you look at BMI which is again pre-diabetes. You can
see the pathways starting back in teenage years and just follow them right
straight through. And of course the rate of cancer is going up as us baby
boomers get there. It is going to be complicated cancer and Medicare is going
broke. Social Security is going broke. Our national debt it —
MS. SHATZKIN: But we all need antidepressants.
DR. GREEN: Mindful of the time, I would like to propose something. We got
all these just in mass rebel against this if you don’t want to do it. I really
would like for us to start consolidating our thinking around themes and
opportunities again. And as a strategy or actually a tactic for doing that, I
am wondering if you would be willing to just run the table where every person
sitting here gives voice to a theme that they heard today that is particularly
important from their perspective and also articulate something that they see as
pressing or really promising, giant, juicy opportunity that is staring at us
that you heard today, a theme and an opportunity. You can pass if you wish on
either one or both, but here is your chance to push back. Would you be willing
to do that? Would someone be willing to try to help capture these things?
Themes and opportunities. All right Stephanie, would you mind if I started with
you? The good news is Stephanie gets a fresh slate. Poor Mary down there by the
time we loop this table — can you start us off, Stephanie?
MS. BROWNING: Oh boy, it’s hard. Theme. Neighborhoods and people and places,
and how that ties into the larger picture of public health systems and how to
get that information down into their level. Opportunity, greater collaboration
and training opportunities, the locals.
DR. MAIN: Analytics. That is actually one of the big themes no matter where
you are. The whole issue of making data usable in a very timely way which means
being on top of it analytically.
DR. GREEN: Was that the theme or the opportunity?
DR. MAIN: That is actually a theme and I think that is an opportunity as
well. I like that. Two for one.
DR. STEINWACHS: Don’t you need apps? I thought we were talking about apps
that go with the analytics.
DR. MAIN: I like that. Yes. That is the opportunity as well for some
MS. STEWART: I would say a theme that I have heard today is disconnection
between community and data collection. I would say the opportunity is
connection between community and data collection.
MS. MEREDITH: Similar, but different. I was thinking that a theme is
collaboration whether it is in systems, applications, the way that multiple
layers can use similar information or systems so that they don’t all have to
duplicate. That theme is collaboration and I think it is also the opportunity.
It seems that there is so far for us to go in terms of collaborating really
effectively to resources and to make things more effective.
DR. BREEN: We have been carping on this committee for a long time over the
fact that — is no data lobby in this country and how important we all think
data is, but it really doesn’t get its due and it is not the sort of thing you
introduce into Congress — we want money for data.
But I think what we have seen today is illustration after illustration of
how useful local data can be and how it can be used to build coalitions, to
point out problems, to develop strategic planning and policies that work well
and that can then be monitored and maintain those coalitions over time and
eventually improve the public health. That is what I would say that I really
felt strongly that I learned today.
DR. TANG: The theme is that there are lots of people doing God’s work one
neighborhood at a time and not to their fault they are doing it in silos
because that is the way it — et cetera. There is plenty of volunteer labor —
even the people who are getting paid are volunteering — to their passion, but
there is so little data for them. I think the opportunity for us as a nation is
to provide them — somehow we need to work together and I am sure they are
willing to kick in to collaboratively develop the standards, the framework and
the tools that help them do their job. That is not an easy thing, but I think
we have to do as part of the Federal Government as a public good. That would be
DR. LAND: I guess my long winded spiel earlier is probably what I would have
said, but since I have an opportunity to say something else. I guess one of the
things that I heard which I was really pleased to hear was many of the
presenters talked about data that they are using that are basically social
determinants of health. I found that good. Sometimes we get captured in our own
public health datasets which typically aren’t social determinants of health and
I found that a great thing to hear coming over and over again. To what degree
that is going on nationally or if it is just the groups that were here, I don’t
know. It is something that I think it is important that we ought to be
encouraging and helping find other datasets that are out there to assist in
that regard. And maybe that is where the opportunity lies. What is the
technical assistance that is needed to assist in jurisdictions that don’t have
a lot of resources to not reinvent the wheel but learn from the good work that
is already occurring?
DR. HORNBROOK: I was thinking about instead of teachable moments assuming
the world is digital. There is digital medical record world and a nondigital
medical record world. Hopefully the digital medical record world is getting
bigger under Medicare incentives. There is end-of-life care. You ought to be
instructed what the wording should be in your end-of-life care planning and
giving somebody access to your medical record. What are the words you need to
put in because I just filled out mine — to your attorney that says here is how
you get new access to my chart if I am unconscious or not able to give it to
The same thing for travel clinics. If you are going to leave the country,
tell every person who is leaving the country how to protect themselves when
they are leaving the country about having view access to their chart, the
electronic medical record system.
The same way if you take your drugs outside to get filled at Costco,
Wal-Mart or whatever outside of your health plan pharmacy, the patients ought
to tell the pharmacy to send the data back to their health plan.
The same way with public health clinics. The public health clinics take care
of Kaiser patients because Kaiser patients go there. They tell us. They take of
our members because members think those clinics are better for them or more
friendly or they are looking for something specific. They don’t want to —
again, there needs to be a teachable moment of telling them how to connect back
to the medical records system.
This whole notion of emergency care. The teachable moment is your car
accident. You could have had a really bad experience. Somebody should have
access to your chart in that ER. And by saying that to all people who have
digital medical records you tell the rest of the world who don’t have digital
medical records but they are out. They are missing. Maybe that is a teachable
moment for the rest of the world to put pressure on moving the digital divide
to help deal with them. Now I can’t deal with the privacy people who are going
to come back and scream and yell. At least we can have a debate but it is not
going on right now.
MS. SHATZKIN: So I guess one of the things I am realizing is that what you
see as a theme also has to do with what you already know. For me one of the
themes today was a concern I heard about the ability of our departments of
health to remain effective in the changing environment of health care
information and that was new information to me o some level. It is a theme I
The opportunity I take away is how important it is to engage community
activists in setting and moving the health care agenda and I think that that is
an opportunity I am certainly going home to start building on.
MS. JACKSON: The cultural shift, making peace of having everyone at the
table and listening to what they have to say and getting everyone involved. The
idea of having the folks who were involved in designing — records and
designing some of these plans and that helps to mitigate the privacy aspects of
it and the consent aspects of it. It just blew my mind. It was a big eye opener
DR. STEINWACHS: To try to avoid duplication I am picking up on what others
said, but changing the wording. Limited resources. I got that message over and
over again. And particularly as soon as you move out of that part that is
funded now — the high tech — and seemingly more resources, but even there
people probably complain of limited resources. That takes you to a concept I
think we talked about a lot here about shared resources and there are two areas
in which that sharing might occur. Todd talked about some at the HHS level and
the community level. There is infrastructure and the others in training, those
It seems to me it would be a great opportunity to try and take a systems
view of this and say what can be done that everyone can benefit from at the
national level and that is not going to apply in every community and every
place. What can be done at the community level that might be generated and used
in other communities? We were kidding about apps a little while ago. You look
at the model that Apple has and the databases we are talking about, many of
them potentially lend themselves to that. There may be some ways to get
investments and analytic and so on that everyone can benefit from. Otherwise we
just don’t have enough resources to go around.
DR. GREEN: I am going to pass on both.
MS. MILAM: I am going to build on both Tracy’s and Debbie’s comments. We
heard over and over about the importance of trust and actual participation, but
we also heard about the disconnect in terms of privacy, meaningful privacy in
terms of our current construct. And in the clinical setting we heard a little
bit today about HIPAA or reference to law and regulations and good or bad that
defines the clinical setting and it is fairly clear what you can do and what
you can’t do.
One of the opportunities I thought I heard is that it is not so clear in the
public health setting that there is no easy framework that an organization can
just grab a hold of that is a public health type of organization that can know
at what level they can disclose identifiers on the web to a researcher, to
another epidemiologist, to another state agency, to the Federal Government. I
think I heard a big opportunity to provide some easy to follow best practices
for states and communities and neighborhoods to get their arms around in terms
of freeing up their data but doing so that in a way that would engender the
DR. FRANCIS: Sallie said it wonderfully. Let me just add to that. I don’t
think anybody has quite mentioned disparate understandings of what communities
think is core to improving health. Now that is sort of social determinants, but
maybe not quite. That when we think about that it may really be that we need
different types of information and then linking to what Sallie just said
different paradigms for how it is that we ought to consider protections of that
information and govern whatever system we put into place on that score.
MS. GREENBERG: As predicted, I agree with all of you. But I guess to me the
bottom line is that all health is local as paraphrased a former speaker of the
House but also I think again to quote one of our former members. Lloyd — is
the House commissioner in Arkansas — whereas all health is local and obviously
it matters. That is where you get healthy. That is where you stay healthy and
that is where you get sick and that is where you get better hopefully. But it
isn’t true that if you have seen one community, you have seen one community.
There are a lot of similarities. HIPAA made the Medicaid community realize that
they had always said if you have seen one Medicaid program, you have seen one
Medicaid program and it just couldn’t be anymore and this is where the
opportunity is I guess. Now we are national standards under HIPAA, but they
have to follow. I am not recommending legislation, but there is not something
similar so that in a way public health departments are out there to some degree
on their own and making somewhat on their own, but there are a lot of
opportunities I think for synergies. We have heard that.
I think the challenge for the committee is to think in terms of almost the
recommendations that cannot stifle in any way innovations, but can encourage
this kind of collaboration and learning and sharing of resources and maybe more
of what I was given permission for. Somebody mentioned — I don’t know whether
it was you, Nance, but the message to public health that they need to be
involved in HIEs and we have talked about for a long time. But also that they
need to be involved in these extension service and that is something that has
concerned me from the beginning because I have heard a tremendous need for
hearing best practices, training, technical assistance, education, all of that.
And I just wonder to what extent the — there is a huge amount of money going
into these regional extension services for health IT and I really think we need
to dig deeper to see to what extent they are or could address —
MS. KANAAN: One thing that is surprising to me hasn’t been named yet and
that has stood out for me is — and this is kind of an extension of the points
about social determinants is the point about using data to impact policy. And
an important corollary which is so many of your communities have found allies
and champions among your elected officials. Many of the communities have
examples of that and I think that is extremely significant to really take aim
at public policies and find effective ways to influence it. That is the theme I
would call out.
I was trying to get at some way to bring in something I believe in which is
the importance of asking the right questions. I do think there is a lot of
evidence to that as well. I will just identify that as an opportunity that link
to what Garland just said. I thought that was a really interesting idea that
warrants further exploration, the idea of public health community framing
queries for clinical information systems where they really think about what are
the most important questions to ask in lieu of having all the data that we
would really like to have. I think that is really an interesting exercise to
try to identify the right questions.
MS. JAMISON: Following up on that a little bit I was really impressed with
the addition of the qualitative data to Garland’s three types of data.
Frequently in public health you ask an epidemiologist a question and you get an
analytic answer. I think it is really an important opportunity to think about
qualitative data that some of our communities have indicated is very important
to them in terms of motivating their community, involving their communities.
DR. OVERHAGE: It is always challenging to figure out what the most important
thing is. It seems to me that there is a lot of energy and passion that we
heard about, but a lot of beating heads against the wall as well on several
fronts meaning there is progress being made that a lot of the energy is spent
overcoming barriers and hurdles rather than doing the work. The opportunity I
think that might represent is that the — at least it seems to me that one of
the ways to help with that is trying to align many of these specific goals and
objectives. We talked a lot about clinical data exchange and public health
needs and the collection of data for business operational purposes — laws for
research — it is hard to do all that.
If we don’t align those things carefully and sometimes put a comma on one of
them so that the others can come along with it, I think we are going to
continue to struggle. I think the opportunity is to align in a way that enables
overall progress even if it means slowing somebody else down so that we get
somewhere instead of beating our heads against the wall because it is too hard.
DR. CARR: I am impressed with the emphasis on the data infrastructure and
thinking back to our last meeting when we heard from Todd Park and again this
morning, I am reminded of our conversations and discussions in our quality
meaningful measures. To give a qualitative example in 1994 the agency for — it
was in AHRQ too, but the predecessor of AHRQ said ACE inhibitors are good for
heart failure. They reduce mortality by 30 percent. And 10 years later we made
a measure and we look at that now today. There is a lot of wringing of hands if
you are not at the 98 percent compliance, only 97 percent and you won’t get
your value-based purchasing money.
With that example in mind and looking at the very beautiful display that
Sonoma County has put together for all of the indicators I would make a very
concrete recommendation that we do the kind of thing that Todd showed us this
morning except I can’t make it work. If we had all of that together and we
pulled all these data sources and everybody had their starter set of
information on their key indicators and where they stood in their reference
frames as Sonoma County has, we could take work out and free up the people to
do the more drilled down work that is needed. In summary, to follow the health
care model of a finite set of measures that would benchmark and that you can
measure yourself against and work toward improvement.
DR. CHANDERRAY: I think the theme is to develop policy for public health and
public education. To develop this policy I think we need to understand the
barriers to develop this policy and also implement it. I think the hub of the
— to me is the patient and the physician and they are being ignored in the
process. And a lot of participation of this kind of development involves a lot
of institutionals and major academic census where — don’t see much involvement
of the community physicians that are out there and provide the bulk of medical
care in this country. I think we need to involve them and the opportunities
that we have is in identifying limiting factors that prevent patients and
physicians in participating in this delivery of public policy is to remove the
fear of participation. The fear of participation by the patients is like I have
indicated earlier is depriving of the myriad of funding from Medicaid check and
also the loss of employment if an employer finds out about the health what kind
of penalties they would —
And most of the physicians many that I talk to are worried that this data
gathering and meaningful use is nothing but a Trojan horse for healthier
agencies to look at and penalize them for the wrong that they are doing in
their practices, and also the fear of potential malpractice when this data
becomes public. I think if you remove those limiting factors, I think we can
have the major part of health care — the patients, the physicians participate
in the process and develop this quality that will make it very viable.
DR. SUAREZ: I think three themes that I — some of them have been already
mentioned and I want to emphasize them here; public health information
infrastructure, workforce and technical assistance. I think the first one goes
back to the theme that I mentioned earlier which is this risk of the digital
divide. What I heard actually was that it is not just a single dimension kind
of digital divide between clinical peer and public health. Within or between
public health it is not the same as the City of New York Health Department
compared to the health department of some other jurisdiction. And even within a
public health agency different programs and this whole issue of the silos
across agencies. So public health information infrastructure and the risk for
the digital divide is a critical them in my mind. Workforce and technical
assistance are the other two.
And when you think of those three and probably a few others, you have to go
back to the whole concept of the electronic health record incentive’s program
and the whole national initiative towards improving that option of EHRs and
health IT in this country. These are the same critical element that we are
incorporating today. Workforce development, technical assistance through their
regional extension centers, investments and improvements of health information
infrastructure — certainly by supporting that option of electronic health
records. It is almost like drawing parallels of all the components of our
current initiatives around an option of electronic health record systems in
this country and applies to public health and kind of the same need that exists
I think opportunities I would probably see first of all the need to pursue
mechanisms that support this improvements of the information infrastructure in
public health, improvements in the training — workforce, the availability of
technical assistance, those kinds of areas. And I think Justine touched on
that. I think identifying best practices on things like health indicators and
applying those templates of things that are being done in some states and some
localities to other localities, applying some of the concepts that Todd
mentioned at a national level how we are trying to look at health indicators
world and take into the micro levels of a state or a local community.
I think one that Mary and I were talking about during the break was I think
everybody was very encouraged by this particular hearing and the opportunity
that it provided to create communication channels or foster communication
channels between public health agencies and the federal government and so
creating and finding appropriate vehicles to maintain as a critical topic for
us and foster that communication between federal, state, local public health as
well as the rest of the industry would be a very important opportunity that we
should pursue. Those were my themes and my opportunities.
DR. BERNSTEIN: I think I am going to undergird the observation that what you
draw as a theme is what you already know. I heard a lot this afternoon about
the need for education of consumers as to the usefulness of data sharing and
even sometimes of providers as to the possibility of legal data sharing and
maybe that means education of overly conservative lawyers and law makers. And
for me that is an opportunity to get people behind community projects to
improve care. But of course it has to be actually true that the data can’t harm
people if it is shared and that is not yet universally true. I think our
challenge is to move in the direction that makes that more true for people.
DR. GREEN: You may have noticed by now that when we got to this end of the
table where the community is, everything got longer. That was supposed to be a
compliment to our guests.
MS. FLYNN: Okay, so a theme. I don’t know that I am going to come up with
anything new and as you suggested, probably reword things. The theme for me was
the expansion of health data, a collection beyond the clinical and including
environmental factors in that which is basically the social determinants —
And opportunity. I guess I look to what was reported out by Todd this
morning as an opportunity to build upon with the establishment of a launching
of that one-stop shopping database that will help in the resource sharing and
maybe even in the training and it is a good first step that can be maybe some
other aspects can be built into that. I see that as an opportunity.
DR. RO: I am officially taking off my public health Seattle King County hat
so I can say what I want to say. What I would challenge this committee to say
is who is not around the table and who is not here at the table speaking their
voice. From the Asian American and Pacific Islander community our challenge is
that we don’t have data on our ethnic communities. A lot of the aggregate
measures that we have talked about, a lot of the data tests that currently
exist we are just not there. The question is yes there is a lot of data, but if
you don’t have data and evidence already on your community, you are pretty much
at a loss still. Please keep that in mind.
As an advocate we are always asking the question of what data drives
decision making. We talk a little bit about cost effectiveness data that we
need, but I would also say that one of the areas that we didn’t touch upon
today though some of these tools that Todd talked about maybe helping — hats
to Sonoma — is around data visualization. How do we demonstrate the data in
effective ways that will resonate with policymakers, resonate with communities?
I don’t think we do that anywhere good enough. I can tell you as somebody who
manages a group of epidemiologists and social scientists I don’t think they do
it good enough.
And then on a completely other hat I also sit on the HHS advisory committee
to minority health. Data is one of our number one priorities because we work on
eliminating disparities and we have been writing letters to the secretary. We
have never interacted with your committee. There are so many committees in our
Federal Government. I would almost guarantee that several of you probably sit
on three or four of these committees. If we don’t pull together, we are just —
talk about lack of efficiency. Please reach out to the other committees as
DR. SONDIK: Well, I’m certainly not going to disagree with that. I learned a
great deal today actually and I thank you all for coming. I heard three
different kinds of data: one is sort of standard public health data sort of the
indicators that we think about. Second is what is not standard, the contextual
data the second panel really focused on. And then the third is what is not
really focused on in public health which is the clinical care data. I really
found that interesting. It really hit me over the head about those differences.
But it seems to me if what we are after is improving community data,
improving community health rather, not improving community data per se then if
we want to do that then we have to have some measures. We didn’t know what it
is we are measuring, what are those indicators. What are the key things to
know? I was struck by the fact that except for Garland really nobody talked
about the value of the information. And I think that is absolutely crucial. I
think we have to talk about that.
If we want to improve community health, we have to have at least some
canonical set of things that we are going to look at, some core set of things
that we are going to look at and we have to have some set of interventions to
know what to do about it even if we make them up from things that have happened
in the past. I think they need to be clear as to what those are and in general
they relate to behavior, they relate to care, and they relate to this — area
of data which is the context or the environment to changing that.
I think as we think about data I think we have to think equally about how it
is going to be used and that to me is crucial that it be part of the
MR. STEFANAK: As a former IOM study director, I always think even from the
first meeting in terms of final report. This is my draft, final report and
outline from what I have heard today. I think it is clear what is the topic
really. We are here to make sure we understand what the topic is. I think the
topic is actionable data at the neighborhood and subpopulation level. What we
have heard from are the actors who need the actionable data. Sometimes it is
the health department. Sometimes it is various groups that come out of NIH
studies or just grassroots development.
Where does this data come from? I think we have heard of three general
sources that I think all need to be looked at in terms of conceptualization and
the ways of improving because that is where the data is coming from. And the
first is there is a lot of client-censored data from program data at the
community level. South Carolina is the premier example of that of saying people
are using education, the education system, social service system, the criminal
justice system. They are homeless. They are in the housing data. And the issue
is how do we integrate and be able to practically use this data at the
individual, the neighborhood level, whatever the action is needed.
The second type of thing we heard about was primary data collection at the
local level and some cases very ad hoc. It is sort of sending high school and
college students out. In some cases it is based on BRFSS and YBRFSS and the
HIS. Are there better, more efficient ways so every community doesn’t have to
reinvent this wheel and then how do we deal with the resources to do that?
And then the third category I think we have heard about is clinical
electronic data. That is really divided into two areas. One is the HIE. We get
the data from that system. Do we need to do it through the bidirectional
exchange of data is directly between public health and clinical providers and
maybe we need to start that small? Maybe it is the community health centers in
that community until we understand how to get the data out, how to protect it
and how to use it appropriately. Those are the three data sources.
I think the questions about all those data sources are how do we get the
data out and organized. What are the analytic tools to convert data to
meaningful information? My biggest experience in that area was electronic
hospital discharge data was around for a very long time. Until somebody figured
out how to do inventory sensitive measurement, it really didn’t become useful
in terms of any kind of actionable community planning evidence.
And then I think the third thing we have heard is the analysts themselves
and the workforce training set of issues. I think those are the topics that I
think I have heard from today and their underlying themes within those topics.
MS. MADDUX-GONZALEZ: Well, I don’t know that this is new, but it may include
some of the things that already said. The theme that struck me is that we have
many areas of disconnect between public health and medical care in spite of our
shared goal of improved health. But the opportunity is I think through having
access to data and being able to share that data effectively amongst ourselves
and with our communities. There is tremendous cost ability to really connect
the dots of how community environments, population health and individual health
are all connected. In that way from the public health side from the health care
side we can develop strategies that are really aligned to effectively improve
health through combinations of community-based strategies and the medical care
DR. GREEN: That was fun. I think we have a good shot at getting out of here
just slightly early, which is almost always a good thing. My answer to the
question of the theme would be — the overarching theme for me was we are
missing the data. We are missing an infrastructure that we need to get from
where we are to where we want to be and the opportunity is define and fill that
The way I would like to wrap this up if it is okay is — I confess to Don
that when I go to meetings all day long and the challenge is to listen, I have
to have tools and techniques to help me listen. And one of the things I do is I
write down as closely as I can exact quotes from what people say during the day
and I am usually happy if I get five or six out of the day, but I have four
pages of them today. I am going to beg your indulgence. I am going to pretty
arbitrary say back to you what some of you said today as accurately as I can.
If you want to, you can turn this into a — you can bet on who — name the
people who said it. I wrote their names down.
DR. STEINWACH: And what is the prize?
DR. GREEN: The prize is we get out 3 minutes early. Health is all that
matters. I am sorry. Health is in all that matters. We have not met our goal of
asthma control because it was too ambitious. We need a big tent to include
everyone needed to improve health.
We would greatly appreciate your help in getting for public health the same
sort of support for information development that health care practices are
getting. South Carolina is very sensitive to government intrusion. Protect
people’s privacy. Have a framework of trust and over time you can expand what
can be accomplished for citizens and patients.
The trust we build is what lets us do our work. We are moving from community
involvement to community ownership. People really do care but we have to give
them something to rally around. I am still sitting on 2005 data that I can’t
find anyone with the skills to analyze. This is a very eye-opening experience,
I must say.
I don’t know where eyewitness testimony and qualitative analysis fits in,
but a local voice telling the story is pretty powerful. Structure not soup,
services not silos — start off and starts when you are ready. The Bronx can be
seen as a dull background from which comes some shining light. It took an
amazing amount of work to determine that the providers should be given credit
for a well-child visit.
If you think patients don’t trust insurance companies, try doctors. I have
spent most of my career in government and consider myself energetic, but not
compared to you community guys. What you see as a theme is something you
I think that today we have learned that to use other quotes something is
happening here. What it is ain’t exactly clear. But whatever it is it is going
on in a lot of America’s communities. And whatever it is there is a lot of
federal action and federal leadership. Simultaneously there is something going
on at the local community level and there is something going on at the highest
levels in our nation. We are just not sure how to connect them up.
If you will forgive me for overreaching just a little bit here, but I think
we might have been co-witnesses today to some of the laboring pains that are
going to be involved in birthing a new epidemiology that has never before been
possible until we have reached the part we have with our information
A lot of our conversation revealed a shift from sequestering data often for
private gain to liberating data for public good. This is a huge shift.
We started off saying that our goal today was to bring community voices
inside the beltway and into the room. I rest my case. We have done that. We
really have brought powerful voices into the room and inside the beltway. I
want to thank you so much you folks who came on such a relatively short notice
for doing all the pre-work. I think you have really been great. I want to thank
so many of the committee members for adding a day to your travel to show up and
be here. And as Marguerite said, we got a lot of help from staff and I don’t
mean for this to be a comment about anyone else, but I just want to say that
without Susan Kanaan this wouldn’t have come off the way it did and we are
really grateful to you.
(Whereupon, at 5:25 pm, the meeting was adjourned.)