[This Transcript is Unedited]
National Committee on Vital and Health Statistics
Population Health Subcommittee Workshop
Using Sub-County Data to Promote Multi-Sector Approaches for Community Health and Well-Being: Identifying Gaps and Opportunities
September 27, 2016
- Welcome – Bruce Cohen and Bill Stead, NCVHS Co-Leads for Population Health; Monte Roulier, Facilitator
- Community Health and Well-Being Measurement Framework,V3 12 – Bill Stead and Bruce Cohen and Gib Parrish
- Working Session: Community Health and Well-Being Measurement Framework
- Current Reality and Desired Future
- Peter Eckart, Illinois Public Health Institute
- Brita Roy, Yale University
- Leah Hendey, Urban Institute
- Kevin Barnett, Public Health Institute
- Working Session – Current Reality and Desired Future
- Federal Roles
- J. Alice Thompson, Centers for Medicare & Medicaid Centers
- Wayne Giles, Centers for Disease Contgrol & Prevention
- Jason Broehm, US Department of Transportation
- Elizabeth Sobel Blum, Federal Reserve Bank of Dallas
- Claire Wang
- Advances in Data Technology: Making Data Available and Easier to Understand
- Bob Phillips, American Board of Family Medicine
- Roxanne Medina Fulcher, IP3
- Mike Reich, Seabourne Consulting
- Collaborative Possibilities
- Summary and Next Steps – Bruce Cohen & Bill Stead, Monte Roulier
P R O C E E D I N G S
MR. ROULIER: Good morning everyone. Welcome to the NCVHS Population Health Subcommittee. I want to just quickly turn it over to the two co-chairs to formally get us started. Bruce Cohen and Bill Stead.
DR. COHEN: My name is Bruce Cohen. Since this an official Federal Advisory Committee meeting, I need to do a little business first, which means asking the committee members to identify themselves and declare whether they have conflicts or not.
I will start it off for the committee members in the room. I am Bruce Cohen from Massachusetts. I am a member of the Full Committee and co-chair of the Population Health Subcommittee. No conflicts.
DR. STEAD: I am Bill Stead. I am from Vanderbilt University. I am a member of the Full Committee, co-chair of Pop Health, no conflicts.
DR. COHEN: Other committee members, please identify themselves.
MR. LANDEN: Rich Landen with Harris QuadraMed, member of the Full Committee, the Standard Subcommittee, no conflicts.
MS. GOSS: Alix Goss. I am a member of the Full Committee. I am co-chair of the Standards Committee, the Review Committee, and I have no conflicts.
DR. PHILLIPS: Bob Phillips, a member of the Full Committee, the Pop Health Committee, no conflicts.
DR. RIPPEN: Helga Rippen, member of the Full Committee, the Privacy Committee, Data Working Group, and then the Population Group too. I have no conflicts.
DR. MAYS: Vickie Mays, University of California Los Angeles, member of the Full Committee, Privacy, Population, and chair of the Work Group and the Review Committee. I have no conflicts.
DR. COHEN: Great. Hopefully, that is as formal as we are going to be today. The rest of the day is going to be incredibly interactive. Thank you. I see that we have a quorum so we can do our public official business.
I am going to just start today with a brief overview of how we got here. You will be able to follow on the slides. The purpose really – we have four important goals for today. We want to enhance the public/private collaboration, focusing on sub-county data. We are going to be using the term sub-county, smaller area, community, neighborhood to pretty much cover similar concepts. Depending upon where you are from, your language might vary, but really the focus here is on small geographic or other affinity kinds of communities for which we need to focus on collecting better data.
The intent here is to really see how we can improve HHS data collection, focusing on these small areas.
And of course there are other secretariats besides HHS that have been doing this work as well. It would be great if we can figure out how to align and collaborative the efforts across secretariats.
The other major thrust of today is going to be looking to reinforce a multi-sectoral approach for measuring community health and well-being to broaden the disease-focused ideas and really consider community well-being and health and quality of community life from a much broader perspective and hopefully move HHS in that direction as well.
In our next session that Dr. Bill Stead will help lead and Monte will facilitate group discussion, we are really going to discuss a framework that embodies this notion of a multi-sectoral approach towards data. Throughout the day, we are going to be trying to identify opportunities and gaps in data that exist at this level and try to brainstorm about ways where the federal government can play an increased role to generate these data and promote small areas to collect data as well.
As I mentioned before, this is a group effort. We are building on the shoulders of many of you who have been in this area longer than the National Committee. We are not trying to reinvent the wheel, but we are trying to expand and align all of these efforts.
How do we get here? The National Committee is one of the oldest and best respected Federal Advisory Committees. We have lots of diverse interests. I would say over the last 10 or 15 years, we have been refocusing on population health. Over the last ten years or so, we have really been focusing on strengthening our understanding of how communities use data.
In 2011, we had a workshop to begin talking about communities as learning systems. All of this information for those of you less familiar with the National Committee is on our website and the links to the reports and recommendations that we have shared with the secretary there as well.
Clearly, we heard from communities that they need quantitative data, but certainly quantitative data alone is not the answer to addressing their issues. We have heard that continuously throughout our discussions.
One thing that we have added that I think was an area that has not been dealt with much is concerns about privacy and security. As more small-area data become available, communities need to better understand and consider some of the potential issues around protecting the data that they use. The National Committee has developed a toolkit for security and privacy that is on our website that is a really wonderful guidance as communities begin to use and collect their own data.
We have also heard in 2014 that there is enormous variation in the ability to use data and the readiness for data. As data collectors, we need to really work with communities to meet them where they are and to help support developing the skills and experience around collecting and using data.
I would say that the current phase of our activities began in 2015 and our workshop in November where we focused on measurement framework. This was prompted by our internal discussions and conversations with Dr. Denise Koo working with Karen DeSalvo.
I just got an email from Denise this morning. Today is her last day of federal service. I want to wish her well in her new life and thank her for her incredible input and insight in collaboration with us on these activities.
Denise helped us frame the Framework Version 1, which you will hear a little more about in our workshop in November of 2015. It made it clear to us that we had a lot more work to do.
In 2016, we commissioned Dr. Gib Parrish, who is right over there, to do a comprehensive environmental scan, not focusing on health indicators per se, but to do a survey of the field. This environmental scan is available on our website. It is a really great resource for those of you who are in this area of indicators.
That has led us to today’s conversation. We built on Gib’s work and Denise’s work and feedback that we have gotten from many of you about trying to develop this organizational structure as one of the elements in our efforts to move the Department of Health and Human Services forward in thinking about sub-county data and broadening our worldview about what data we can help support community initiatives with.
I am just going to stop here. I just wanted to provide a little context. I will turn it back over to Monte who can take us through what this day is going to look like.
MR. ROULIER: Thank you, Bruce. Good morning everyone. As Bruce mentioned, I have the privilege of serving as your facilitator today. I am going to do my best to really help maximize the time and the wisdom and the insight of this group over the course of the day. I have been really excited about spending time with this group today and I have had a chance to spend a couple of meetings with you the last two. I know there are a number of you who had been here before obviously on the committee. I have noticed there has been a lot of interest in what the Population Health Committee has been up to and I think particularly as it is honed into, as Bruce has pointed out, really honing in on multi-sector collaboratives and how do we really support the tools and capacities of these local groups, these place-based efforts to drive health and well-being and really recognizing this key ingredient of sub-county data in order to do that and to do that well. I think that is why there has been increasing – one of the reasons why there has been increasing interest in this work
I think we are really fortunate to have a really rich mix of perspectives here in this room. A couple of slides give you a little bit of a flavor of that. We have obviously from our federal partners. We have a number of HHS. Maybe we could do the old-fashioned data visualization in the form of hand raising if you do not mind. How about HHS folks here, some facet of HHS? How about other federal representatives, friends from other agencies, other areas outside of HHS? Wonderful. And then we have a number of health organizations, a broad bucket there. You can see associations, health care organizations of all sorts. Who is representing the health sector as a health care organization of sorts? Great. Some of you may actually have twofers here. Foundations and other not-for-profit organizations that are here? We definitely have twofers here. Are there any other big buckets that we have missed that do not fit into those? University folks. How about university? Thanks. This only begins. I think we are getting a flavor of the alphabet soup that always happens when we bring a lot of different associations and government groups.
One of the things I am going to ask is we be mindful of some of the acronyms that we are using as we introduce ourselves. I will not go through all of those on the slides. What these slides do not tell us is really that this room is filled with a lot of thought leaders. This room is filled with a lot of influencers besides the categories. I think having you all show up and agree to spend the better part of a day together holds enormous promise. As Bruce went through our proposed outcomes for the day, I think our agenda is pretty straightforward. Those of you who want to follow the agenda, it is here in the packet in that first left hand category.
Essentially, we are going to spend the first part of our morning anchoring ourselves in the Version 3.0, the Framework that many of you have been working on for some time. We are going to have a chance to give a little bit more input and more thought to that. And then we are going to spend the latter half of the morning actually really focusing in on what is really our current reality in this field of fields. What is really our desired future as it relates to this area that we are exploring?
After lunch, we are going to have a chance to hear from some of our experiences from some of our federal leaders. We are going to have a chance to really hear about where data and technology is moving and what that might imply for some of that around helping local communities, multi-sector collaboratives, access and use, data of all sorts, including sub-county data.
We are going to really end our time together with really surface and again what we see as possibilities. Some of those might be in the form of recommendations that the committee will take back and sort through. Some of them may be some other ideas for some collaborative opportunities given various stakeholders in this group. But that is essentially our agenda in a nutshell.
Does it make sense at a high level? Because we are going to be in small and large groups during a lot of this, I am going to propose just a couple of working agreements. I know how hard it is to get to meetings on time these days in DC. Thanks for helping us start pretty much on time. We will absolutely make sure that everybody is out here as the agenda suggests at the end of the day here.
Share air time. There is going to be lots of different opportunities in small and large groups. This is absolutely the place to test new ideas. There are restrooms. We are going to have a couple of formal breaks. They are right out there. Meet your needs. If you need to stand, do that. We will keep this as informal as we can in this space. Gentle on people and rigorous on ideas. We sometimes have that backwards in our culture these days.
Lastly, learn through dialogue. Since it is political season, I think we know what debate is or I think we do. I am not sure. Debate is assuming there is one right answer and you have it. It is about collaborative. It is about winning and dialogue with all these rich experiences and perspectives. That is why we gather this group here is to really discover how we have different pieces of this. It is puzzle learning. It is collaborative. It is about exploring common ground.
When we are in the debate mode, we are listening for weaknesses. We are listening for flaws. We are listening to reaffirm our positions. I am hoping and that has been the case the last couple of meetings is to really be in a spirit of dialogue to balance bringing your experiences because we want to really build on those and also being really open minded. For those of you who have worked with us before have heard us often say that dialogue. Come as you are, but leave as you come. Hopefully that we are all teachers and learners all the way through here today. That is the spirit that we want to make sure we capture today.
Any other thoughts or questions about the day and mode of operating or things that would be helpful to you?
I have heard informally talking with many of you that you are excited to be here. I am really excited that you all are here.
Bill, I think you are going to kick us off on this next assignment.
DR. STEAD: Thank you, Monte. First, just to complete our official nature, I saw Linda and Walter come in. Please introduce yourselves and state whether you have conflicts or not.
MS. KLOSS: I am Linda Kloss, member of the National Committee on Vital and Health Statistics, co-chair of the Subcommittee on Privacy, Confidentiality and member of the Standard Subcommittee. I have no conflicts.
DR. SUAREZ: Good morning everyone. I am Walter Suarez. I work for Kaiser Permanente. I am a member of the National Committee and chair the committee at this point. I welcome you all on behalf of the National Committee as well. Thank you. No conflicts.
DR. STEAD: Over the next 40 minutes, we will review the purpose of the measurement framework, summarize the landscape we uncovered through the environmental scan, briefly review Version 3, and how its domain support epidemiological life course and health equity perspectives of upstream and downstream determinants of health.
For those of you that have not tracked the insider baseball over the last year, Version 1 is the version that Denise Koo drafted with the Office of the Assistant Secretary of Health. Version 2 is the version that we developed with the input of the environmental scan for broad vetting. We put it out in June. Version 3 is what has resulted from the feedback we received on Version 2. The folder has the three versions so you can see how they progressed. All of this is designed to get you ready for the small group activity where you will be able to give us additional feedback.
I want to begin by reading a story by Dave Ross who is one of our members, describing the reaction of community leaders in DeKalb, Georgia to Version 2 of the Framework. These are Dave’s words. Dave was going to try to be with us on the phone. I do not know if he is on. We thought it best for me to read it in the room. For the past year and a half, the Task Force for Global Health has been in conversation with several of the Atlanta Metro Area County Governments about their options for developing a health promotion agenda.
The NCVHS draft framework presented with a unique opportunity to explore the framework’s utility as a tool for helping community leaders focus their attention on health problems that need action.
I will offer the following overview of the project that is beginning to emerge in Atlanta. I will provide this as an example to explain why I think the framework can make a significant and important contribution to advancing population health.
My story begins with a discussion I had in 2015 with DeKalb County Chief Operating Officer and the chair of the DeKalb County Board of Commissioners. DeKalb County, Georgia is the home of the CDC, Emory University, the Carter Center, CSTE, the National Association of Chronic Disease Directors, and the Task Force for Global Health. We are county rich in health expertise, yet also a county where local leaders have almost no readily available data to help them improve and protect the health of their citizens.
The county’s chief operating officer oversees daily management of government operations while the chair is an elected official responsible for establishing overall direction for guiding the county commissioners. They rely on their public health department to advocate for health improvement and to provide necessarily population health protection.
The great recession of the past few years led to significant reductions in county government staff including the health department. Consequently, when they think about health improvement initiatives, they confront a staffing and funding challenge. The recession has resulted in a county that has stalled in its understanding of health disparities, opportunities for prevention, interventions, and stalled in having an active connection between the health department and the county’s political and management leaders.
They are seeking a logical and efficient way to establish a county-wide health improvement agenda. They are fully aware that social, behavioral, economic, educational, and other factors play an important role in determining individual life course as well as population health status.
They also understand that the county is not monolithic. It is highly diverse, patchwork of economic subgroups, economic enclaves, ethnic enclaves and neighborhoods.
They are aware that their county faces health issues that are not simply resolved by a visit to the doctor like a high prevalence of overweight children, increasing deaths due to gun violence, and asthma caused by bad air. They know that the burden of these problems is not uniformly distributed amongst the various subpopulations in communities that make up the county. Despite this awareness, they have little evidence that motivates an action agenda.
The health department has been reduced to offering basic mandated services. Absent data that paint a clear and compelling picture of the county’s health needs, a tool that shows sub-county-level data around social and behavioral determinants. They remain paralyzed to act.
In the course of our discussions, I showed them the draft framework under consideration by NCVHS. They recognized the logic of the domains and the inherent validity of the subdomains and example indicators. They instantly understood that having sub-county data on these subdomains and indicators would open a view of the county they govern that has heretofore been invisible to them.
They quickly provided examples of subpopulations they know about such as a section of low income Hispanic and Asian communities that abut a high income, high-education neighborhood. Presently, they only have access to state data or at best county aggregate data and only a few health metrics. They know that they need sub-county data to make visible underlying inequities in health and/or opportunities for health promotion.
I asked them a simple question. If the NCVHS framework existed as a simple-to-use computer tool and the tool was able to access sub-county-level data for their county and possibly surrounding counties, would they use it? Their answer tells a story. Without reservation, the answer was yes and they added an emphatic when can we have it.
Based on these interactions and with the framework guiding us, we are mobilizing county government officials, the health departments of three counties in Metro Atlanta, health care provider organizations, public safety officials, business sector leaders, and the boards of education to form a health-planning task force aimed at understanding where health needs exist and formulating proposals for action.
Problems these leaders know exist in general yet lack the facts to act in specific ways or in specific populations include early childhood violence prevention, land use planning to promote walking, and bike friendly environments, food deserts and substance abuse. The framework has provided a straightforward, easily understood means for seeking data that will highlight problems and indicate possible directions for interventions. Most importantly, the framework has given them confidence that this tool will lead them toward a countywide health advocacy agenda, which they presently lack. That is why we are here together.
Briefly, the purpose of the framework is to assist communities in gathering and using information while enabling the federal government to better organize and unify its approach to collecting and generating data at the local level to improve community health and well-being. Specifically, to strengthen multi-sectoral efforts at the local level, to help HHS and other federal agencies and private-sector partners identify and close gaps in the availability of data at the sub-county level.
By sub-county, as Bruce said, we mean the smallest possible geographic unit that permits meaningful planning and project development. Depending on local characteristics, the meaningful unit may be a neighborhood or a small town or a group of communities or even counties.
It is supposed to offer communities a blueprint of the key issue areas, domain and subdomains, and to ease conversation across sectors to identify opportunities, design interventions, and measure progress. It is to promote public and private collaboration to build on the successes of numerous existing efforts.
Let me draw your attention to the frequently asked questions that are in the folders at your tables. They add color commentary to the purpose by highlighting what the framework is designed to do and what it is not designed to do and to answer other questions raised during the vetting process.
The framework is designed to provide a structure for thinking and talking about measurement across determinants. For example, providing a way to organize a menu of indicators and metrics. To assist in aligning federal data efforts across secretariats, to facilitate community-level measurement, to draw attention to multiple domains to design interventions and measure progress, and to enable assessments through the lenses of health equity and life course.
The framework is not designed to replace other measurement framework efforts, provide a comprehensive list of indicators or to prescribe a set of indicators. It is not designed to rank or compare communities. We want communities to be able to use sub-county data to compare pieces of themselves to figure out how to apply resources. It is also not designed to stand on its own. I draw your attention to the Roadmap to Community Level Health Measurement, which is also in the packet, for an overview of the subsequent activities we anticipate through public-private collaboration.
Let me briefly introduce the environmental scan and then turn it over to Gib Parrish. Participants of the 2015 workshop confirmed the need for a parsimonious measurement framework and were of one voice that the time to act was now. What we know is imperfect, but we know enough to act. We need to do so.
However, the Version 1 draft needed a lot of work. In particular, the domains were at various levels of granularity and omitted key determinants. We also heard that non-health sectors might be ahead of public health and health care in this work.
In response, we contracted with Dr. Gib Parrish to work with the Population Health Subcommittee to develop an environmental scan. The goal was to scan health and non-health sectors, i.e. transportation, housing, education, environment, et cetera to identify existing measurement frameworks, core domains, indicators, metrics for community health and well-being.
And then to provide a meta-synthesis to guide development of V2 of the Framework. If you have not seen the first several pages, those tables will be very helpful to you. And then note examples of indicators or metrics that are available. And by available, it means they are accessible, estimate-able, or collectable at the sub-county level.
With that, I will turn it over to Gib.
DR. PARRISH: Let’s go to Slide 3 because Bill made a fabulous introduction of what I did. Basically, I did what Bill said. We took suggestions from the meeting last week to design health sector indicator systems, rankings, and indices. I compiled with their help 47 such things. They are listed here. The main categories that we found are listed here with the number of systems in each and a few examples. We found, for example, community and neighborhood indicator systems. These are ones generally developed at the community level by various community members and organizations.
There are well-being indices. These are typically developed by national or international organizations. Poverty depravation and inequality indices. These are mostly used at the national or international level. A lot of them have been developed by the United Nations or its affiliated organizations.
There are a number of livability metric indices and rankings. This is the one group that have been used and developed in the private sector. Money Magazine, for example, has best cities to live in the United States. They are actually quite an interesting group of using indices and sometimes rankings. There are a couple of other sector-specific indices. I did not dwell on this because they were quite specific to a sector, for example, economic indicators.
And then at the request of a number of population health subcommittee members, we actually did look at a pretty good cross section of health sector indices and there were 15 of those that we looked at. A few are listed here. You can look in the full report for the listing of all of those.
Basically, we made a few observations about what we found. First of all, there were many more indicator systems, indices, and rankings than what we found. There are probably several hundred of these. I looked at 47. But just to let you know that there are more. I think that the group that we identified was pretty representative of that larger universe, but we certainly did not go through all of them.
They have been developed by many different kinds of folks. As I already mentioned, community or local organizations versus expert groups. The Institute of Medicine has developed several groups. There are universities, national and international organizations as well as government agencies.
Again, experts versus community members have been used for different ones. Sometimes there have been community surveys to get input. Other times they have been almost purely done by experts.
There are two groups I just want to point out. The Urban Institute and its National Neighborhood Indicators Partnership has worked since the mid ’90s, developing very nice work on community indicator systems and the Community Indicators Consortium, which actually the meeting today and yesterday here in Washington has also done very nice work, pulling together information from various community data sets.
As Bill just mentioned, there are a few other federal agencies that are a bit ahead of where we are here. There is recent very interesting collaboration that occurred between EPA, HUD, and the Department of Transportation on a set of livability principles and then from those, each of those organizations has developed sets of indicators for use at the community level.
The terminology. There are a lot of different terms that are used. We stop with the use of the word domain as the highest level term for categorizing these indicators, but other things like category, topic, sector, et cetera. Below that typically people use the terms indicator, measure, or metric.
There are different conceptual levels of the domains that are included in these different sets that range all the way from a concept such as well-being itself, which is a very high level concept to something like receipt of means-tested benefits, which is very specific. That varies in these different indicator systems.
Sometimes a community indicator system may actually put the same indicator in different categorical areas. For example, percent of population with a high school diploma might actually be included under education, demographics, the workforce, and the economy. They sometimes will spread the same indicator across different domains. And some people typically think of an indicator and a domain, but that is certainly not the case in many of these indicator systems.
Finally, some of these systems pull all of their data together into some sort of a summary measure, an index or ranking as opposed to just using the individual indicators where some others in fact just use the individual indicators.
This is just an example, as Bill mentioned, of the report where it lists, for example, these are some community indicator systems and the main domains that are included in there. No need to look at the specifics here, but to get a sense that we compiled all these. We got several hundred different domain names, roughly about 150 that were pretty unique.
From that, we actually pulled together – this is the meta-analysis or synthesis. We pulled together which particular domains were most frequently seen in these various indicator systems, rankings, and indices. And the ones that are highlighted in green here are ones that are actually in Version 3 of the framework, which Bill mentioned just a few minutes ago. But there were some others that we found, which actually were either subsumed under these or on occasion are not included. For example, arts and culture or something akin to that is frequently mentioned in a number of the indicator systems, but we do not really have a category for that in the framework.
This just puts side by side the three different frameworks, Version 1, 2, and 3. This is in your packet. You can pull that out. Bill was mentioning that in his remarks as well. The only two items, education and housing, which were in Framework 1, went all the way through all of the frameworks. The ones which are in green are new to the frameworks. For example, health, environment, economy, public safety, et cetera were new in Version 2. And in Version 3, food and agriculture and community vitality were new, but otherwise there is some continuity particularly between Version 2 and Version 3.
The final thing that I was asked to do in the scan and that we worked on was to identify some sources of small area data. This slide lists some of these here. The American Community Survey is probably the best source and the most frequently used source of small area data in the various indicator systems that we reviewed. Certainly, the Decennial Census as well. And then there is a lot of locally-generated data that people have used for developing community-specific indicators. Examples being school system data, tax assessor records, planning information, voting records, housing authorities, et cetera.
This just gives a very quick overview of the information that is available from the American Community Survey, at least the 2015 version of the survey. Many of you may be familiar with this. I will not go into the details, but there is a lot of information ranging from housing, housing composition, food assistance, insurance, disability, et cetera. It covers a number of different interesting areas that again were found in the various indicator systems that we looked at.
That is it. I tried to leave a couple of minutes in case there is anybody who has a question about what we did. This is a very quick overview. The full report, which outlines all of the indicator systems, is on the website, as has been mentioned, if you want to go look at that. Let me just stop for a second and see if anybody has a question about what was done or what we found.
DR. COHEN: All of the slides that you see will be on the NCVHS website.
DR. PARRISH: Thank you very much. And thank also all of you who contributed to your comments to this process. It was very helpful as we moved through the process in terms of making the final framework as good as it is.
DR. STEAD: Thank you, Gib. It has been really wonderful to work with you on this. Version 3 of the framework has 10 domains and 23 subdomains. As Gib mentioned, we define domains as categories or spheres of activities and conditions and information that constitute or characterize human societies, nations, populations, or communities.
We tried to level the domains to make them parallel in scope. We define subdomains as more focused categories within domains that include issues of concern for community health and well-being. The structure of the domains and subdomains does not reflect weighting of the contribution to health and well-being.
Let’s just read through them briefly. The first domain is health. The four subdomains are health care infrastructure, health behaviors, health conditions and diseases, and health outcomes. Next is environment with natural environment and neighborhood characteristics. Then education. Infrastructure and capacity, participation and achievement. For economy, income and wealth and employment. Food and agriculture, food availability and nutrition. Public safety, infrastructure, perception of public safety, crime, and injuries.
Community vitality has social capital, governance, civic engagement, and social inclusion. Housing has infrastructure and capacity, quality, use and affordability. Transportation has infrastructure and capacity, quality, use and affordability. And then demographics have total population, the Affordable Care Act variables and other demographics.
The framework provides an outline, if you will, of the sectors and categories of indicators and metrics within sectors that can be drawn on. Page 4 to 9 of Version 3 presents a sample of the framework filled out with a limited set of example indicators in metrics. We define indicators as specific narrowly defined activities and conditions whose state or level is measurable. Metrics are quantitative measures, including units for expressing the metric such as rate per 100,000, the population measured, and the source of the data.
The sample shows examples that are currently available at the sub-county level and examples of ones that are not as a starting point for our effort to identify gaps.
Just to summarize how this framework fits with others. It was developed with the environmental scan as its evidence base. That and the feedback we received is the base for what we are proposing.
The guiding principles we use for development were that the similarity with frameworks being used by neighborhood indicator projects. We wanted it organized in a way that aligned with federal agencies to promote collaboration. We want each federal agency to see metrics that they are familiar with in words they are familiar with, sitting side by side in a way that can be brought together for collective impact on health.
It is not considered a replacement for other framework efforts. Rather, this serves as a convening framework to support alignment, federal involvement, and collaboration.
There are several conceptual frameworks and models that exist to show the relationship amongst upstream and downstream determinants of health. This is a slide of George Kaplan’s model from the University of Michigan. It is a multi-level model that shows how individual factors interact with social relationships and community compositional and contextual factors across the life course and in association with the environment to determine health and well-being.
This is a graphic by Neal Haflon at UCLA that depicts the change in the relative magnitude of the influence of individual, family and community determinants of health across the life course. The measurement framework is intended to help communities see the breadth of domains and subdomains that interact to determine health and to selected indicators or metrics that call to them.
If a community decides to take a life course perspective and align policy and resources by stage of life, they can select indicators that apply to an age range, such as school lunch programs in the food availability subdomain, or measure metrics, such as food insecurity in the nutrition subdomain as rate per 100,000 for different age groups.
If the community decides to take a health equity perspective and align policy and resources to eliminate avoidable disparities between sub-populations, they can select indicators that speak directly to bias such as perceived racism in the social inclusiveness subdomain, or measure metrics, such as food insecurity in the nutrition subdomain as rate per 100,000 for different demographic groups.
During July, we emailed Version 2 of the measurement framework to a wide variety of public agencies, private organizations and individual experts to obtain feedback specifically asking the following questions. For each subdomain, is it useful in the community context? For the framework, does it make sense? If not, why? To what extent would it be useful to help galvanize community activities to improve community health and well-being? Are there domains or subdomains you would add or modify to make this more complete from your perspective?
We received over 100 comments from nonprofits, state and local governments, health organizations, providers, academic institutions and federal agencies as summarized on this slide.
The feedback we obtained fell into four categories. First were concerns about domains and subdomains. We added the pieces in Version 3 in response.
Second were requests for specific indicators. We have tried to clarify both the purpose part of the framework and the frequently asked questions that this is designed to organize whatever indicators and metrics that exist. We are not actually trying to include those in the framework at this juncture.
Comments on the relationship between indicators and questions regarding how the framework would be used.
This slide is a different view of how the domains and subdomains have evolved as we move from Version 2 informed by the environmental scan to Version 3 based on the vetting of Version 2 over the summer.
Between Version 2 and Version 3, we added achievement to the education subdomain to reflect the move beyond attainment and measurement at community levels. We promoted food and agriculture to the domain level to equalize scope. We extensively reworked community vitality domain, calling out social capital, governance and social inclusiveness. It is safe to say that the governance area – all this is around the capacity of communities. It is safe to say that the indicators and metrics in that space is among the ones that are least developed. It is clearly – fall into the gap category.
To reemphasize a point that Gib made from the environmental scan, the overlap in domains is intentional to allow us to create parity across sectors. An indicator such as bike lane accessibility may be in both the infrastructure and capacity subdomain of the transportation domain and in the neighborhood characteristics subdomain of the environmental domain.
Let me turn it back over to Monte to set us up for the small groups.
MR. ROULIER: Thank you, Bill and Gib, for going through all that very clearly. I think for some of you, if you are absorbing this for the first time, there is a lot to absorb. Many have done some heavy lifting on this over the course of the last 12 months plus. I think our intent here is not to open this up. There have been a lot of iterations, as you have heard, around feedback on it. There certainly will as an evolving framework continue to be opportunities for that.
I wanted to give you all an opportunity here in this first part of the morning since we are going to be working in large and small groups is to actually get to know some of the folks of the table. Maybe just go around your table and what brings you here today. What is your interest and hope for this meeting? A lot of you are here not because you have to be. You are using your discretionary energy to be here I think for the most part.
And then any other kind of high-level thoughts on the framework. In particular, there are some ideas out there around where this might be valuable. We heard David Ross’ message. Where do you see the greatest value? We are going to loop back to where we go with this framework over the course of the day. This will be the first chance to weigh in a little bit on that.
I am going to give you about ten minutes as a group and we will check back in with you.
DR. MCGINNIS: A question for clarification to Bill. Actually two comments and two questions. They are brief. First comment is the taxonomy is terrific. It seems to be a very nice way for communities to assess the key issues – secondly, the systematic. I am a first-time viewer of this. The systemic way —
I am Michael McGinnis from the National Academy of Medicine. Just to repeat quickly. The first comment was a kudo for the nature of the framework itself, which I think reflects in a very systematic and coherent fashion the key issues involved.
The second observation is the systematic way you have gone about it, not only in terms of the consultative process, but I was especially intrigued by and interested and informed by your frequently asked questions. That raises the question I have or two questions. One is when you are doing a taxonomy, generally the domains that you identify are either grouped or clustered with some intent. The order, for example. I was not quite clear on what intent you had with the order of the domains because community vitality for some reason does not fit into the flow of things. It is a minor question. But I was just curious as to whether there was background on that front.
And the second question, which is a little more important, is in your list of frequently asked questions, you said you specifically were not intending to provide prescribed list of indicators. But presumably, as you indicated you are going down the line at some point on formal indicators and because communities need a finite set of indicators to compare themselves to other communities, what is the thinking of the subgroup on that count?
DR. STEAD: Thank you, Mike. Let me take them in order. First, to my knowledge, there is no logic in the order of the domains. My colleagues at the back of the room are all shaking their heads yes. It may make more sense for us to alphabetize them because we are in fact interested in them having parity. We are trying to have them have parity, have them connect the key stakeholder groups, not trying to organize them to reflect any form of causality. That is a good point. I think we probably ought to alphabetize them. Other suggestions would be welcomed.
The question around prescribed indicators. What we are expecting is that the framework will be used to allow us to create a menu of indicators. We think that menu would allow indicators to fall in as many domains and subdomains as appropriate. This is trying to get parsimony at the domain and subdomain level, not at the indicator level.
It is also not trying to allow people to rank or compare communities. We are not trying to compare two different counties to one another. We are actually trying to let a county compare the side-by-side hot spots and cold spots and identify at ground level how to refocus resources to make a difference. This is very different. We have tried to absorb and provide ways to plug in the IOM core metrics idea. But this is a very different approach because we are not trying to get parsimony at the indicator level.
We are hoping that this will move efforts and we are coordinated with them such as Healthy People 2030. We are hoping this will actually move that effort more toward parsimony then to more and more indicators. But we are not trying to get to the level of parsimony that core metrics would have for comparison across the country.
MR. ROULIER: Michael, thanks for those really important clarifying questions. I think one of the reasons why set up the agenda the way we did and having this as a little bit of a level set in terms of what might be the possibilities in terms of a menu metrics, what might be some ways that we could work together to further evolve this. I think part of what will happen in the middle of the day might give us some insight into answering that again at the end of the day.
Again, thank you for helping out. I am going to invite you just again to take 10 to 12 minutes in your small groups again to just get to know each other, what brings you here, what is your interest, and where you are starting to see some of the value in this framework.
(Pause for Group Discussion)
MR. ROULIER: We are just going to start our conversation that I think is going to take us throughout the course of the day really. But I just wanted to give you a chance to pull out what are some of the nuggets, some of the themes that seemed to be coming out of your conversations. First, I am curious if there is anything that you heard in terms of why you are here as you are listening to other folks and then we will turn to the framework.
I would just open it up to whoever would like to speak up. If you could use the microphone and share your name, that would be terrific.
DR. DEUSTER: I am Patty Deuster. I am from the Department of Defense and we are starting a pilot project called Building Healthy Military Communities. I think this is exactly what we need. We did not know about this until about a month or two ago. I am just delighted to be here.
DR. HOMER: My name is Charlie Homer. I am from the Office of the Assistant Secretary for Planning and Evaluation or ASPE at HHS. The first observation was just an incredible richness of expertise, knowledge, and passion at our table and I am sure at all the other tables. I just enjoyed immensely learning about many of the initiatives and programs and organizations around the country.
I think in terms of the conversation, I would say we elicited a healthy tension between the openness of the framework and the numerous opportunities for communities to identify their own variables of need against some of our desire to say it would be awfully nice for communities to be able to learn from each other. If they are all using different metrics, it is going to be a little hard for communities to learn from each other. I would say that was a beginning of a robust dialogue, which of course was not resolved and I expect will never be.
MR. ROULIER: Well said, Charlie. I am just curious. Very diplomatic and nice way of putting this. Creative tension between flexibility and openness and the need to have some common metrics that would allow for either comparison or learning across. How many other tables touched on this creative tension, if you will? Any other insights that you want to add, recognize we are not going to try to resolve that in the ten minutes?
DR. STOTO: I am Mike Stoto from Georgetown. Some of the other issues were the availability of expertise in counties to be able to use measures. The develop measures is actually pretty limited. The natural tendency of people to want to benchmark to compare one another as a way of learning, not just learning about what works, but learning about how they stand and whether this is important.
And the third thing is that we have seen in a lot of activities that it is pretty much the same issues that are the important ones in a lot of different communities. It is not like there is all that much uniqueness. Obesity, opioids. You see over and over again in community health needs assessments.
MR. ROULIER: Thank you. This whole idea of not just the availability, but the capacity to identify and appropriately use is an issue I think that will continue to pop up today.
PARTICIPANT: Data availability and sub-state metrics were a big topic of conversation here at the table and the lack of.
MR. ROULIER: And why we are here.
DR. PERLA: Rocco Perla with Health Leads. We talked about the opportunity to do some real sense making here and looking at the evolution of the versions and how that was refined over time. I think we also jumped to implementation. How do you actually take this data and move it into action and the opportunities there?
And then a little bit about the multi-sector opportunities. If you would create a master Venn diagram and put all the stakeholders together, is there that sweet spot where these measures leverage all of the assets that we have as a collective community to drive population health?
MR. ROULIER: That is a nice set up and again a tension that I hope we carry throughout the day around what does this look like to put into motion.
One of the things we were just talking in the small group in between – this is Version 3. Presumably, there is a Version 4 and there is some openness to thinking about how this really evolves and starts to manifest in some ways that would really be helpful to local communities.
DR. LANDEN: The eight of us over here brought a very interesting set of backgrounds all of us unanimity about this tool fills a need both for professional purposes, some for individual where we live at home. Unanimous endorsement of the tool that it meets a need and will be used eagerly awaited.
A couple of specific comments from the group is one of us would like to see some more inclusion specifically around gun violence. And the second point was an interesting thing about what about comparability over time. It was not clear to any of us whether this could be used as a metric over time for an individual sub-county unit or not. That might be an issue we want to look at as the larger group.
MR. ROULIER: You said comparability over time so not just the comparability, but really looking at impact over time.
DR. BEATLEY: Noreen Beatley, Healthy Housing Solutions. I am actually paraphrasing I think what Andrew said more than anybody else. We talked about the availability of data like the other table, but more than that is the control and the management of it and the collection of it. One, finding the source is really difficult, but then who keeps it up and who makes sure that it is right and that it goes back to being comparable? We are hopeful that the electronic data system that has been put in place has some value and can be used, but that is still questionable right now.
MR. ROULIER: Thanks for putting that out there. I think you just named another tension that I am hoping actually some of the panelists might actually speak to around the debaucherization in some ways, of data and the quality control and who controls that, updates it.
DR. WAXMAN: We also really found it valuable. I am Elaine Waxman from the Urban Institute. There was an area where we were not sure where certain important measures would go and that is child care and other child well-being that is not related to education so child care not attached to a school, foster placement and things that were not immediately obvious.
MR. ROULIER: Didn’t really see those clearly in existing set of domains. We will touch based on that. Just so you know, we are capturing this in a couple different ways – recording, but we also have one of the best recorder synthesizers on our team here, Susan, that is capturing it by the computer.
DR. MAYS: Vickie Mays, member of the Committee. One of the things that came up is the issue of tools that may be needed to do analysis. As we get into these smaller and smaller community levels, we are really going to have to think about whether or not we have the statistical methods and tools that we need in order to be able to produce results that are going to be solid and evidence based enough for them to be actionable. As it stands now, where we need data the greatest are often in unique smaller populations, but we do not yet have the ability to be able to present a case that is statistically sound enough to be able to go for funds. As we think about tools, that is one of the tools that we need some thought and funding for.
MR. ROULIER: Thanks Vickie. So some case-making tools. I assume there are some other analytical decision-type support tools.
DR. BLUM: I am Elizabeth Sobel Blum from the Federal Reserve Bank of Dallas. One thing we talked about was let’s say we had perfect data in terms of collection, analysis, management, et cetera. What do we do with it? What decisions are made? I think there has to be a conversation about what is the goal. Is it about utilitarianism? Is it about we are trying to create the greatest good for the greatest number of people? Are we trying to reduce disparities? We have to be very explicitly about that because as we have seen with the Affordable Care Act, at the bottom of the debate is access to quality, affordable health care, a right or a privilege, that is a very values-based question. When we talk about data, how it is going to be used and for what purpose? What is the end goal? Have a mission of what our goal is so that we know how are we going to be using that data in a way that reaches that goal.
MR. ROULIER: Thank you for putting that out there. We are going to get a chance to hear from you after lunch. Wonderful.
DR. WANG: I am Claire Wang from HHS and Office of Assistant Secretary for Health. In this table, we talked about there is a common thing in terms of how local health departments and local public health agencies are able to in my boss Karen DeSalvo’s word public health 3.0. They have to be the chief health strategists for their communities. In order to do that, the data sharing and data linkage is a challenge of cross sectorial data linkage. In a lot of communities as well as in the federal government, that is very difficult for reasons that need to be worked out. Maybe less technical, but more political is to cross link data from different agencies and different bureaus. That is just an additional layer to this discussion.
MR. ROULIER: The layer and a real opportunity it seems a premise of this group.
MS. KLOSS: Thank you. I am Linda Kloss, member of the Committee. I just want to call everyone’s attention to another work product of our national committee working hand in hand here with the work of the Population Subcommittee we have published. It is on the website of stewardship framework for community health data, starting to get at the issues of de-identification of data and some of the data management issues that have come up. That was published about 18 months ago and it may be of some use to your organizations and certainly we are committed to keeping that moving forward as this project evolves.
The NCVHS’ Subcommittee on Privacy, Confidentiality and Security of health data.
MR. ROULIER: Thank you. Anything else that anybody wants to say before we take a brief break?
DR. NORRIS: Just to follow up a little bit on the issue of the use of data, which was mentioned. That is that in my involvement with community health assessment and in a number of different places over time, one thing that was quite interesting was to see the importance of local involvement in the selection actually of indicators and projects that they wanted to do and that sometimes they could not find from existing sets of indicators anything that really met their particular needs that they viewed. And that community process of getting people together thinking about what they are interested in often was the buy in that was needed to actually try to do something about a particular problem. That is something else where some of the local decision making I think is important in the overall process.
MR. ROTHWELL: Charlie Rothwell, NCHS. A couple of thoughts. One would be that whatever the indicators are decided upon, I hope that this would be maybe tested in a few communities before it is blessed so that we can see what it would take to manage it.
Number two is I can remember many decades ago when folks came up with a good idea that now is called Healthy People. Unfortunately, Healthy People was so healthy that the indicators became so many that we really even tell where we are nationally. I think that is something to learn here is – I think the word was parsimonious – I think we need to make sure that if we are going to recommend something to counties, and Mike pointed that out, who do not have the capabilities that folks do maybe at the national level. We need to make sure that we are not wasting their time. This is something that they can really utilize. Just a thought.
MR. ROULIER: Thank you for offering that.
MR. KAHN: Kahn(phonetic) US EPA. First of all, having been in the workshop last November for framework Version 1, I really liked to praise and commend the committee members and the contributors to see the Framework 3, which is a phenomenally well-developed, very comprehensive domain in terms of achieving sub-county level health care data.
A few comments if I may please. In the environment domain, of course being at EPA, we do have a slightly different take on the environment. It is a natural environment and built environment, which actually is related to several domains here. And also what we call social environment, which is – say the same thing about that.
Especially in the environmental domain, there is an environmental justice screening tool at EPA. We are thinking how we can incorporate into this framework here.
And also I am very glad to see the food and agricultural domain. Being at another national committee for food and agriculture subcommittee. Food availability also. Food safety is a very important issue to be remembered.
Lastly, I just like to second or third on the comment on the data management and availability. It is incredibly difficult to realize very fragmented even as getting health care data from Indian Health Service for tribal communities. We found it extremely difficult and fragmented. Thank you.
MR. ROULIER: Thank you for being here and for the great framework that you all have been producing at the EPA. A couple more comments.
DR. LAURENT: My name is Amy Laurent. I am representing the Council of State and Territorial Epidemiologists. But in my day job, I am an epidemiologist with Public Health in Seattle and King County. We also talked about the issues of the data fragmentation and how you work across sectors.
But I also want to talk about the concept of being the chief health strategist because one of the issues that we have is that public health throughout the United States is so very – the ability of all the health departments is so highly variable. You have really well-resourced places like Public Health in Seattle and King County, but even in Washington State, there are counties that do not have an epidemiologist. They have a part-time epi or somebody who is doing community health assessments.
One of the other pieces that I think could actually be helpful is also taking the framework and building out a logic model to help people understand how do you get from point A to point B. Why is the natural environment important? We take for granted that people understand those concepts, but I think that that is not always the case.
And the other thing is really helping the framework get disseminated because we have done a lot of environmental scans and seeing what Gib has done. We did a lot of that. That could have made our jobs a lot easier. Just making sure that people really – that this does really get well disseminated and having maybe some champions in state health departments can be one of the ways to help do some of that.
MR. ROULIER: Great ideas. I saw a number of heads shaking as you were talking about some pathways to really give some guidance for communities that may not have epidemiologists or other kind of experts that maybe goes along with some of the capacity building.
This is the last comment before break.
DR. HUNTER: Ed Hunter. I am at the de Beaumont Foundation. I guess two quick comments. One of them – on your point of what do you see as the greatest potential value. All of us that are coming from our history in the data world go right to what is the indicator. How is it comparable? What is the capacity? How do you manage the data? I want to back it off one level to what are we trying to tell people as important. What I like the most about this framework and the biggest value is actually shown in the evolution from one, two, and three. It is pushing public health and health discussions at the community level from what is the health indicator, what is the outcome, what is the obesity rate, what is the death rate to these are the things that matter. In the 3.0 model that Karen talks about, this is the discussion. It is cross sector. It is multi-social determinant issue. Even if we never get data, we are telling communities these are the things you should be aspiring and worried about and trying to get data on.
Second quick point. Speaking to the capacity issue and – Seattle and King County, public health does matter, not just health care infrastructure. I wonder if there is a way we ought to be signaling if we are trying to tell people what to aspire to. Public health capacity matters. If your community is looking at what is a good indicator, there must be some indicator on having a strong state or local health department that can actually do some of these things, pull some of these indicators together and help coordinate across the sectors.
MR. ROULIER: Thank you for that. Rebecca, it sounded like you had one quick comment.
MS. HINES: One quick tying up comment. I really appreciated the last two comments. You will notice on the agenda that one of the goals here is leaving here hoping to foster a public-private collaboration. This committee is a body that just makes recommendation. We do not operationalize this. There are a lot of people in this room. We have the head of ASTHO. We have other folks here who could take some of these suggestions. I encourage you today to think about roles that you could play and contributions you could bring to this from where you are sitting. If you are in a different department, if you are BLS or EPA or wherever you, how you can think about bringing this together from your particular organizational perch.
MR. ROULIER: Great segment. Let’s go ahead and take a break and come back at 10:30. We are going to get started right at 10:30. Thank you.
MR. ROULIER: Great conversation earlier on. I think one of the higher level takeaways just to name it was that V3 heading to V4 to be determined is very directionally correct that there is some real utility value there and a lot to still explore, which I think is great news and a testament to a lot of the hard work that has happened again over the last couple of years, but particularly the last eight or nine months. Well done to all of you who have been moving that forward.
This last half of the morning we are really going to continue to delve a little bit more deeply. Where are we now? What is our current reality as it relates to multi-sector collaboratives trying to use data and particularly sub-county level data to drive health and well-being? What is really our desired future to set us up to be a little bit thinking strategic around what are our opportunities to act –
We have a great panel that has agreed to kick us off. Kevin Barnett, Brita Roy, Leah Hendey, and then Peter Eckart. I am going to allow them to go through and spend just a few minutes giving a little bit of background on themselves and their organization and their perspective on this work. And then we are going to delve into some questions. Keep this pretty interactive and conversational. And then we are going to kick it back over to you and invite you into this conversation.
Maybe start on this and work our way across to start.
DR. ECKART: Good morning everybody. I am Peter Eckart. I work at the Illinois Public Health Institute where I am the director of Health and Information Technologies there. For the purposes of the conversation today, I am also representing a Robert Wood Johnson-funded national program office in Data Across Sectors for Health. Data Across Sectors for Health is a collaboration that we are doing with the Michigan Public Health Institute. I think that when I looked at the agenda for today, I put our work into two elements of the objectives. The first one is to put forth a multi-sector measurement framework to support public and private collaboration. I am going to really focus my remarks right now on how do we catalyze collaborative efforts to continue this work.
I am going to talk a little bit about DASH, but I also want to talk about another national program. This one is housed at AcademyHealth, their community health peer learning program. My close collaborator, Alison Rein from AcademyHealth, who directs CHP, is here. I am going to talk about the work that she and I are doing together.
DASH is funded by the foundation and CHP is funded by the ONC. They are both really created almost exactly at the same time to test the premise of can we integrate health care data and data representing the social determinants for the purpose of better understanding, what is happening in communities, empowering community leaders, and having an impact on community health improvement.
We have ten grantees in DASH, including Public Health in Seattle and King County. Good to see you, Amy. And then CHP has 15 grantees of their own. What we have all together is a focus on local collaborations, taking a multi-sector approach to share data.
Most importantly for us is that we have an equal focus on developing a learning collaboration so that we are understanding and documenting what works, what does not work, putting that together and then disseminating it to the field.
Alison and I met at Concordia last year and we immediately saw the benefits of aligning our work. We went from asking what we could do together to asking what couldn’t we do together. That commitment to working each other as our friend Soma Stout talks about unprecedented collaboration. That has really defined everything that we have done.
We started out over a year ago, contributing to each other’s call for proposal. We are part of each other’s selection process. We were a part of developing the technical assistance process. We run one virtual platform for our grantees. We do all of our national presentations together. We do all of our webinars and our conference calls all together.
It is our sense that because the work that we are doing is so aligned that it makes sense not to create new silos, but to try to integrate our work. We think that benefits certainly our cohort of grantees who are better by having access to more information, more potential partners, and more examples. But it also contributes to what we are charged to do, which is to understand what is happening in the field and trying to have an influence on the national conversation.
We are very clear that neither DASH nor CHP are inventing the wheel around community-based multi-sector data sharing. And certainly the folks in this room have a lot of experience.
But what we are saying is that this work really benefits from us being able to talk to each other to build some relationships and to build really connections to and from each other.
To that end, DASH and CHP are forming a network of networks to try to leverage and extend the learning collaboration. We call it All In: Data for Community Health to reflect the idea that we are absolutely committed to this idea of catalyzing collaboration and also trying to throw open the doors to relationships with folks who are specifically working on the kind of work that we are doing and committed to sharing their lessons.
We are joined by national partners including the Colorado Health Foundation. We have a really robust relationship with the BUILD Health Challenge. It is not our purpose to create a super structure for this work, but we do know that there is up to 20 different national program offices that are working on this and literally hundreds of unaffiliated local collaborations. We are really trying to understand what is happening, where it is happening and then to work with partners like many of you in this room to build our relationships to extend our influence. Thanks.
DR. HENDEY: Good morning everyone. I am Leah Hendy from the Urban Institute. I am a senior researcher in the Metropolitan Housing and Community Centers. I am coming to you from a little bit of a different perspective. I am excited to make all these new connections more into the health world. I am the deputy director of the National Neighborhood Indicators Partnership that was mentioned earlier.
NNIP was formed about two decades ago. It is a collaborative effort between the Urban Institute and about 30 organizations, which we call local data intermediaries. It is a peer learning network. I just want to briefly describe a little bit about NNIP for those of you who may not be familiar.
These are really intermediaries in the true sense of the word. They are translators. They assemble, transform, and maintain data and then really help communities use and understand it. The second part to disseminate information and apply the data to achieve impact is really what this is about. It is about building the civic capacity in communities to use and understand information and to improve the content of their work.
What makes NNIP a specific model for local data intermediaries has really been our long-term focus on neighborhoods from the beginning and on distressed communities. These are the three principles that we ask all of our partners to uphold to. It is around building neighborhood information systems that are updated across domains. We do not want our partners just focusing on housing or crime, but focusing on all the different areas because people experience their neighborhoods in a holistic manner and that solutions can be really found across sectors.
Again, as I mentioned, the real importance about facilitating the direct practical use of data to build capacity of both institutions and residents who are working to make change in their communities. This really was about democratizing information. In the mid ’90s, information was a lot harder to obtain than it was today. Mapping was a lot more difficult. But this is really taking information, making it available to community members and groups that are working in these distress places so that they have equal access to information that they can participate in the solutions and the changes going on in their neighborhoods.
DR. ROY: Hello everyone. My name is Brita Roy. I am at the Yale University School of Medicine where I wear multiple hats. I am the director of Population Health. I am a general internist. I also do research trying to investigate positive health assets or positive social structural and psychosocial factors that are health promoting. In addition, I have the good fortune of being on the Metrics Team for the 100 Million Healthier Lives Initiative, which is what I hope to focus on today.
The Metrics Team consists of a small, but mighty group based all over the country. Rohit Ramaswamy is at UNC Chapel Hill. Carley Riley is at Cincinnati Children’s. Matt Stiefel is at Kaiser Permanente. And then Soma Stout is the executive lead for 100 Million so she is based up in Boston.
100 Million Healthier Lives is in a really unprecedented, as we said, collaboration across sectors, across communities, across organizations with the goal of achieving 100 million healthier lives by 2020 in an equitable and sustainable manner.
Our charge was to figure out how to measure this. When we thought about what a healthier life is, we adapted the definition of health from the World Health Organization. We include physical and mental health as well as social and spiritual well-being.
We thought it was very important to include these domains of quality of life or well-being in addition to an overall measure of physical health, a summary measure, which we use life expectancy to measure health and well-being.
Well-being is broad, but we really thought and the communities we are working with really thought it was important to find a way to measure this at the ground level. We worked to create a very parsimonious measure of at this point adult well-being. It is a seven-item measure that does measure across these four domains, including self-perceived physical health, self-perceived mental health, adequacy of social and emotional support, a sense of financial well-being. Both of those are lumped into social well-being. And then finally, a sense of meaning and purpose in life, which is how we measure spiritual well-being.
This assessment is really available. Communities have started using it. It takes most people less than a minute to complete. That is essentially how we are measuring well-being.
We are also looking at ways to measure life expectancy also at the sub-county level. And then we are testing a method to actually pair the two together to measure well-being adjusted life expectancy, which we have named as WALE.
But as you can imagine, each community is actually working in a very different way to create healthier lives. Communities will have different drivers, different determinants of these different domains of well-being that they are working on and each leading indicator might range from housing homeless veterans through creating healthier and higher quality lunches at schools.
We have been helping communities think about what those more proximal drivers are and those more proximal outcomes are for their specific initiatives and then also tracking them to the different domains of well-being by which they are influencing healthier lives.
When we think about drivers and leading indicators, we also recognize that communities are working at very different levels. Some communities are working at the individual level on specific health outcomes, which we are helping them to measure, but then also at the community level and society level. That is a lot of what this group is focused here on today.
The community level is perhaps at a city, perhaps at a neighborhood level, but you can think about community-level measures in physical health, mental health, social well-being, as well as spiritual well-being. And then finally, societal level is a higher level, typically policy-level measures at the state, maybe county or national level.
It is a very high-level summary of the initiative and our metrics framework. I am excited to be part of the discussion today.
MR. ROULIER: I think we may get a quick preview at some point later this afternoon of some of what that might look like in reality.
DR. BARNETT: My name is Kevin Barnett. I am based on the Public Health Institute in Oakland, California. I am not a data maven. I love data, but I am a little intimidated to be in the midst of all of this data focus here, but excited at the same time.
My work for much of the last 22 years has been doing research in field work on how non-profit hospitals fulfill their charitable obligations. Now my training is in public health and city planning. My orientation to this work is fundamentally social justice and thinking broadly how we engage our communities meaningful and how we build an ethic of shared ownership.
The evolution of data to what allotted the assessments that had been done historically. We are all assessing our communities to death, hospitals, public health agencies, committee action agencies, FQHCs, United Ways. We are all asking them what creates health and then at least historically use county-level data. We wonder why the data are the same three and five years later that they were the last time we looked at it.
The good news is thanks to the work and leadership that you folks have provided, the great work of folks like the Community Commons and others that are really getting this geo-coded in a way that is meaningful for our communities. There is great potential to really move this agenda forward.
I am going to be brief, but I look forward to our conversation. My orientation is really how do we leverage the resources of hospitals and health systems. Increasingly in the last five years, I have really worked in this intersection between health care, the broader health sector and the community development sector. How do we leverage the roughly $200 billion that is available for areas like affordable housing, grocery stores, child care, FQHCs to really built the infrastructure for improving health? And how do we measure that process? I look forward to our conversation.
MR. ROULIER: I should have mentioned that we were not getting deep background on folks that are speaking, but their bios if you want to learn a little bit more about each of them are in the packet.
Just a couple of questions for each of you and then again we will play out this for a little bit. And then I am going to pass it over to this group. Brita, maybe just starting with you. With 100 Million – I do not know if you mentioned that you spend a lot of time, not only helping with the Metrics Team, but also helping communities to identify and actually use metrics to drive improvement. You have been doing that with some of the scaled communities funded by RWJ. I wonder if you might share what have been some of the biggest lessons and experiences working with these collaboratives.
DR. ROY: The first group that we had working with under 100 Million is the scaled communities, which are funded by RWJF. There are 24 different communities around the country, each with different capacities, and each with different experiences of trying to create healthier lives among their community members. Each of them has a different need and therefore different initiative that they are working on.
What I spend a lot of time doing now is actually working with each of these communities, thinking about what is their intervention, what are they hoping to change, and what are both short-term and long-term process and outcome measures that would be useful to them first. Just thinking if they could measure anything in the world, what would they like to have?
An example is a community that is working on improving the healthy availability at their local schools. While they certainly want to measure just the number of healthy options, they also thought it was important to know if children are buying them and then if they are eating them and maybe if some of this effect diffuses to their home environment. Are they actually changing the habits of what is purchased and eaten at home? These are all worthy measures. But then the question is, do they have access to these measures? Where will they get this information?
Then the next conversation is always about whom else in your community cares about this issue. Who else is maybe collecting data on this issue? And often it leads to maybe some unlikely partnerships. The community members might partner with the schools and those cafeterias to actually get data on what is being purchased. They might look and see what gets thrown in the trash. They might actually start looking at obesity rates in the community or perhaps do some local surveying or send questionnaires home to see if there are differences made at home.
I think those are the two big things, both short-term and long-term outcomes and then collaborating with others that care about the same issue.
MR. ROULIER: In terms of what you are learning or the capacity to do that – to do that in terms of their experiences what you have observed the readiness to do this.
DR. ROY: There are often ranges based on their prior experience and their resources to be able to gather data. Some community organizations have robust systems or robust partnerships with others that are able to gather data. If they are unable to gather data, it often feels a bit more limiting until they can establish some of these other partnerships.
MR. ROULIER: Thanks Brita. Maybe just pass that over to Peter for a moment. I know through DASH, Peter, you have established this readiness assessment tool and kind of discerning what readiness looks like. Maybe you could share what are the factors you all have been focusing on and what does that look like in practice.
DR. ECKART: One of the things I was happy to see as Gib described the framework is that it was not intended to supplant other frameworks. With DASH, we established our own framework to understand the kinds of things that contribute to successful community-based data sharing.
We broke that into two different pieces. The readiness to collaborate and the readiness to share data. This showed up both in an original environmental scan that we did for local collaborations and then a later scan that AcademyHealth did around the national program offices.
I am not embarrassed to say this, but sometimes the lessons from these kinds of things are just so obvious that I hesitate to share them, but of course we keep making the same mistakes over and over again. We do an assessment and we vote on which problems we are going to address and then we say let’s build the data system and then we build the data system and we do not include the intent of residents or other participants in the community. Then we are surprised that people do not use it or cannot use it.
Obviously, one of the things that we have identified is that strong relationships matter very much, continuing relationships past experience.
I would also say that in terms of these multi-sector relationships that we are all moving ourselves towards, there are some tricks to the trade of that kind of communication. Leah and I were just talking as we came up this morning about the labeling for the ten domains and the framework. Some of the language is not necessarily familiar with reflecting her understanding of the built environment. There is no way to create a set of domains that will be perfectly clear to everybody.
Part of this is we need to do a crosswalk between the language that we use and the language that you use. We have to fully understand our own values and the value proposition that we are going to use to share data.
Within both DASH and AcademyHealth and some of our other folks involved in All In, our projects right now are about midpoint in the development of projects. Folks are starting to move from the focus on the collaboration to the sharing of data. One of the things you say is we finally got all that collaboration stuff and now we can get to the technical piece that we really need to do. In working with the DASH grantees, we have seen that if you are not circling back around to all of your partners in the community, even if you did a fairly good job of community engagement to start, you are going to lose them if you are not walking together along this path.
Again, it seems so obvious, but we really do not pay attention to those really important things.
MR. ROULIER: Thank you. I see lots of heads nodding with you. Leah, it was alluded a couple of times. You all through the National Neighborhood Indicator Project have been at this for a while. When we had a brief conversation, you were talking about building a culture of data use and how that might dovetail with some of the work that you have been doing around the open data movement at the local neighborhood levels.
DR. HENDEY: I think this touches on some of the things Brita was talking about earlier. This is about building a culture of data use. It is not just about putting some indicators on a website, but really building curiosity to use data and understanding how that can help improve work. I know a lot of grantees are out there doing data collection activities or using data in the work because they have to because their foundation requires it and because the federal government requires it.
All of these things, but trying to move people this is a compliance activity to these data collection activities and thinking about indicators can really help improve their own performance, increase their capacity, deliver services to the communities they care about. Trying to change the frame, which people are thinking about.
And perhaps it requires reducing this high stakes implication of some of the data requirements we have for folks that they are doing it for this purpose and they are going to be graded on whether or not they are using these indicators, have they produced the number, but how do we build that culture of learning into these organizations.
For NNIP, for example, one of our partners is at the University of Pittsburgh and they run the Western Pennsylvania Regional Data Center, which is an open data portal. One of the things they have done to build a culture of data use in their community is to convene data user groups. These are cross sector convenings. They have people from academia. They have the data providers from local government. They have other interested data users, nonprofits, advocates, community members that come together and talk about an issue.
They have done this on a couple of things. One was environmental justice. But to really lift up what are the data needs in our community to discuss where do we have data that we think answers those needs, where perhaps we need to push further in developing new sources. But you really get a sense of how the data might be used.
I know often, from a data provider perspective, people do not understand how their data might be used in community. They are collecting it for administrative purposes to do the program work to do all these things. There is a reporting system. That is how the data comes in.
There might be adjustments that would make this data more useful for our community. Having those interactions, building those relationships that Peter is talking about are really part of building the culture of data.
I think the open data movement probably less familiar with some of you who are on the health protected confidential data side, but there are still opportunities to create aggregated indicators that could be released. There are public data sets that are out there that governments are more and more making accessible. But this is again an issue of it is not about counting the number of data sets we have published for a city or a county. It is about where can we prioritize opening data that is most important and moving again from a transparency exercise to creating value for the community.
I think important for us and what we have our partners help do and help cheerlead around is are we again democratizing information and not repeating the same issues that if people already have good access to data and technology that we are not recreating that power structure that there are other people who have the opportunity to use and access this data in important ways.
MR. ROULIER: Thank you. I would love to bring Kevin back into the conversation too. You have alluded that you have been at this charitable giving community benefit work for a long time and it is evolving quickly. I am wondering what you might add as what you are observing as that world is changing and multi-sector collaboratives around CHNAs, community health needs assessments, and community health improvement plans, et cetera.
DR. BARNETT: I do think there is immense potential at this particular point in time, from a starting point. Of course, we have to recognize that hospitals are where they are by an accident of history or location decisions. We can go back 30 years and people like Rundle and Shofare(phonetic) showed us, and basically made the argument that location means more than articles of incorporation. In essence, if you are located in an area that is proximal to affluent people, you are likely to serve affluent people.
We have these situations in our communities across the country particularly in our urban areas where you have some hospitals that are located very close to more affluent populations and other hospitals carry the disproportionate burden for serving lower income people who are in socially and economically disadvantaged communities, which just happened to be primarily racially and ethnically diverse populations.
In that context, with the advent of data mapping, we did a study 2014 working with the Community Commons to develop a tool called the Vulnerable Populations Footprint tool that enables people whether they are community members, other organizations across the country to quickly load up any city or county anywhere in the country. You can see where the hospitals are, where the FQHCs are, where all the political jurisdictions are, where the concentrations of poverty at the census track level, high school non-completion, as well as the census tracks where you have populations that are 100 percent above the top quintile.
These kinds of conversations and our ability to begin to increasingly overlay data, many hospitals and health systems in communities that are really interested in advancing practices will overlay what we used to call ambulatory care sensitive conditions or prevention quality indicators and naturally in these lower income census tracks, the utilization is twice as high. We have the means to use and share these data in a way that can really help and begin to drive conversations and move the agenda.
One of my struggles is as important as a lot of this data is and will be going forward, there are also – we need to begin to look at how are we measuring organizational change. The people that are doing this work and the reference was made to epidemiological capacity earlier at least a few years ago, about a quarter of our health departments have someone with epidemiological expertise. Most of those folks are focusing on communicable diseases. We do not have the capacity in government public health to do comprehensive planning, assessment, and facilitation and evaluation. We need that to support this work going forward.
Similarly, for the most part, while the leadership of hospitals is recognizing that the train is leaving the station that we are moving towards a system where they are beginning. They are going to begin to be held financially accountable for keeping people healthy and out of their institutions. Mind you, we are in different places across the country. If you are in a state that has not done a Medicaid expansion – if you are in rural areas, you are more likely going to stay for longer a period of time and fee for service.
But there is recognition that we have to move in that direction and in part out of basic competitive urge, we have systems that are competing to move us quickly as they can along that path. They are now discovering that there are all of these factors outside of anything they do in care coordination that are driving poor health. We are at a moment where if we engage with understanding that the power that we have of being able to put these data up in public settings or diverse stakeholders to look at together, we can begin to drive the dialogue in the direction that we have.
Part of the issue is how are we measuring changes in the way these organizations do business. By that I mean what kinds of committees are being established, senior leadership accountabilities, new competencies, integration across departments, formal partnerships, policy advocacy. There is basically a roadmap that we need to begin to validate in a much more clear way.
It just suggests that there are other metrics that are precursors to the kinds of metrics that we are talking about that are needed to validate and facilitate the kind of acceleration and the direction that we want to move.
MR. ROULIER: Thanks Kevin. A couple of folks have wanted to put out a couple of questions to you all. I am going to ask those and then again I am going to pass it back to you to bring some small group conversations to this.
One question. You can jump around. Not everybody has to answer. Particular trends or bright spots that you find particularly compelling that we should be paying attention and looking at potentially building on. Some specific examples.
DR. BARNETT: Again, I am most excited right now about the intersection between health improvement and community development. This is an opportunity to really drive home to hospital and health care leadership, A, that this is not just about them solving poverty, but, B, that there are some other major stakeholders out there that can help drive this process. Getting them involved and thinking about not only – it is not just that it is a good thing that we can take a little money out of our investment portfolio and invest in the predevelopment phase of a housing project, but that we are actually thinking about how this more broadly helps us contribute to an ecological approach to health improvement, a community revitalization, and at least in the near term, begin to help solve some of the problems that we have for the variety of people with chronic diseases. Opioid epidemic was mentioned. Isolated seniors, homelessness.
There has to be a business case dimension to this. It is not just something that we do because it is nice, but it is actually tied to the core business strategy of our organizations.
We are beginning to see some acceleration. We are still really at the 1 percent level of this, but I think there is great potential to move this forward in the coming months.
DR. ROY: I will quickly mention two bright spots. One is also in the health care sector. Hennepin County actually has created a great collaboration and has linked data from the health system to other county-level data, linking patients in their Medicaid system that have been incarcerated, their education data, whether they qualify for certain types of food or housing subsidies. Actually linking that data across these sectors has been really helpful in allowing the health system and the county to be able to truly work together to address some of the determinants of health and directly measure the outcomes and costs of that.
From a community level, I think community is now with the NNIP. New Haven actually happens to be one of them. We are actually starting to measure well-being at the community level. One benefit to that is that all sectors can actually see how they play a role in improving the well-being of the community whereas not all sectors necessarily immediately see a role in improving health or health outcomes. I think that transition to either well-being or what is proposed here as multi-domain approach is really useful to catalyze in some of these collaborations that are much needed.
DR. HENDEY: I think this plays right off of Brita’s points, perhaps why we are all on the same panel, around integrated data systems just to define that term. For our work, we have approached this. There is mainly from the human services orientation versus the health side, but systems that integrate across government agencies at the individual level so they might include Medicaid and TANF and SNAP and education records and birth certificates and all of that into one system.
We have been working. We supported the Annie E. Casey Foundation for a couple of years now on connecting our partners to agencies or universities that have these types of systems. One example I would give is our Cleveland partner at Case Western Reserve has been doing this for Cuyahoga County collecting more than two decades of information at this point.
They took a look at lead poisoning for kindergarteners. In Cleveland, 40 percent of kindergarteners show up with a positive lead test. This is pretty high rates. But what are the interventions and where can we think cross sector about the solutions?
They found that the time spent living in distressed properties because they were able to pull all this administrative data together and create a month-by-month address history for children from birth to age 5. They were able to see that distressed properties whether it is tax foreclosures or other delinquencies were associated with 15 percent lower scores for kindergarten readiness. And looking at whether Head Start is impacting that difference or not. These families also had higher rates of child abuse and were more likely to be moving more frequently. Where are the interventions? Obviously, we need to have an intervention at the house to remediate some of the lead paint issues, but also there could be other points for other service delivery to occur for these families.
I think that is one of the bright spots I point to is really thinking about both the people data, but then how can we connect it to the place data and how can we increase people’s ability to collect place information and to keep that. Those address histories are actually incredibly valuable and important. Maybe it just should not be overwritten.
DR. ECKART: This is top of mind for me because I was just in Dallas visiting one of our grantees in Parkland Hospital and their data shop, PCCI, are doing this really interesting project addressing diabetes and hypertension through a collaborative project around food insecurity. In that place, they are integrating systems from the food bank, the North Texas Food Bank, which serves a multi-county area, but then also putting software into hundreds of local food pantries, creating a de facto case management system for human services. This project is really a proof of concept of integrating food data with health care information.
The thing that I think is really interesting and it is especially stark on this evolution of the Framework Version 1, Version 2, Version 3. Food shows up finally in Version 3. When you think about the very basics of human development, food has to be pretty central to that. I am very excited to see it represented in the framework.
The other thing that was really interesting in talking with the community partners about this integrated data is that they say because food is so basic, they really see this as a gateway opportunity. They might capture folks who have needs or assets in a number of different domains. They may first be entering the system for food.
MR. ROULIER: If I am not mistaken, that was some of the overwhelming feedback around food, food insecurity. That has to be really primary and you all adopt it.
You all set a nice table here in both talking about some of the challenges that we have and I think alluding to really where the future is heading and some desired attributes in the future and capacity. I want to turn it to the small groups for a little bit. We will bring it back to large groups. About 15 minutes. There is a slide I think you might want to put up there, which is building on the conversation. What is your view of the current reality where the field is? A couple of different comments that have been made there. What do you start to see the desired future for these multi-sector collaboratives using data to drive change? Again, you can build off some of the comments here or add your own. I am going to give you about 15 minutes.
If you like the larger flip charts, that is great. We are not trying to capture any great detail. I will leave that up to folks if they would like.
(Pause for Group Discussion)
MR. ROULIER: Before this last little segment for lunch, we would love to again just surface what seemed to be most significant that came out of your conversations. There are a number of things we could talk about, but let’s just take the next 15 minutes or so to just hear. Again, if you wouldn’t mind using the microphone and share your name as we are doing it.
DR. REAMER: Andrew Reamer, research professor at George Washington University. I focus on federal statistical policy. There are four ecosystem building efforts going on that I think this effort can connect to. I want to mention them. One is that Congress created recently a commission on evidence-based policymaking. It has a charge to come up next fall to look at the desirability and the feasibility of creating a federal data clearinghouse that connects federal administrative and survey data all across the federal government and then brings in state data as well from federally funded programs like TANF and food stamps.
At the same time, Congress has funded the Census Bureau for $10 million to start building a prototype of a federal data clearinghouse. There are efforts to create linkages among survey and administrative records that can feed this kind of framework that we are discussing today.
A vehicle for getting access to those data is the Federal Statistical Research Data Center network. How many people are familiar with that network? Not very many people. The Census Bureau set up and now it is a multi-agency led effort. There are research data centers around the country. I think there are 20 some, mainly at universities, where researchers can get access to confidential micro data that are held by the federal government. NCHS has been a partner with the Census Bureau for many years on this.
There is an opportunity particularly if there is a federal data clearinghouse for researchers to go into these. Research data centers get cleared to use confidential micro data and create small area indicators.
In the non-federal sector, the Sloan Foundation, the Gates Foundation, the Arnold Foundation are collaborating to create a series of data enclaves around the country in which state governments, local governments, nonprofit organizations can pool their data for research purposes. The Sloan Foundation is trying to model what is something in Britain called the Administrative Data Research Network. The idea is to create sector-specific nodes around the US. Clearly, there will be one for health care.
And then the last thing is the private sector – there are all these private sector data firms like Facebook that have tons of data that are useful for indicators. They are not official statistics, but you can do lots of interesting things with them.
The chief economists from Facebook, from Google, from Microsoft – three companies are meeting in two weeks in Silicon Valley to discuss ways of making new data available for economic research. Imagine what might be possible in creating health indicators from Facebook. These are all things that are happening in real time. They are all very early, but I think it is important context for this operation.
MR. ROULIER: Thanks Andrew. I think at that last point, you could generate about three hours’ worth of conversation. I do want to just think about that. I think we will come up with a little bit in the context of the kind of data technology segment in our work. Are these some of our opportunities? What are some of the tradeoffs of some of these private sector opportunities? Thank you for adding those assets. Let’s get succinct here if we could.
DR. FULCHER: My name is Chris Fulcher. I am at the University of Missouri and I am part of the Community Commons teams. We talked at this table. We were using a car analogy and a car parts analogy. We are all dealing with data like car parts when we were going around the table. You make a great carburetor. Those pistons are fantastic. That chassis is really working for you. But people do not really care about the car parts. They care about the vehicle.
When we are dealing with the vehicle, the big concern I have is not about democratization of data because we have been talking about that for over 15 years. It is really getting into the next level. What are the intermediaries? The organizations are those companies that are putting together the data as manifestations of how you visualize or report on that.
That came up because there is a book I read recently. It is called Weapons of Math Destruction. It is how big data increases inequality and threatens democracy. It is a fascinating book because I deal in the data world so much. But it is a little disconcerting because we focus on efficiencies and scaling and we lose sight of what we are really trying to address.
This idea of really being transparent with the algorithms that go into the predictive models is really something we should be demanding. Let’s get past the discussion of democratizing data to really demanding transparency around the algorithms that undergird so much of what we are about.
We are all the car parts people here. What is the next step in terms of how data can actually be consumed by communities as those manifestations of visualizations, reporting tools? What goes into the reporting systems? What is left out? The idea around people or data driven decision making is really quite scary. It is really we focus on people-driven decision making informed by data where and when appropriate. I have to really think about that.
MR. ROULIER: I think last meeting, Chris and colleagues, you gave an analogy or a metaphor about trying to find pumpernickel in the supermarket. I always can count on you for that. Thank you.
What else came out that was significant that you would like to lift up into the larger group that should be on the radar?
DR. PERLA: Rocco Perla with Health Leads. We are kindred spirits with you guys over there. We actually started to talk about what is the purpose. Is it to drive the car with the parts and shouldn’t the measurement be in purpose of that end? And Linda was really pushing us on. What is the purpose here?
There was also a theme around what NCVHS really can stand for with regards to providing that frame and setting that orientation around what the intent of this data. You think about learning collaboration or relationships. Probably not the stuff you think about when you hear about national committees on statistics. I think it is a powerful framing and we were all talking to different elements of that idea.
MS. GOSS: I am Alix Goss. I am an NCVHS member. I just wanted to highlight a couple of the things that seemed to really resonate across the comments you all have been making. We thought a catalog of the collaboration opportunities that are going on in this space might be really helpful for a lot of folks. We need to create this curiosity about the data and the view of what the opportunity is to pivot people to build the capacity so that their communities are driving where they are going over maybe some of the federal and state programs motivating on more of a compliance aspect.
We talked a lot about the tools and the innovation that can help streamline the landscape for the resources to bring it back to the human conversation and the culture of how we want to use that data and that we really need to evolve our thinking and our landscape by supporting people with training and tools. But that really devolved into the limited resources and the burden or the challenge that communities have across the domains, not just health, but if you look at EPA, everybody is sort of heads down, trying to deal with their portion of the big puzzle and we need to figure out how to make it easier for them to get to the quality of the data over the quantity of the data. I tried to summarize all that for the table. If there is anything else you want to add, let me know.
DR. LANDEN: Similar to the previous table, we had our concept here. Our thinking about the current situation is mostly what is out there now in terms of multi-sector collaboration is one off. Communities have to go it alone is one of the ways that was expressed here.
With multi-sector collaboration, it is difficult. If you are starting down the path, you do not know who is out there. You do not know what is out there in terms of data or systems. You are uncertain about how or if it can be integrated. You do not know what success looks like. You do not the value that you will have after you integrate whatever is out there. How do you measure collaboration? How do you do that?
The role of privacy was discussed both philosophically and then most practically, there is state and federal laws and regulations that vary across the country. There are organizational rules and restrictions around data access and use. There is the fundamental issue of data ownership. What is the role of opt-in and opt-out and consent of those individuals, those organizations who the data is about or who own the data?
Switching to the future state, we would like to see knowledge of what data is out there. How do we find out what data is out there and who is out there? The ability to access data and integrate them. And then again, similar to the last table, resource availability, staff training, tools and tool kits. Bottom line, community should not have to go it alone.
MR. ROULIER: Thank you. Maybe if others could piggyback off this last point of what you are seeing. What is the desired future? What does success start to look like knowing that it really looks a little different for every community and region? What are some of the thoughts that came out of your table?
PARTICIPANT: Kind of going back to what Andrew said is the ability to share the data. And then going back to the ownership is that success means you own it, but you let other people know you have it and that you are willing to share it. We are talking about health care data. But what you are looking at, there is a whole lot more data that does not have any of those HIPAA, any of those restrictions on it. But people still like to hold it close. Success means that people start understanding I have data that can help you. Let’s figure out how we work together and we do not keep it so close to ourselves and we build more tools that are easy to use.
MR. ROULIER: Easy to use tools. You really honed in on what kind of desired future state around the shared use of data.
DR. STOUT: Soma Stout, 100 Million Healthier Lives. Our table talked a lot about how in a way you have to see this as a developmental journey as these various sectors come together to create health. At some level, a big part of what we can think about is how we make that path easier. What people are ready to measure at different stages of that journey might be very different. The DoD is thinking about the fitness of their installments or the health of their installments and helping to create the simple tools and measures. How do you know that your environmental healthy? Measure your air quality, your water quality. But just breaking it down, making it simple and making it easy.
Similarly, in 100 Million, we have been talking a lot about the pathway for health systems or for businesses to do this. That maybe begins where it meets people where they are at. Maybe you are just ready to do that with high-cost members, but then begins to think in a very different way about how you can leverage all the levers you have to create health and well-being in your community.
When we think about what data we want to have available and how, I think an ability to be in learning mode and actually allow a lot of flexibility to meet organizations, to meet sectors at their developmental space and to have as sectors come together to discover what those measures are going to be that actually work for both seems to be a really important piece of the puzzle.
DR. LAURENT: Amy Laurent here. One of the things that we talked about for success is also around the sustainability because a lot of times if you are working in the community, there are these one off projects that once that funding goes away, that project does not get sustained so you lose the ability to have that data. Or if you using some sort of technological solution, often times what we experience is the cost of that continuously goes up and sometimes that can make that also unsustainable so trying to find solutions for that is really important.
DR. STOTO: We spoke a little bit about the community health needs assessment that Kevin brought up and really the gap between the promise that he lays out and the actual practice. Some of the reasons for that gap are what incentives did the hospitals have to actually take this seriously beyond checking off the box for the community benefit requirements. The sense that is everybody is doing it on their own, trying to figure it out without any real guidance about how to do it.
The question about how hospitals in a certain area can collaborate working on common measures and the common geography, yet at the same time have their own activities that they are doing within that and some of the models that, for instance, Chicago and Cooke County are trying to figure out how to do that in a way that maybe a model for others to look at.
MR. ROULIER: Thank you. This afternoon I want to be really intentional about there are some assets and opportunities – that we capture those.
DR. DELANY-BRUMSEY: Ayesha Delaney-Brumsey, Vera Institute of Justice. Jumping off a couple of points including the comment behind me, I think one of the things we talked about at this table is really figuring out how to leverage spaces outside of health care to make this a sustainable and useful effort. Because coming from the justice space, although I am a psychologist by training, we are really interested in thinking about how the justice system can be used as a lever to actually improve health and really reduce the negative impact of the justice system on individuals.
For example, in New York City, there was a real interest in thinking about stop-question-and-frisk, and how that is impacting communities of color disproportionately. There was no data from the police department. It required a fair amount of advocacy on the part of justice professionals, as well as community activists, to get that data. And then once they had it, it would have been great to connect that with health data so you get to really see how that might be impacting utilization, trauma, et cetera. That is still a ways away, although I am seeing a lot more interest for creating those kinds of shared data systems.
I think it is important to think about outside of the typical health care partners to really know how to merge that data with of course a nod to really thinking about who has it and what the privacy protections need to be.
MR. ROULIER: It is an interesting paradox. They are different sides of the coin, which is how do we share the data across sectors and how to make sure that we have these partnerships that have the capacity to start to practice in some new ways around how they are making decisions and the values and the context of this work, which I think is a good reminder. It may be the obvious.
DR. MAYS: I just want to pick up on what we are talking about here in terms of these different groups because I think Ayesha has a really good idea here. That is when we start talking about population-level health and what it is that we want to surveil in terms of well-being. Thinking about how people in communities are experiencing some of the things that we are seeing in Charlotte and places like that.
There are county and state and federal officials who keep asking, don’t you have any data that can help me see in my community if it is a hotbed and things are going to pop off. Can you, for example, help me understand how people are experiencing their life so that rather than having the incidents that we are seeing that we have some prediction and then can intervene?
I think this notion of for the health and well-being, connecting with very specific areas, which I think right now it is like policing. It is like jails. It is like a lot of the things that we are seeing, but we are not thinking about putting that into our health care data.
We would love to have linkages, but some of that data has such privacy, security, and confidentiality that we will not be able to link. I think it is also a time to ask. Can we put some of those questions in our big data sets like in HIS and some of the others?
MR. ROULIER: Thanks. A quick comment here in the corner.
DR. FRASER: I think just one analog to that is some work going on in health to predict opioid overdose geographies. It is a similar approach to using health data to predict where the next overdose might be based on sociodemographic characteristics, but also ER missions and some of those other early warning signs. I think there is an analog there for a broader approach that you are making. I think it is really smart. Thanks for that.
MR. ROULIER: And Bob and maybe some others might pick up on some of these points as we are thinking about the predictive, the analytics, and the ways that communities might be using this data to drive action.
DR. WANG: I just wanted to take the opportunity to echo the comments about policing because I noticed in the framework that public safety – and I realize it was only using examples, but it talked about perceptions of safety or crime. It did not talk about those things that are particularly important to some communities like policing activity, lethal and non-lethal force, which is really critical and is difficult for people to talk about. But if we do not put it out there in the health domain, we will not make progress on those larger social issues.
MR. ROULIER: One question that had come up earlier on is that the role of the feds, which we are going to talk about over lunch. We do not want to explore this in depth, but wanted to throw this back to the panel for a moment and ask them. If there is one thing that we could really be having the feds think about and their role in this part of the ecosystem, what are —
DR. HENDEY: For us, many of our partners are working with local administrative data, but all of our partners rely on the federal data sources that are released. I think progress we can make towards timeliness and small area is always appreciated. I think we have some good examples for where the federal government has stepped in with private sector to negotiate data and release it more broadly. Like in HUD, they work with the USPS, the postal service, to get data on vacancy and make that available at the census track level for free to nonprofits and government. That is really doing a service more broadly. I think we can think of more examples where we could make progress on some of those data sets.
But I think also to be an advocate for the federal government, they need more capacity and they need more funding and skills around the data just as well and building internal data use capacity. We have some good examples. The Department of Labor has a data analytics unit in their chief evaluation office. Getting the data providers together with the policy people so we are not just doing data in a silo, but really working on building that into the infrastructure of programs.
Just to iterate my last point, making sure places in the conversation, I know many of the federal agencies have restrictions on what information they get from grantees, but often county is the lowest geography they have available to them and can we make differences to push that down to addresses of census tracks or cities or neighborhoods in the data that is being pushed up to the federal government.
MR. ROULIER: Great conversation this morning. I think as we hear each other talk, some of it is not new territory, but it is worth reminding ourselves of some of the limiting factors. But we also I think just hearing ourselves talk also, there is some extraordinary assets that many of you in this room are leading or connected to. Particularly, as we head towards the end of our time together, how might we leverage some of these assets and pull those together?
I want to thank the panel for kicking us off with great conversations. Thanks for really generously sharing. Lunch, as you may know, is on your own. There are a number of lunch places around here. There is actually a sheet for some restaurants that are nearby. We have given you an hour to grab lunch. I think there is a number of places close enough. We are going to get started as close as possible to 1 o’clock.
I think there are three or four of you who are going to be on the panel to kick us off. If you could be here a few minutes early that would be great. Have a good lunch. See you back at 1 o’clock.
A F T E R N O O N S E S S I O N
MR. ROULIER: Welcome back. And thanks for making it back. Did you have a good lunch? Yes? It sounds like some good conversation for sure.
So, again, as we’re starting up this afternoon, we’re going to come back to a really important part of this ecosystem and focus of our work, in many ways, our federal leaders and participants and a lot of this change work.
So we’re going to have a chance to hear from several of you who agreed to be part of an informal panel as well as we’re going to invite other voices that have federal perspectives to kind of share, what’s been your experience?
And those of you who have been here this morning, what is it you’re hearing around your experiences really working with local collaboratives and smaller estimates? So each of you I think has a handful of minutes just to kind of populate the room with some ideas. And we’re just going to leave it pretty wide open.
Again, after that, we’re going to kind of move to, what does this kind of new growing data technology portend for the work that we’re trying to do given some of the challenges we identified this morning?
And we’re going to land back on getting somewhat concrete around recommendations that we want to offer up through NCVHS to the Secretary of HHS, and what are some collaborative opportunities that we think would be worth pursuing given the assets that are in this room?
So that’s our game plan before the rest of the day. Make sense? So I want you to be thinking about as we move towards possibilities of what things that you really want to promote. But let’s just kind of be in an open learning mode in the meantime.
And I guess Alice, we’ve got Alice Thompson, who will maybe kick us off. And, again, bios are in your packets if you want to learn a little bit more, but I think you’ll learn through some of the conversation here.
MS. THOMPSON: Yes. Great. Hi. Alice Thompson, from the Centers for Medicaid and Medicare and the Innovation Center. I’m really happy to be here and listen to all of the conversations. This is an area that CMS has been working in, in small pockets for a while, but over the past couple of years, there has been a concerted effort to really think through how we can focus on population health. And today I’m going to talk about some of the ways that we have been trying to do that and get at that.
There are other opportunities I can talk about. I’m focusing today on measures, but Accountable Health Communities resonates around a lot of the discussion today as well, and I’m happy, in discussion, to talk more about that.
So as always, at CMS, we have our Triple Aim and everything sort of goes back to that. It’s better care, smarter spending, and healthier people. At the end of the day, we occupy sort of an odd place in that we are a health insurance company basically. I work in the Innovation Center of a health insurance company, so that with it, has a lot of benefits as well as challenges in terms of getting at population health. But in the end, this is sort of the overarching goal.
So in our population health work – and I should mention that this is going on in a number of places within CMS – I am centered in the Prevention and Population Health Models Group within the Innovation Center. Some of the models we have put out, again, Accountable Health communities, Million Hearts, and the work that I work on is focused on how we can leverage measurement as a way to drive population health, and that has a number of implications.
So the sort of mission for that work is really to develop and test novel population health measures and then how to incorporate those into CMS’s value and quality reporting programs as well as innovative models coming out of the Innovation Center.
So just to sort of give you a sense of how we’ve been thinking this, CMS measurement and reporting has really been focused a lot in the past, especially the quality measures, around process. Right? So did you screen someone? What’s going on in that clinical sector? And checking those boxes in terms of process of care.
So in our group, what we’ve been trying to think through is, for CMS, how do we move the needle on those measures, and move them in two ways? So the first is moving from sort of the process measure to the outcome, so a lot of the things that sort of line up with the framework being presented here today. What do we really care about in terms of outcomes for people? What does “healthier” mean?
And then on the other side, it’s moving from that individual level to a population level or community level and making sure that the measures really represent the communities that we care about and sometimes that are the most disadvantaged, and really shine a light on those disparities. So that’s sort of where we are with moving the work.
So in terms of our goals with this, we are really focused on, again, moving to these outcome measures. CMS has a long history of collecting a lot of information that isn’t necessarily useful to people outside of CMS or necessarily useful for anyone but an actuary, and we really want to be able to see our work sort of start to align with work of other agencies within HHS and more broadly.
Through this work, we’ve been doing a lot of collaboration with CDC, and I’ll talk a little bit more of that in a minute. And, again, just to think about how we can move from the clinical sector, because a lot of the – we – obviously you guys have known this for a long time – but at CMS, are coming to grips with the fact that a lot of the health conditions we most care about can’t sort of be moved just within the clinical sector and within the – yes.
So how do you incentivize multimodal evidence-based intervention? So in terms of smoking, for example, there are things that can be done by physicians in that setting, but those are enhanced by community resources and being able to make those linkages, and so really trying to push CMS outside of the doctor’s office into the community where people live.
So I’m going to go through a couple of examples of the work that we’ve been doing.
Last year, we proposed a smoking prevalence measure for the Hospital Inpatient Quality Reporting program. This was a measure that comes from the BRFSS at CDC, and the idea here was that we could hold hospitals accountable for a county-based smoking prevalence measure. This was reviewed at the Measure Application Partnership, which is CMS’s sort of process for reviewing measures.
There was a lot of discussion around why we thought that you could hold a hospital accountable for a county smoking prevalence measure, what the role of the hospital would be in terms of that measure. But, again, this sort of underpins the idea that our clinical providers, hospitals, and health systems are actually huge actors in the community in terms of moving population health, and if we start holding them accountable for those things, we can actually incentivize them to reach out beyond their walls and make the connections needed to change health.
So, again, we submitted this measure to our internal process, and there was a robust discussion in December of 2015. In the end, the Measures Application Partnership recommended further development, and so we sort of took that and have been thinking through how to expand the use of a measure like this, sort of across CMS, and thinking about a smoking measurement strategy for programs like the Medicare Shared Savings Program and the Merit-Based Incentive Payment Program, which is coming out of the MACRA legislation.
So if we think about, “How do we hold a hospital accountable?” maybe the answer is actually we hold all providers in the community responsible for this measure because it does require collaboration across not only clinical sector, but also the other sectors. So we are in the process of developing a suite of measures related to tobacco along these lines.
Just really quickly, I want to point out that the tobacco measure that we selected was really a test case, that we believe that this could be applied to other types of measures and other outcomes, and that smoking represented a wonderful opportunity to be able to look at a obviously very clearly identified problem.
Everyone sort of agrees that smoking continues to be a public health challenge. In some areas, there have been declines, but there are still pockets of very high rates of smoking that are impacting health, but also that quit lines and things like that are opportunities to reach out into the community. And so we do see this as a test case, and this is sort of where we began, but look forward to further discussion on opportunities for that.
The second area is a contract done under the Measure and Instrument Development and Support Contract through CSSQ, the shop at CMS that develops measures, specifically focused on population health measurements. So they did an environmental scan and determined gaps in measurement areas. And then in the second phase, we actually picked out priority areas to develop measures and/or a blueprint and roadmap for further development.
So just to sort of talk about it, the health behavior outcomes that we selected, again, smoking as well as obesity. So those are the two areas that we identified as potential gap areas, so we need measures for those that CMS could use, but then there was also readily available data that we could sort of more quickly move on.
The other thing that we came out of it with was a blueprint and a roadmap around multi-sector collaboration, and I think this applies a lot to the discussion we’ve been having today, is thinking through how CMS can best sort of situate themselves to be able to facilitate multi-sector collaboration and the work that needs to sort of occur again across multiple sectors that have to do with health, lots of the things we’re talking about today.
And so, lastly, we’ve been exploring a number of different strategies, including working with state innovation models that come out of CMS that do a lot of data collection and analysis at the state, county, and sub-county level.
We’ve been collaborating with federal partners, as I mentioned, with CDC looking at ways to leverage existing data sources. A lot of the sources we looked at had challenges sub-state, and then sub-county was even harder. So just figuring out ways that we can leverage those resources, then coordinating activities across CMS and HHS, and we have some convenings that are coming up related to that so we can get a better sense of what’s going on and how CMS can fit into that work that’s already happening.
So thank you.
MR. ROULIER: Thinking about questions you might want to ask.
So, Wayne Giles?
DR. GILES: Sure. It’s a pleasure to be here and to talk with you all today. You should have one-pager on your table that talks a lot about one of the projects that I’m going to talk about.
So, as was mentioned, I’m at CDC. I run the Division of Population Health. Within that division is the Behavioral Risk Factor Surveillance System, or BRFS, that Alice mentioned in her presentation.
And I also have a group that’s been doing for a number of years a lot of work around small area estimation, and they have largely been focused – their methodology allows them to get down to the census block level. The nice thing about that is they can then aggregate up into multiple different configurations. So they’ve been doing work around looking at data from the BRFS and other data sources, looking at congressional district, looking at data by county.
And we’ve got this new project that they’re doing now, which is the 500 Cities Project, and if you want – I can do it myself, can’t I? This is the collaboration between CDC, the Robert Wood Johnson Foundation, and the CDC Foundation. It was launched about a year ago. And really what we are interested in doing is, can we provide estimates for the 500 largest cities in the United States for a number of indicators? And within those cities, as was mentioned earlier today, can we get below the city level down to the census tract level so that leadership in those cities can look at different census tracts or different neighborhoods in those cities, and compare and contrast, and that’s really the intent of the work that we want to see and want to see happen.
This is the 500 cities. So a couple of states only have one city in them. We made sure there was at least one city in every state. California has the most cities. It’s got 120 cities in California. About a third of the U.S. population is covered in one of these cities as part of the project. And I should say you can actually go to the website CDC.gov/500cities, and you can get a complete list by state of all of the cities.
Really what we want to do, if you all are familiar with the state maps, very similar to what we do with BRFS, but how do we go to that, to the map that’s on the bottom, where people can really look and see, and policymakers can really target the work that needs to be done and know how to do that?
We’ve got 27 measures that are part of this. We’ve divided them up into unhealthy behaviors, health outcomes, and prevention, and it really very nicely targets the subdomains within the health domains of the framework, so I think that’s some really nice opportunities.
In terms of the unhealthy behaviors, we decided to focus on the five unhealthy behaviors, for those of you who remember the Alameda Heart Study that looked at major risk factors for heart disease, but a number of chronic conditions. So it includes binge drinking, current smoking, no leisure-time physical activity, which is I think about linkages with Department of Transportation, obesity, linkages with Department of Transportation, sleeping less than 7 hours. But those were the indicators that were all in Alameda.
And I should say – I don’t know if any of you saw, but America’s Health Rankings actually just came out with a report last week looking at these same behaviors and ranked the 50 states in terms of it, so there are some really nice things that are coming out. I think a lot of people are convening around these five behaviors.
These are the health outcomes. There are 13 health outcomes that we’re looking at, major causes of disability, such as arthritis, but also asthma, high blood pressure, cancer, high blood cholesterol, chronic kidney disease, et cetera. You guys will post these slides, right? So I don’t need to.
DR. GILES: The slides will be posted if you want to see the individual indicators.
And then the prevention measures. This includes things like lack of health insurance, the proportion of people who have seen a doctor for a routine checkup, taking medicine for high blood pressure, cholesterol screening. There are nine prevention measures that are part of this work as well.
So since we’re from Atlanta, I figured I would show a couple of maps of the city of Atlanta. But this is just looking at the prevalence of binge drinking for the city of Atlanta, but you can see the northern parts of the city, higher rates of binge drinking than the southern-more areas.
This is looking at chronic obstructive pulmonary disease and sort of the variation that we see by census tract.
And then this is looking at sleeping less than 7 hours on average per night, again data from the Behavioral Risk Factor Surveillance, but you can see some of the variability that we get.
And then I also wanted to show Los Angeles as well, but this is the data that we see from data from Los Angeles again for sleeping less than 7 hours.
So a couple of things in terms of next steps. We will be in December 2016, probably early December 2016, we will be releasing for these 500 cities a series of map books that will include all of the 27 indicators, and they will be up on the website. With that will be access to the data, so people will be able to download the data, the census tracts, et cetera, et cetera. So that will be in December 2016.
And then in early 2017, we will be creating an interactive website, and there you will be able to look at the maps, you will be able to change the quintiles or quartiles for the maps. People will also be able to enter an address and see the map in terms of census tract. So you will also be able to compare cities as part of the work as well. So that will be in early 2017 as well.
So thanks for the opportunity to present to you all.
MR. ROULIER: Thanks so much, Wayne. Very exciting.
So Jason Broehm, Department of Transportation.
DR. BROEHM: Yes. So I am Jason Broehm. I work in the Office of the Assistant Secretary for Transportation Policy at USDOT.
I want to take just a moment to recognize that for a number of years I worked at CDC, so I have a history in public health. In fact, Ed Hunter, my former boss, is sitting in the back.
And I also wanted to say that while I was working on public policy and congressional relations primarily, Ed did give me an opportunity to go and do a detail at the Federal Highway Administration working in their Human Environment group for about 5 months a couple of years back, and that was really a building block towards some of the things I’m going to be talking about here today.
So one thing that predated that experience, the detail assignment, was that for several years CDC and DOT had been working jointly to develop a transportation and health tool, which we launched last year in October. That brings together, not health data strictly speaking, but 14 datasets from various parts of DOT and the Census Bureau primarily, looking at things like alcohol-impaired fatalities, commute mode shares among transit, walking, bicycling, and driving, Complete Streets policies, land use mix, proximity to major roadways, various things that have sort of a built environment aspect and I think probably a theme that you’re getting at a lot through your discussions today about these factors that play a very important role in public health.
So sort of building on that – and I should say one more thing. The data for that are based on the availability of the data that we’re pulling from and the scale at which they’re collected. We have state for I think all 14 of those. And then for a combination of those, we have them either at the metropolitan statistical area or the urbanized area. So it’s a little bit of a mismatch, but that was the best we could do.
I should also say that some of the feedback that we’ve gotten in presentations from users and people who are interested in the transportation and health tool is that they would really like to get below that level. So that’s something that’s sort of a frustration, we have to tell them these are the data that we’re relying on and the scale at which they’re collected.
But I think there is an interest in getting down and drilling down along the lines of what Wayne was talking about to a neighborhood or above level that’s at least smaller than sort of the regional level.
So in the experience that I had at the Federal Highway Administration, one of the big tasks I had was planning a big meeting that brought together CDC and DOT staff from across both agencies, across sections of various programs, and looking for intersections between those two. So we did two of those in 2014 and 2015, and actually last week we just had a webinar that sort of built on that and continue to share information about programs.
And through sort of running through this, transportation and health data and linking the two up and looking for ways that we can bring them together is a theme that’s been recurring. So I’ll talk about several projects that have come out of those conversations.
One was to get some either revisions to existing questions or some new questions into the National Household Travel Survey. This is run by the Federal Highway Administration and administered really on a fairly irregular basis every 7 or so years I would say.
Right now, we have one in the field and those new questions that we were able to get in. And when I was at CDC, I was working with folks at the Federal Highway Administration, and after I moved over last year, continued to build on that to make sure that those questions got into this round of the survey.
So it’s in the field right now from the end of March of this year to April of next year. And from what I hear from the program director, that they have interim data reports already that show that the response rate on physical activity questions has actually been pretty good, so there hasn’t been a drop-off on those. They’re expecting to release the public datasets late 2017 or early 2018.
Another area that we’ve explored and sort of hit a little bit of a roadblock and set aside was BRFSS. There was some conversation at the 2015 workshop that we held in Atlanta about whether we could get some transportation questions into health surveys generally, and we sort of focused a little bit on BRFSS.
I understand that one of the issues was we need to come up with money to get that question in, which is a challenge. And the other was that I believe it was a three-quarter threshold of states would need to approve those new questions. So those are a couple things that we just ran into and set aside.
One of the projects that we did identify in the first workshop in 2014 that is I think any day now hopefully getting the data linkage happening finally, we had to go through a process of putting in a research proposal to the Research Data Center at the National Center for Health Statistics at CDC to link up some built environment and transportation datasets with individual health records within the RDC.
So we’re looking at intersection density, traffic density, and block density within the RDC, linking those up with individual survey responses, geocoded at the location, and then looking at various geographies around that, around each of those, and just really as a pilot in four large states, look at the associations with things like physical activity, obesity, heart attacks, asthma, as a way to see how we could do this and how it might be broadened to a more national scale at some point in the future. And we’re looking initially at California, Florida, New York, and Texas. So hopefully we’ll get some promising findings from that and maybe be able to build on that in the future.
I’ll just mention a couple of other things where there have been conversations and it’s still maybe sort of evolving, but this may or may not set off any light bulbs for you based on the conversations you’ve been having. And I looked into your framework, and I know transportation has been worked in there over time. And so a couple of areas where there is a lot of work going on and hopefully we can build on this in the future.
One I know that our colleagues at CDC in the Physical Activity and Health Branch have had a lot of interest in, better mapping of walking networks, sidewalks. These data are largely collected at the very local level, so there is no national map of all sidewalks, as much as we would like for that to exist.
But there has been a lot of interest in some research activities at the Federal Highway Administration building on – they’re really promoting connectivity and networks of pedestrian bicycle infrastructure. Rather than looking at individual projects, they’re looking at how they connect up and how they get people to where they want to go and need to go.
So this is something that’s still evolving in terms of mapping and really trying to support local governments and the more regional scale metropolitan planning organizations in collecting these data, and hopefully at some point in the future making them available for some more national analysis.
And then there has been a lot of work going on both with the Federal Highway Administration and the Transportation Research Board on counting pedestrians and bicyclists. Often it’s evolved through more like a trail focus, but it’s also being incorporated more into the street network. I know there are some counters here in D.C., and many other cities are working on that as well.
I’ve thought that one really interesting and helpful thing to do would be to have counts before and after you put in a bike lane or before you improve a sidewalk in a certain area to see sort of what the – you may not be able to immediately make the link to the public health outcomes, but at least you could see the change in activity over time, and that’s something that I think we’re working towards hopefully.
So Federal Highway Administration funded a pilot with 10 metropolitan planning organizations across the country that didn’t have existing counting programs to help them buy the technology that they needed and then helped them deploy it and do everything they needed to get the counts, collect them, and make use of them.
Also, they’ve updated their travel monitoring guide, or TMG, to include a chapter on bicycle and pedestrian counting. Historically, there has been quite a robust amount of effort to focus on counting motor vehicles as they’re going around and the street network to have general estimates for vehicle miles traveled, which is aggregated and looked at nationally and reported. But they’re looking to really incorporate into that existing system pedestrian and bicycle counting.
And then last I would just say that their Travel Monitoring Analysis System, TMAS, is currently being updated so that they can start to accept the counting data that’s being collected across the country into this national database like they keep for motor vehicles. So that’s another emerging, I think somewhat exciting, way that we can get better national data around physical activity through active transportation.
So I’ll leave it at that and look forward to questions.
MR. ROULIER: Okay. So Elizabeth Sobel Blum.
DR. BLUM: Hi. I’m Elizabeth Sobel Blum. I’m with the Federal Reserve Bank of Dallas. Who here is familiar with the Fed System?
(Show of hands.)
DR. BLUM: Oh, that’s a nice surprise. Usually people say, what does a Fed do? And is it Janet Yellen? So Janet Yellen is our head.
I have to give a disclaimer, of course, whenever I talk in public about these are my views and not necessarily those of the Federal Reserve System.
Okay. So I’m from the Dallas Fed. So the Federal Reserve System is the Federal Reserve Board, which is here, the Federal employees there, and Janet Yellen heads the Fed System.
And then there are 12 Reserve Banks. And so I’m from the 11K, which is – actually a little bit of trivia. Do you know if you take out like, for example, a dollar bill and you see what letter of the alphabet it is, that shows you which Federal Reserve Bank the – when the money was produced by the Mint, it went through one of the Federal Reserve districts, who then just put it in the money supply. So that’s how you know what bank it went through. So if you see “11” or if you see “K,” you know it went through the Dallas Fed’s District.
So our district is all of Texas, northern Louisiana, and southern New Mexico, and specifically from the community development function. And the golden thread that ties all of the Community Development Offices throughout the Fed System together is that we promote community and economic development and fair and impartial access to credit. Let me repeat that one more time, we promote community and economic development and fair and impartial access to credit. So basically our responsibility is promoting the social determinants of health, even though that’s not what we traditionally called it.
So the reason that we do what we do is that in 1977, the CRA, Community Reinvestment Act, was passed, and it turns out that’s when it specifically became illegal for banks to redline. So what banks used to do legally was they would look at a map and say, “Okay, we’re going to draw a red line in a neighborhood and say we’re either going to totally exclude this neighborhood or we’re going to accept deposits, but we won’t extend credit.” So people can’t get small business loans, they won’t be able to get mortgages, they won’t get other forms of credit.
And the point in 1977 of CRA and other fair lending laws and regulations that were passed basically said you can’t discriminate on anything besides credit worthiness. You can’t do it based on race, ethnicity, immigration, all that stuff; it’s all based on, is someone creditworthy? So it was supposed to level the playing field.
And so the community affairs, sometimes called community development, function at the Fed was created to provide technical assistance to banks on how not to redline, how to meet their CRA obligations. And included in their CRA obligations are investing in community development, from affordable housing to community revitalization, et cetera.
So what you see now is that over the past 5 or 6 years, the Federal Reserve System has been getting increasingly involved in the conversation about healthy communities.
Sometimes you might see, for example, if you’re in one of the districts like Minneapolis or San Francisco Fed and the Dallas Fed, there are a lot of conferences and events with, for example, the Robert Wood Johnson Foundation, where we’re bringing together the community and economic development industries with the public health and health care sector industry showing – CDC to us means Community Development Corporation, not CDC in Atlanta.
And so we need to get beyond acronyms to have a shared understanding and learning and appreciation of what each other does because we’re all working in the same neighborhoods, but instead of just brushing elbows, we need to actually be linking arms to be working together because obviously we cannot do it alone, and we see and appreciate the inextricable link between health, income, and education.
So in terms of the role that we play, we play the convener role and the disseminator role. We disseminate information. So we invite people to partnerships, and then we attend partnerships and say, “Oh, did you know that such-and-such entity is also actually engaging in this, and they actually have data on XYZ?” So the point is making sure we’re helping make those connections be bridged and play that translator role in terms of, “Oh, in your environment, did you know this is what Transportation is doing?” or housing or education or workforce development.
So that’s the role we play. In terms of dissemination, we do it in terms of publications, we do a lot of qualitative research, some quantitative research, it depends on which department you’re talking to. And we very much work as a federated system. So it’s not like top-down where the board says, you need to do XYZ, and everybody says, okay. It’s about each investor reports to the board, This is what we’re doing based on what the needs are of our respective districts.
So we all recognize that we’re in this together, that we’re trying to address the issue of social contact being broken, where no longer just because you work hard and get a good education, therefore, you’re going to be in the middle class or higher. That’s very much no longer the case.
I generalize this, but what you see is the middle class is shrinking. And so we care about this because we know health is an asset, that we know as the middle class is shrinking, there are severe problems in terms of health, income, education, et cetera, et cetera.
So I’m happy to be a resource to connect you to your local Federal Reserve Bank. If you go to FedCommunities.org, there is actually a little place in the upper right-hand side of the website where you say give what your location is, and it will tell you what your Federal Reserve Bank is, and so you can reach out to the local community affairs or community development staff of the district that you’re in and ask them how to get plugged into the community economic development conversations so that they know what you’re doing and you can help them with what they’re doing, too. So it’s a much more collaborative approach.
MR. ROULIER: Thank you. So I know there’s more than a handful of Federal leaders here that might want to just offer up pretty informally some experience that they have working with some sort of local community collaborative with the small area or smaller area data, sub-county data.
Claire I think was maybe – is Claire Wang in the – Claire would you mind? I think I put you on the spot.
DR. WANG: Sure. I’ve been told to share. Hi. My name is Claire Wang. I am a physician epidemiologist. I’m a researcher coming from academia, but spending the year at HHS in the Office of the Assistant Secretary for Health, where we coordinate a lot of public health activities, from the federal to state to local levels, so having been in conversations with many of you in the audience.
So I’ve been told this morning, well, yesterday, to share a little bit about Public Health 3.0. That is an initiative. It’s a call to action by Acting Assistant Secretary for Health Karen DeSalvo, to strengthen the local public health infrastructure. So early I share that the goal is to empower the local health leaders, public health leaders, to be the chief health strategists for their communities.
And really built on the idea that despite the fact there are the incentives for health care providers to getting the population health and care about the communities’ health, oftentimes it’s only when people became patients, and there really needs to be a little bit more accountability at the local level. And local health departments are really at that juncture of being accountable for the health of the entire population.
At the same time, the infrastructure at the local level has been facing a lot of challenges, funding being one of them. When the economic downturn hit in 2008 and ’09, more than half of the health departments had program cut back.
And facing that, you know, we’re also looking at a lot of the data, including Raj Chetti’s(?) paper, looking at how life expectancy in different communities, not only the top earners and bottom earner households experience very different life expectancy, the gap, but depending on where you live, that gap is either narrower or wider, and that really speaks to the fact that there are some communities around the country who are doing the right things, and being able to build these collaboratives so that the community is health promoting.
So Karen DeSalvo, in working with many thinkers and leaders in CDC, including John Auerbach and Denise Koo, being one of the leaders in this initiative, and Charlie Homer, from ASPE, started thinking about and have convened a series of conversations. And so I was told that I will share some of our findings and what we heard.
Since late last year, we have had five regional listening sessions around the country talking to different communities. So that includes Allegheny, Pennsylvania, Nashville, Central California, Kansas, and Spokane. And in these regional meetings, about 100 people each time would show up, and from sectors including public health. But also the housing, transportation, and other community development organizations in the area would show up and share about why these communities are taking action to build these umbrella organizations so that they can share resources, braid and blend and measure and sharing data and build collaboratives and have leaderships that go towards that direction.
And the reason why we call these communities 3.0 communities is because in a 1.0 era – I’m going to the history part – early on, there was a lot of emphasis on developments in screening tests, lab sciences, and a lot of infectious focus, towards sort of the 2.0 era, where the local public health functions are more defined, and accreditations started to be important in defining foundational capabilities.
Meanwhile, chronic disease started to become a threat to health and population health, but in this era, a lot of the funding and a lot of the studies are still pretty much disease-specific and categorical.
So that creates a challenge for today where all the communities started to talk about social determinants of health, health generations, more than health care, so that really needs a different approach to engage different sectors and using different sectors of data.
So in these national convenings, we are going to have a final report coming out on October 18th. The national meeting is going to be here in Washington, D.C., so talk to me about registration and so on. And on October 17th, I’m also organizing a roundtable panel event with George Washington University on the making of Public Health 3.0 leaders. So talk to me about that as well.
So on October 18th, we’re going to release the summary report on what we found and what we heard and the recommendations that we’re making organizing five themes and which we consider the five essential ingredients of Public Health 3.0: leadership and workforce development, infrastructure, including accreditation, the third one is about funding, more sustainable and more flexible funding, and – what am I missing? And, of course, data analytics and metrics.
So these recommendations will come out. So earlier in our process, on March 22nd, many of you guys are in the room in this roundtable on data metrics and analytics. The experts got together and talked about, what are the data opportunities and what are the visions for Public Health 3.0 era data system?
So the vision is really threefold. One is the local health departments really want timely and locally relevant data, which is very central to the theme today. And not just health, but measuring multi-sectoral data, but also focusing on health is not just the absence of disease, so really getting to measuring the well-beings and quality of life and being able to do the things they want to do.
And, finally, actionable. So it’s not just about we can measure, so we measure, but really driving how actions can be taken at a local level.
So I will share five general findings that we just put out in the summary report of that roundtable convening in March, and now it’s posted on the Healthy People website, and that’s where the Public Health 3.0 initiative is sitting.
What we heard is, yes, there are a lot of data gaps. Some communities are still struggling to get prevalence and incidence of chronic diseases, for instance, and connections to mental health services. All these are data gaps that are being talked about.
On the other hand, there is emerging consensus that everyone has a lot of data sitting somewhere in their electronic file cabinets. But the challenge at the local level is really to make sense of them. So that could mean connecting data or having visualization tools, having metrics that matter to the community, and the analytics to say, “What does it mean what you are seeing these data and where the problem area will come up next?”
So in metrics, many people talked about the need for a national dashboard of measures because right now there are just so many measures, and there is a lot of burden in measuring.
The second one we’ve been hearing is HIPAA sometimes is a real or perceived challenge or barrier to data sharing. So there is a huge need to connect health and human services data, but oftentimes within the agency, it could be the federal, state, or local level, there is a perceived – and I think many people alluded to this – that people tend to hoard data and feel like they have to protect them from sharing even with colleagues across the hallway.
And so we’ve heard local CIOs in health departments say that when they want to reach out and get education data, they need technical assistance, not technical in terms of the analytical part, but legal. So they need legal guidance on how to draft that data user agreement so they can use it over and over instead of a one-up handshake.
The third one is the importance of local voices and how local public health agencies can be a great convener because what they could do is let’s say the federal or state level will have these core measures or data resources available, but then they can add local voices and context. And so, for instance, in Detroit, they are layering, the health department is layering, their lead measures data in a map and overlay with their – I guess it’s development data.
So Detroit has one of the nation’s largest scale of demolitions, and going forward, the health department is really worried about the lead levels that come with demolition projects. So on two sides, they layer the data on where the demolitions are happening, and ongoing monitoring of lead levels to make sure there are no hotspots that they need to attend to. So it’s an ongoing process.
In Colorado, the state has their hospital data available in the city of Denver. They layer that with the dispensaries for marijuana, and it is very local questions, very locally relevant questions, but drag data from a central data resource, and you are to layer their own data and their questions.
The fourth one is many people call for enabling a data infrastructure so you can capture cross-sectoral return on investment. And I think in the discussion of social determinants of health, when you’re intervening on, say, financial security or housing security, oftentimes the return on investment is accrued in a different sector, so the data connection has to happen in a strategic way so you can capture that return on investment. So if you are spending the money, but someone is benefiting, you should be able to claim, to acknowledge, that return on investment in a different sector.
The final one, the fifth one, is a higher level one, is this cultural change has to continue. When the Obama administration started the Open Data Initiative, that really changes the default of data from the default being closed to default being open. So I think that cultural shift needs to continue. And that also speaks to sort of the HIPAA point earlier.
So, finally, I think we’re making a lot of progress. And generally there’s a huge opportunity and a lot of innovations are happening. We learn so much about what CDC is doing, DASH, Community Commons, and so on. So I really just wanted to scale it up and perhaps recommend an interagency working group on thinking about social determinants of health and how different data elements can cross sectors.
MR. ROULIER: Great. Thanks, Claire. Somebody said that you might have something to share. So it was terrific. Really interesting.
So any others that might like to offer up?
DR. HOMER: Charlie Homer, from ASPE. I just wanted to briefly mention, I think people are aware, but our office at HHS serves as the liaison to the activities driven from the White House focused on communities of concentrated poverty.
And there have been a number of initiatives across the duration of the Obama administration. Many of you know the Promise Neighborhoods Initiative, but that also includes Strong Cities, Strong Communities, they include the Promise Zones, they include the Choice Neighborhoods, they burned – it was a criminal justice initiative out of Department of Justice, and my colleague Jeannie Chaffin might also talk briefly about the Rural IMPACT Project, which is another initiative addressing rural poverty.
So these are initiatives that are intended to be quite different from the way the Federal Government has historically operated. They are not put out with the typical funding opportunity announcement that says you will do these 43 things, they’re really to identify specific communities. Again, applicants needed to be from communities of concentrated poverty, they needed to demonstrate that they had a coalition or a group, a variety of partners, that could identify what the priorities were, and each community came forward with what their specific priorities and areas were related to how they would improve the conditions of that community.
That had its strengths and its weaknesses, of course, as we’ve sort of had in the concept of the measurement that we had before. But it is very community-driven. It has encouraged, enforced, in fact, a significant change in the way the federal workforce works with communities, and we’ve sent a fair amount of detail people out from Washington and from the regional offices to specific communities.
We have had more recently an extensive sort of training program for federal staff across a variety of departments. And this is all departments. HHS plays a relatively small role in these initiatives overall. Substantial training.
Related specifically to the data early on in the project, ASPE contracted with Urban Institute, which described some of the data sources that are available, including the Urban Initiative that was mentioned, as well as Dennis Culhane’s project. I think that’s Data for Actionable Intelligence Social Data for – what is it?
PARTICIPANT: Actionable Intelligence for Social Policy.
DR. HOMER: Actionable Intelligence for Social Policy. Thank you. University of Pennsylvania. And, actually, just last week the White House convened a meeting to try to foster partnerships between universities, which have the expertise and the data, with these local community partners, and we’ve made available at least the data inventory that Urban provided as part of that work.
So I’m happy to provide any more information.
And, Jeannie, I don’t know if you want to mention any more about the exciting Rural IMPACT Project.
One last thing, we are running a learning collaborative amongst the Promise Zones focused on early childhood. As a pediatrician, I couldn’t forget to mention that.
MR. ROULIER: Thank you.
DR. CHAFFIN: Thank you, Charlie.
Jeannie Chaffin, from the Administration for Children and Families. And I really appreciate that Charlie mentioned the place-based work that’s gone on in the administration because what that sort of demonstrates is that, yes, the data is important, but there is also the practice behind solving these complex community problems.
And, you know, oftentimes our particular funding streams have difficulty with creating that practice because they’re focused on a very specific problem, and I think a lot of what we’re talking about here is, how do you cross many different sectors and disciplines? And so I think this administration has tried to push that a little bit.
But I would challenge everybody in this room to continue to think about that work. And when we have a new administration, where does that work fit and where does that work live, not just the data, but the practice of it?
Charlie mentioned several place-based. Another one that we have been pursuing in the administration is really on rural places. Much of the place-based work has been in urban, but the White House Rural Council has a ten demonstration site project to really demonstrate two-generation services, so high-quality services for parents and children at the same time, in ten rural places around the country.
And one of the problems that all ten of those sites are grappling with is integrated data. So they are realigning resources in their community to serve parents and children better, but they have this great challenge with, how does the data work to do that?
So a lot is going on at ACF around this space. I was sharing with colleagues that invited me to lunch today, as great as your data is sort of in the health field as far as being interoperable, we’re way behind you in the human service field, in sharing data across food stamps and child care and LIHEAP and all the various human service, but it’s so important to the problems that we’re trying to solve together.
So I know many of you are going to continue the conversations this week, and I challenge you not to forget us as you’re thinking about some of the human service, how to bring that into what you’re doing.
MR. ROULIER: Thanks, Jeannie. I still remember, Jeannie, we were talking about this the other day, being part of the Commons team, and when Jeannie kind of spearheaded an effort with MAA, when she was with Missouri Action Agency, and created this tool that allowed folks to kind of quickly populate a community health needs assessment to kind of do the reporting requirement for all the community action agencies, and you were in the room, and folks were so excited because 3 months or so, or at least a couple weeks’ worth of work of collecting data, all of a sudden, they had it right in front of them, and they were like, “Our work is done!” and you were up in the room and said, “Your work now has just begun because now you can engage folks, you can bring in data.” And that was a precursor to a lot of the CHNA stuff that now Kaiser funds on the Commons.
So I’m curious, there are a lot of places we could go. It’s fascinating what we’re hearing, and we’ve just kind of skimmed the surface. But in the next 15 minutes, I wonder if we could be somewhat strategic in naming, what is it that you’re hearing that excites you as a possibility that we should really be paying attention to? And what’s really the implications for doing this work, as federal leaders, to really bring small area estimates to scale and to spread this work? So what are the implications for doing this better, faster, and what is it that you’re hearing that really excites you, possibilities?
DR. BEATLEY: I would like to start.
MR. ROULIER: Please.
DR. BEATLEY: I’ve heard a lot about research and researchers using this work, but the other thing that’s really exciting is the idea of letting the public get more involved. I mean, I’m kind of sad that you guys didn’t hear about the HUD project, their Healthy Communities Index, but their Healthy Community Assessment Tool that was piloted in four places and specifically targeted at the neighborhood level, so that neighborhoods – and to engage the people, the community, and the community activists, and looking at it so that cities could understand, how do we work at the neighborhood level?
And I think it’s great that you guys are now looking at very small areas, because county is great, but it crosses over too many places, and it doesn’t identify. And when people can identify their own area, they get much more involved.
MR. ROULIER: Yes. So engaging them and really engaging the capacities and the wisdom of communities. I know Susan over lunch was like, “I was curious if the term ‘citizen science’ was going to come up at any point.” There is this whole movement really around how we better engage our residents and local leaders.
Can you get a microphone here for our friend from EPA?
PARTICIPANT: So I would like to follow up on that comment and thank you for this opportunity again for being able to talk about EPA perspectives, especially in engaging communities involved in the scientific evidence to lead to ideal public health outcome and health and well-being of the communities.
So we do share the many common experience and ideas with all of you already spoken, but if I may circle back to the workshop number one last November, when we talked about EPA perspectives, we talked about ownership, and we talked about ownership of data per se.
But actually when we work with the communities, especially those environmental justice communities who feel that they are underserved or even neglected, when we approach them and say we’re going to monitor your school’s air quality or community’s air quality or drinking water quality, rather than using actually small things like not using monitoring or surveillance, but rather then involve – actually ask community leaders and members to list their needs and what they like to see in our scientific research makes a huge difference. And I’m talking about, of course, community engagement and actively participating community members.
We recently had a workshop presenting EPA tools for land use and city planning, and we invited city and community planners and leaders to ask them what they like to see in our framework and tools.
So if I may say, that’s the vision of success on this framework, that I know this framework is much, much more complicated than just land use, but in same avenue that we can actually make the communities, that they can use these tools and domains and indicators to better their health and well-being, you know, depends using their health impact assessment or environmental assessment issues. So ownership is extremely important.
And then, again, whether it’s drinking water qualities in Flint, Michigan, or violence issues in Charlotte, North Carolina, or Baltimore, Maryland, we all need to be very, very sensitive and quite susceptible to their needs, the community needs, and unique needs they might have.
And case in point is we recently asked the tribal communities to – we are thinking about, based on our scientific data and evidence, not to consume too much fish because of high level of mercury and PCBs, but we didn’t know, but the community leaders actually told us that will have devastating impacts on their community because the fishing industry and fishing consumption is such deeply rooted in their culture, society, and even spiritual well-being.
So that kind of example must be included in this framework. So thank you so much.
MR. ROULIER: Yes. Thank you. Yes.
Peter, our colleague from DASH, was saying let’s – you know, we have to state some of the obvious. I mean, some of the obvious is, how do we actually engage communities, the wisdom and assets, build tools that work for them, and in partnership, this co-creation notion?
What else jumps out?
DR. NORRIS: One thing that I think would be interesting to consider is we talk frequently about community, and there’s a definition of sub-county areas within this, but what is the optimal community level? Is it geographic? Is it population type? Is there something that we could define in terms of the data level as well as the actual putting programs into effect that, well, is ideal in terms of this whole – all of this that we’re talking about? And it could conceivably vary place to place. In some places, the population size might be 8,000; in others, it could be potentially a million.
But I’m wondering if anyone has looked at what the best construct is for community – and I think probably people have and I’m just not familiar with it – but the best construct for community in terms of both data level as well as actionable items that come from the data in terms of community well-being, community health, et cetera.
MR. ROULIER: Yes. That’s a really good question. It’s a really big question, I think, too. I wonder if anybody wants to quickly respond to that. And I want to also head back to Kevin and a couple others here.
DR. SCHMELTZ: Hi. Michael Schmeltz, from EPA. And just to address that question, you know, we’ve been working on a project at the EPA to understand vulnerability mapping to human health outcomes due to climate change, and one of the things we’ve realized, you know, going through a workshop and interviewing a lot of experts, is that to identify a community, you really want to identify a geographic scale and a timeframe and a – I forget the other one.
But you want to really identify something that you can follow over time, because if you’re changing the geographic area over time, you’re going to miss certain populations that are changing over time or people that move or populations that disappear from certain areas. The other one I thought about was scale in terms of geographic area as well.
So just looking at those three things to determine, what is a community? And it may be determined by that, just kind of geographic factors.
MR. ROULIER: Yes. I don’t know this will happen for sure, but I suspect as we move into a little bit of the data and technology section, to really think about, what does this portend in terms of looking at scale and different kind of geographies in a rooted place?
Did you have one? And then I’ll head back to Kevin.
DR. STOTO: I want to respond to Gib. I think it’s a really interesting question, but a really hard one.
MR. ROULIER: Yes.
DR. STOTO: And part of the reason that it’s hard is that the answer would be different depending on the use and the topic and so on.
So maybe I would rephrase it, is rather than the optimal one, what are some definitions that would work reasonably well for a lot of different purposes and situations?
MR. ROULIER: Yes. I remember when Michael McGinnis was sitting here earlier on and the kind of launch of some of the U.S.-based Healthier Communities, there was this premise of a broad definition of health, which we continue to kind of build around, but also a broad definition of community, of neighborhood, of school district, and it kind of depended on what – those kind of matryoshka dolls – the community where you were taking action. Right? What context made sense, whether it was policy – and so I still think about, what’s the broad definition of community?
Kevin, did you have something?
DR. BARNETT: I did. There is a lot to be excited about here in the use of data at this level, and yet we have communities that are persistently concentrated poverty and challenges. And we’re talking about solving problems within those communities.
I guess my question to us is, what’s our theory of change here? What is it that gets us from solving a problem and identifying a bright spot and addressing some of these – I mean, a lot of the work that we see that we actually report as successes are gentrification of neighborhoods because the majority of people get displaced. So there is a deeper issue that we have to grapple with.
So we talked earlier about the lack of epidemiological expertise in our public health agencies. How can we use data in a way that drives home and where we’ve identified we’re making progress towards actually beginning to fundamentally change the equation here?
Because as I see it now, we’re still sort of – we’re getting excited about helping people identify stuff, but we’re still not dealing with the structural racism, with the larger issues that are preventing us, including the fact that our foundations are not aligning their funds in a way that are actually producing the dose to actually produce an impact, our institutions are still operating in a proprietary way.
I just wonder if we can begin to identify and lay out, what’s the progression here that we want? And how do we get to some of the bigger ideas, the bigger outcomes, that we want?
MR. ROULIER: Yes. Thank you for putting that out there.
Soma, maybe one or two more, and let’s come back after break.
DR. STOUT: So I’m going to build on what Kevin just said because I keep wondering as I hear this, first of all, on the one hand, being incredibly excited about all the work that’s going on, and wondering when it translates to real results for real people.
And I wonder if there isn’t the possibility of actually having – putting forward a real result for real people that catalyzes us to collaborate in a different way, whether that focuses on, could we think about neighborhoods of concentrated poverty where we know we have to figure out how to solve the cross-sector challenges around social drivers? Everybody would agree that matters. That pushes us to do it.
What if we could, looking at the registry of all of the neighborhoods of concentrated poverty, think that we would want to create improvements and real health and well-being in a certain proportion of them and think about how we bring our assets and resources together to learn at scale and at some speed about what it would take to get there?
So I just encourage us to consider whether it’s that, whether it’s let’s take on the epidemic of chronic disease, whatever it is, that would be something that drives us, I think that’s the way we actually begin to have a real reason to put our pieces together beyond the knowledge-sharing level that we’re doing today.
MR. ROULIER: Yes. Beautiful. I would like to suggest – this is right after lunch – that we land kind of a little bit around this present pregnant possibility, this question that I think in some ways the Kevin and Soma tag team offered up to us.
So maybe let’s take 10 minutes, 12 minutes. We’re going to be thinking about where technology might fit in this. We know that technology is not the driver, but I think that you all are asking the right question, too. What is it that we might do together? What would be worth doing that would be building on some extraordinary bright spots, but actually get to some broader system-sustainable structural change that I think many of us have built our vocation around? Right?
So thank you, all the federal agencies, for being here.
Back in 10 minutes?
MR. ROULIER: So one of the beauties and also one of the frustrations on a day like this is there is so much that we could delve into, there is so much, you know, I think a follow-up conversation I think a lot of us want to have. So I appreciate your kind of holding onto that tension and ambiguity.
But I wanted to kind of put out one kind of more lens out there as we’re thinking about possibilities towards the end of the day, and that is, of course, just as our whole movement towards multi-sector work that’s aimed at health and well-being and greater equity has been evolving pretty rapidly in many ways, and spreading over the last few years, so has the technology for how we think about data and use of technology, and how we deliver that.
So we have Bob Phillips, who many of you know, who is on the committee here, and then Roxanne Medina Fulcher, and Mike Reich, who are also part of the Commons, and they can kind of introduce themselves as they go.
But, Bob, I think you’re going to kind of give us a little bit of a landscape around kind of data and technology possibilities. So thank you.
DR. PHILLIPS: So I am going to start talking a little bit about advances in data technology, but I want to at least give you some perspective of where I’m coming from.
So the roadmap that you have that was in the back of your packet, we are through the first loop, and there are a couple more loops that come after this. And part of the work that comes after this is now if we can come to some consensus on the measures that we should be collecting at the sub-county level, what do you do with them? Where do they live? How do you store them? And who manages that? So that’s part of what I want to talk about.
So the Institute of Medicine, now called the National Academy of Medicine, had a couple reports that came out, the first one in 2011 called “The Role of Measurement in Action and Accountability,” had a couple actually specific recommendations. One was that the Secretary of Health and Human Services transformed the National Center for Health Statistics to provide leadership to a renewed Population Health Information System to enhance coordination, new capacities, and better integration of the determinants of health.
So they actually suggested there be an NCHS. I might actually argue against that. We’ve heard today from Transportation and USDA and EPA that this may need to live above that or at least if there is going to be a platform, need to draw from different agencies to get a good picture of health. But they suggested NCHS.
And then they talked about the need to do this at all geographic levels and coming up with a core standardized set of indicators, which is part of the work that we’re completing today, and then looking at health outcome indicators to pull into that space or to use to help understand which of the other indicators are important, and then coming up with summary measures of population health.
So they continue to walk down that trail of, “What indicators? How used? Against what kinds of outcomes?” and starting to talk a little bit more specifically about the work that I think this group has been doing for the last year.
So a year later, in 2012, they came out with “Primary Care and Public Health,” and this was really about exploring the integration of those. It had a lot in it, but one of the things it talked about – and I’m only going to talk about Recommendation 1, Bullet 3.
So it said that HRSA and CDC, who sponsored this report, should join efforts to undertake an inventory of existing health and health care databases and identify new datasets, and create from these a consolidated platform for sharing and displaying local population health data that could be used by communities.
And they really talked about this platform being about ensuring that communities could use the data in their assessments, their intervention planning, and evaluation. So, again, another message about a common platform.
So I wanted to just point back to these references. They may have some utility for us.
Now I want to talk about a couple nominees, or at least examples to build on. And one of those is out of a little tiny agency within CDC called the Agency for Toxic Substances and Disease Registries, or ATSDR, which actually has something called the Social Vulnerability Index, and it draws on census and American community survey variables with complex modeling to get down to census tracts at least, sometimes census block groups, to help local officials identify communities that may need support in preparing for hazards or disasters, and it has 14 different social factors.
You won’t be surprised that almost all of those 14 are in our list that has been created. And they have things like poverty, lack of vehicle access, crowding, crowded housing, and it pulls them into four different themes. And they actually have a mapping tool. You can find the – if you look up SVI, you can find the county maps, so you can pull up county maps that are already prepared for you, but there is also an interactive map.
So I pulled up Alachua County in Florida where the University of Florida is, and it lets you look at the four different themes. So socioeconomic status, household composition, housing, transportation, and race-ethnicity-language, and each of those themes is a construct of those 14 variables.
And it lets you map – these are maps of ZCTAs, which are ZIP Code Tabulation Area, it’s a representation of ZIP Codes, so it’s sub-county, and lets you look at neighborhoods within them, if you count ZCTAs as neighborhoods. So there are predigested maps for every county.
And then you can also use the interactive map and create your own map. So I just pulled up the regular screenshot so that you’re looking at counties across the country, and then you can dive in and start to play with it. And I’ve actually talked to several other agencies and organizations that use this routinely.
I’ll tell you one of my concerns about it is that there are just 14 measures. There is no weighting, they have not been checked against outcomes, so there is not any sense of which ones are actually more valuable than others, but they’re there. So this may be a nice platform to build on that’s already embedded within CDC. And I only stumbled upon it thanks to Gib.
So thank you, Gib.
HRSA has its own mapping platform that was developed by Health Landscape that’s called the UDS Mapper, which means the Uniform Data Set Mapper. So every Federally Qualified Health Center in the country, managing 24 million patients, has to turn in a Uniform Data Set every year that’s geographically based. And for a long time, it meant that if you wanted to expand your health center or build a new one, you had to either have this data set or you had to create it from your community health needs assessment to try and justify why you needed that expanded opportunity.
When I was at the Robert Graham Center, we had been doing a lot of research over about a decade with health centers helping them better understand how to use and map their data and thought it would be helpful if we got every single health center’s data into this tool so that at the federal level, the state level, the local level, they were looking at the same data sets to understand where resources were needed most.
And so this tool has been up now since 2008. It’s completely open. And you can look across the whole country at Federally Qualified Health Center data looking at geographies of care, and then pull in social determinants data so that you can start to look at the proportion of people under 100 percent or 200 percent of poverty, served and unserved, in any given ZCTA in the country to understand where gaps in access to care exist currently.
So it does have a pretty rich database of multiple layers of social determinants of health care utilization, sites, National Health Service Corps, Indian Health Service, rural health clinics, to try and give you as robust a picture based on the data sets that we have now of what’s happening in every locality.
So you have the ability to look at the UDS data from the health centers, or population data, or combine them. You have the capacity to look at concentrations and determine what those look like.
I pulled up Alachua County again – you wouldn’t know that I was a Florida Gator, would you? – to try and again look at those population data by ZCTA to understand poverty rates within the county. And then, as I said, you can overlay the clinical data on top of that to start to get a sense of where care might be needed.
This is actually where I flipped it. So you’re looking at a percent of unserved in these ZCTAs, and that’s based on the number of people in those ZCTAs who are currently using FQHC resources denominated by the percent of people eligible. So you really get a sense of unserved in an area as well.
So I just offered those not as the platforms by any means, but just to say that two of our agencies within HHS are using different platforms, some of the same data. And then we heard about a whole bunch of other data opportunities that could be fed into a platform like this so that you have the capacity within agencies, within counties, to do some high-level common looks at data.
I think Mike Stoto was talking earlier that there might be some opportunities for us to – there are some common things that we all may need to do or some common places where we may all want to start, but we may want the opportunity to bring other data into an environment like that to look at our specific needs or to bring our own data into that environment to look at our own specific intervention opportunities.
And then the last thing I would say about it is the ability to use geography, not only to display data, but to protect data is an important theme, too. So there are data sets that cannot be shared because they’re identifying and that become worthless when you de-identify them. But if you can put them into a geographic representation, they have value and utility without necessarily threatening the identity or populations of people.
And depending on what geography you’re looking at, data can appear or disappear depending on the threat level to that population. So it gives us another way, these platforms do, of using data without threatening people or communities.
And I’ll stop there.
MR. ROULIER: Thank you, Bob.
DR. PHILLIPS: Sure.
MS. FULCHER: Good afternoon, everyone. My name is Roxanne Medina Fulcher. I am going to have to switch, though, on over to our own presentation. All right. Here we go. And so one more test of – is this the clicker, if I walk around? I tend to do that a little bit. So thanks, Mike.
So as I said, my name is Roxanne Medina Fulcher. I am the Executive Director of IP3, which stands for Institute for People, Place, and Possibility. For those of you who were here this morning, or if all of you were here this morning, I kind of echo Kevin – I see him smiling back there – his comments about being a little intimidated by being in this room.
My background is as an attorney. I practiced poverty law in Detroit for a number of years and then moved into the United Way System, and that’s how I kind of got into this work.
At the United Way System, my role in putting resources back out in the community, I was saying, “Where? How? What data are we using? What are we collecting? How do we make decisions?” And so I come at this work from that perspective.
And so you heard a couple of different times today, you heard Community Commons pop up. So for those of you who don’t know, it is a website, and it’s where we bring data tools and stories together to improve communities. And so I have a number of team members in the audience. Erin is back there, Cara is there, and Chris Fulcher, and then one of our new team members, not IP3, but Seabourne, one of our partners, Mike Reich here. So we’re going to tag team a little bit.
And this is just an example of one of the tools that are out there. And then we’re going to talk a little bit about where technology in general is going to help support this work and then what we might be doing with that.
So what do we on the Commons? So right now we have a number of different toolsets: the demographics, equity, environment, food, economy, health education. Those are the different types of data that we have in there, thousands of different secondary data layers from many of your agencies around the table here.
We started back in 2012, I think is when we launched. And I see Wayne right here, who was actually in our welcome video when we first, first launched. So thanks for that.
In going through, we have access to a lot of different data layers, interactive maps, data, and reports. And just as a little bit of a sample, so the data layers are nationwide for the most part, and they can zoom on in, so into all the way down to the tract or the point level, just depending upon the data set. So that’s one of the different tools you would have access to.
Another tool that was mentioned earlier this morning was the Vulnerable Populations Footprint tool. And this, again, was a collaboration between different partners, I think. We brought together a number of representatives from health systems from Public Health Institute, CDC helped convene, Kaiser Permanente, ASTHO, NACCHO, there was a whole bunch of us around the table to really bring this tool and the community health needs assessment reporting environment to life.
So this was one of the tools that actually identifies the different areas where two indicators meet a certain threshold. So it would be poverty and education attainment. So where those two intersect are the areas in red.
And so on the Commons, you can actually use this to identify those areas, to add different points onto there, and do a little bit more exploration around tools that would help with equity.
The Dynamic Reporting is another tool. And so for those of you who have been on the Commons, this is one of our big areas where people do a lot of the assessments. It started out as community health needs assessment, but we quickly saw the value for it for any kind of assessment, and even beyond assessment.
But what it does, and what you can see here, is bringing together the data in many different forms, not just with mapping. That was the inception of everything. But then how do you get it in tabular form? How do you also see the data in dynamic paragraphs?
And we work with – and these are some of the different graphs just pulled out, so you can see. And it’s really designed to help people understand the data in different formats, so it’s not just in spreadsheets, it’s not just in a map, but how else can you see that and bring it all together?
Our team at IP3 does a lot of curating of this information, too. So it’s not just access to the data, but how do you bring it to life so others can be inspired by it? So we have stories on the Commons. We partner with the journalism school at University of Missouri, we work with guest writers, we try to slew the round, “Who’s doing what in the Commons?” and try to interview them and bring that to life, again mostly to inspire others and to tell the story of what’s actually happening on the ground.
And we also have areas on the Commons that take these tools and put them into context. And so whether it is the same exact tool, there might be a different lens with which people want to look at the information and data, so we do offer and we do have areas on the Commons that puts that into context. So whether it’s around women’s issues, whether it’s around community health improvement, there is opportunity. And we’ve worked with actually many again in the room to put these tools into context.
So over the years, though, we have grown in membership. If you’ve signed onto the Commons, I mean, mostly so that we can keep in contact with you and share information with the people on the Commons, and we’re over 45,000-plus members now, which is phenomenal in growth, but we know that there is so much more that needs to be done. This is not enough. So what is it that we need to do to look at in the future?
Also, not only the Commons has changed, but there has been a change in the field, and we’ve recognized it from our vantage point at the Commons. There has been exponential growth of multi-sector collaboratives. We hope there are more of them, but we do recognize that there are very specific needs and very specific tools that need to be developed for multi-sector collaboratives to succeed, not only in getting access to information, but also to measure whether those collaboratives are having an impact.
There is a proliferation of data access tools and systems. I think so Cara started this work 20-plus years ago. There might have been one or two other people in this field that were doing the same type of work. Now there are so many, which is a great position to be in, but how do we go beyond just the access again? And how do we also – the increased need to effectively use data to drive and measure change.
So as we look forward, we’re looking at, how do we expand our partnership? How do we expand the way that we actually build the technology so that we can dock or connect – or interoperability, those are the types of words that are floating around in our team and with the different partners that we work with.
We’re looking – and the reason this slide is up here is whatever tools and things that are developed, they need to be process-agnostic, they need to be applicable to more than just any one process. And we recognize that within the cycle of community change, that there are so many different points in time when you would be accessing information and data.
So our question or where we are right now is how do we get the right data to the right people at the right time with the right context to make the right decisions? And so to help keep this narrative going, I’m going to turn it on over to Mike, with Seabourne.
DR. REICH: Okay. Thanks, Roxanne.
So my name is Mike Reich, and I’m with Seabourne Consulting. We’re a data strategy and technology firm.
So we’ve been working for probably about the last year with the IP3 folks and with a wider network of partners to think about, how can we build on the success over the last 5 years of the Commons? And as Roxanne said, it’s been really a process of learning and growth, and 5 years later, now, today, or a year ago, there’s a lot more data, there’s a lot more people, there’s a lot more things going on.
And so we embarked, yes, about a year ago, on an evaluation and rethinking some of what we learned, and then really looking into the future. What could the combination of newer technologies, I think a lot of what’s been mentioned today around open data initiatives and new data warehousing and new database technologies, et cetera, give us some new possibilities for what it is we can do? And then there are things like, obviously, social media and mobile phones, tablets, IoT sensors, all this constellation of data, which now presents some really cool things to do.
So I wanted to start with what we’ve learned, and I think Roxanne touched on it a little bit, but there are five things that we can sort of take away from our experience building technology. And I wanted to bring this context because there has been a lot of talk about process and data and governance and organizations, which is usually important, but at the end of the day, this is about people who are on the ground doing something. And what we’ve learned is to them, in and of itself, data is useless. If you just give them a number, it’s really hard to make any sense of it.
So the example I’ve got – well, okay. So the number here, imagine, it’s not showing up, but the number is I think 68. So what does 68 mean? Well, it can mean a lot of different things. The number itself in isolation has no value, but what technology does is adds context, and through that context, data becomes actionable.
So if we go to the next one here, there we go, 68. This is actually the temperature, right? And it’s technology that brings together the context, the other data, the relative perspective on the number 68 which makes it actionable. Because of this, I can decide, do I want to bring a raincoat or do I want to bring an umbrella or what does the – you know, you’ve got all sorts of interesting stuff. I was doing this, I was thinking you’ve got spatial granularity, you’ve got temporal granularity, you’ve got minimums/maximums, you’ve got a whole bunch of really interesting information here which helps you make those decisions. But, again, without that context, without the wrap around it, the number itself, the data has no meaning for people who want to make decisions.
All right. So for those who are math sort of inclined, we’ve got our little conceptual formula. Value is going to be extracted out of this stuff through a function of the technology operating on the data, so those two things together.
All right. So the second thing which I think is also really important and really salient is there really is no right answer here. There is no single platform or technology that’s going to do everything for everyone. And, in fact, I think, you know, raise your hand if you’ve used a platform that was supposed to solve all your problems. Right?
(Show of hands)
DR. REICH: I’m sure there would be a few people who would raise their hand.
And, in fact, when we try to build that, I think we end up getting really distracted and we get focused on trying to do way, way too much. And so we end up with the Excel spreadsheet that nobody can actually use, or the application that’s overly complex, it includes way too much information for everybody. And so you have to be an expert in the application, and you have to go through training, you have to figure out what it is you’re doing with the technology before you can start to understand what the data tells you, which keeps you from actually doing your day job, or your day job becomes using the application, not doing what you’re supposed to be doing.
So I think the third point here is because one application is not going to do everything, we need many applications that are going to cover all these different contexts. And one of the big challenges is, and I think the appeal of a big application from a technology standpoint is, you don’t want to reinvent the wheel again and again. Right?
So I think there has been a very real sense that we need to consolidate and coordinate our effort within a single technology platform, to not duplicate, but over the last 5 and 10 years, from a technology perspective, there have been some really interesting answers to how we can have sort of the best of both worlds.
Like we have something that’s really focused for users, but also allows us to sort of adopt this concept of modularity, so we get a toolbox of LEGOs that we can reuse to build again and again. So we get something that’s very custom but something that’s also very quick to develop.
And the reason that many applications are important is, I think as we think, especially in the context of multi-stakeholder collaboratives, there are multiple stakeholders, there are multiple people who have multiple viewpoints, multiple perspectives, and the way that they want to look at the data and even the data that they want to look at is going to be very different.
So Roxanne showed earlier sort of an abstract model, this abstract model of actions that can be taken, and the arrows in each stage represent data and technology. So they help people understand what it is they’re doing and how they move to the next step, and it’s circular because there really is no end. This is not a linear process. This is a process of evolution of learning and reapplying what you learned to improve over time.
So the fourth thing that we learned really successfully I think was great technology helps organizations operationalize and scale. There are best practices that work, and technology can be an implementation of that, so organizations don’t need to reinvent with Excel or with any other tool what works really well.
And because we’re not locked into a single application, we’re locked into a bunch of applications which are sort of built to fit a specific need, if an organization can share that application to other organizations that have that specific need, they have then the ability to adopt that without having to go through the thinking and the development and all of the energy that goes into creating the software. So it becomes a way that we can scale the work that organizations are doing and the best practices that they’re following much more quickly.
So the four that we’ve really focused on and in thinking about that, that cyclical process, the four things that organizations do that we can help scale are discover. So what’s the data? What are the measures? What are the best practices that others are doing? Technology, we see this with social media every day, is great at putting information in front of people in really interesting ways, information you may not have found in other places.
It can help you collect. I don’t think anybody is using pen and paper forms anymore to collect information, and that’s because technology allows you to do a lot, a lot more quickly. Well, there are some places where they are. I know. I know. But then it gets entered into a spreadsheet and into a database. But ideally, we would not be using pen and paper forms in most cases for data collection.
And the third is obviously analyze. You know, we definitely don’t use pen and paper to crunch numbers anymore, and there’s a good reason for that. The volume of data is increasing exponentially, and so the analysis really only exists these days in the world of technology.
And the last one is monitor. Once you know the questions you want to ask, you need to follow a process of staying up to date on it. What’s the change over time on that process of monitoring and of looking at dashboards as sort of one way of monitoring data? Again, it’s a place where technology can be really supportive and help share best practices and ways of looking at the data and interacting with the data that work really well.
Okay, so the fifth thing that we’ve learned is connect and share. This is a shout-out to Soma, who was talking about this yesterday. So connect and share, but within the context of technology, means some slightly different things. And specifically two principles, which may be familiar to some of you who work in the Federal Government where we want to connect using open APIs. So from a technical standpoint, we want our systems to be able to talk to each other using sort of a common – using voices, right? Using not even a common language, but a common mechanism of communication, and that’s open APIs. And there has been, I think – somebody mentioned earlier the OMB guidance around open data and starting from open as a beginning. APIs are I think also a requirement for agencies to be publishing their data via APIs, and it’s something that we’ve seen I think has really increased the pace of adoption of data within applications over the last 5 years.
And the second is we want to share using open formats. One of the things that we’ve learned as kind of an anti-pattern is trying to get agreement on a data standard, as in a big “S” standard, because, again, there is no right way of presenting this information or storing this information that’s universal.
So instead of focusing energy and time on every – let me put it this way, there are always places where big “S” standards are necessary, but instead of defaulting to a big processor, I’m trying to come up with the single standard focusing instead on open formats that allow us to get the data, it’s not locked in a sort of proprietary format, and so we can get it and then we can translate it into whatever structure that we need, or we can translate the data into the form that is going to be most useful for us in our particular application.
So these five learnings really shaped as we were thinking about what a technology strategy looks like, really shaped what it is we thought about and thought could be done. So we thought it would be fun to sort of walk you all through what it is we’ve been working on and sort of a little bit of a case study. And we’ve spent probably 9 months now working with the 100 Million folks, who you heard from earlier.
Brita was talking, you’ve heard Soma a few times in the back. They’ve been just I think a great resource in helping us understand who are the folks on the ground who ultimately need to be doing things with this data. So all of the performance measures and outcome measures and process measures that we’ve been talking about ultimately have to be ingested and used by some organization and some person.
So we’ve been thinking about, what’s the suite of tools that we can create? But more than that, not just the tools, what’s the sort of model that we want to try and explore of creating technology and open data that could be adopted by anybody to help scale some of these best practices and solve some of these problems that I think were – you know, thinking about like, where do we put the data? How do we actually make this usable to people once we collect the data behind this framework that we’ve been talking about today?
So our case study is going to be Ben, and Ben is starting an initiative to address homelessness in Portland. This is not the only case study, but this gives us a good sort of ground-up view of how somebody is a practitioner in the world who is trying to drive change in their community might use data and technology together throughout that life cycle of organizational development and change.
So the first thing Ben is going to do, he’s probably interested in figuring out, well, homelessness, what is homelessness really? How do you measure that? How do you define homelessness? And metrics is a way of framing an issue. And we saw today you can divide any of these big sort of issues into a lot of taxonomy of subdivisions.
So he’s probably, in our mind, going to stumble on Google, and we’ve got a nice catalog of measures that we have been trying to collect, and the 100 Million folks and Brita have done a fantastic job of pulling together five or six hundred, eight hundred – what are we up to?
DR. ROY: We’re over 1,000.
DR. REICH: We’ve over 1,000. So 1,000 measures together out of literature, out of best practices, folks are actually using these on the ground. So Ben is going to come and find our catalog of measures, and we’re not dogmatic about measures, we’re not opinionated about them, I think there are definitely – we want to try and bring as much of this together as possible so that it’s a resource, a common good, for Ben and for anybody else who is trying to find it.
Over here, Ben is also going to have access to a bunch of related stuff, and all of this is linked together. So the measures and the topic areas are related to the data that informs the measures, the tools that allow you to capture the data or analyze the data or report on the data. It’s going to be related to the organizations that are using the data, using those measures in their work in the community.
So trying to provide as many entry points into all of these different technology resources and data resources that are going to help Ben think about his problem within his community. So he’s going to go to the Tools page, and he’s going to be able to start with three tools that are going to be useful for him.
So the first is going to be an assessment tool to try and understand both things he may need in order to hit the ground running, but also, where are the gaps within his specific community? So he’s going to start with that.
And the first thing, he’s going to click into that tool, and it’s going to ask him about what he is trying to do. So what’s the type of change he is trying to create in his community? And he is going to enter some information about who he thinks the people are.
And then, more interestingly, he is going to enter information about where they are specifically, down to as granular a level as we can get. And so we actually give a tool. And I’ll let you guys figure out how much of this exists. You’ll have to come up and ask me. But a good amount of this exists and is actually in use.
So there’s a tool here that allows you to draw a geographic boundary, and we really encourage people to do that because it gives us a specific local sub-county geography that we can then associate with everything else that they may do into the future, including capturing data on their own.
So he’s going to go through a few more screens that I didn’t show here to pick a focus area and pick some other things so that we know about him. And then he’s going to go and do an assessment of some sort. Right? He’s going to click through some data. He’s going to understand something about his community. And through that, he is going to, and a lot of hard work, he’s going to build a board and a coalition, and a bunch of folks are going to help him in his work on the ground.
Now he’s launched, he’s found funding, he’s ready to go, and he’s at the point of saying, “Okay, now I want to measure, what is my impact? What am I actually doing?”
So because we captured all this great information about Ben behind the scenes, and we’re very transparent about that, but we were able to get all that great information from him about his problem in his community, he’s going to go back to the Tool page and he’s going to select the Metrics That Matter tool, and this is going to allow him to choose some measures that are specific to homelessness that he’s going to use to measure change over time. So he’s going to click through.
And because we know what his initiative is, we know a lot about Ben, we’re able to pre-populate and suggest measures that he might find useful that he can select. So we’ve got sort of his initiative page here where there’s a whole bunch of other stuff, but essentially, he’s able to choose a measure and select either from a pre-set menu that might be framed around the issue of homelessness, so choosing the ten measures that best indicate progress in a community towards reducing homelessness. He can also add his own in there.
He can do his own research. We have a catalog behind the scenes, as you saw earlier. And then he can set some targets. So where does he want to be by what date for each of these? And over time, he’s going to be entering information in, and that information, along with some cool things, like being able to annotate if there was a drop in homelessness, you know, on a quarter-by-quarter or year-by-year basis, or month-by-month, depending on what the granularity of the data is, he is able to annotate why – was there something he did as part of his initiative that drove that change? We’re trying to get at the idea of, what are the best practices? What are the non-quantitative things that might go into driving measurable change within a community?
The other interesting thing is a bunch of these measures are obviously going to be coming from secondary data sets, from government sources, from national data sets, maybe from local or regional sets. Because that’s really a public good, that’s something a lot of people need, and we don’t need to be duplicating census databases all over the place, we have that in a central place that’s accessible via an open API that this application is then able to pull.
So most, I would say 80 percent, of the measures that we’ve worked with so far, actually secondary measures, where the data is coming from these great sources, but he may have some measures which are unique to his community or unique to his initiative, he can enter those values directly in here.
Okay, but one of the questions would be, for, say, one of those measures that he is creating, he actually wants to gather some data from people, and to think about what might be within his community, say, the secondary data set he’s interested in doesn’t capture data frequently enough. So we’ve got this concept of a questionnaire application that would actually allow him to, through a mobile app, go and gather some information from the community directly, maybe from his stakeholders.
So the thing there is it’s not SurveyMonkey. We’re feeding back from the same metrics catalog that we’ve got, which also has detail about instruments. So we’ve got the actual questions. If it was from a survey, what’s the question that was used to get the value for the metric? So we can take those instruments and we can auto-populate essentially a questionnaire for him that he can send out to some members of his community to start collecting some data, and that data, as it’s collected, is going to be auto-populated back into his metrics platform.
So we’re able to do again a lot of automation behind the scenes. We’re not going to solve his problem of having to figure out the hard questions of, “What do I need to measure?” but for a lot of the operational issues of, “How do you get data directly from the community? How do you aggregate it? How do you analyze it? How do you run some of the algorithms behind the scenes to normalize the data?” all that stuff we can handle automatically. So he gets the numbers that he can then use to make decisions. So that feeds back into here.
So the last step is going to be thinking about, okay, once you’ve got the data, looking at run charts on a monthly basis is really interesting, but usually you want to communicate the success to your funders, to your community, to your stakeholders, and that’s sort of the third side of it.
So we’ve got all this really interesting data about what’s been happening, who’s been doing it, when have they been doing it? where have they been doing it? And, again, we can use technology to help automate and aggregate the reporting.
So we’ve got a sort of community report here that’s actually bringing together initiatives that are working around common purposes within a community, as defined by some arbitrary geographic boundary. So we can do the aggregation from a community level to a regional level to a national level to a global level automatically without having to do any more number crunching.
All the data already exist within the system, all the faceting, all the slicing and dicing of it we already have, so building these reports, building the data exports for the folks who want to suck it into our tableau, all of that becomes very, very simple and easy to do.
And then you could pivot and you could say rather than looking at one particular issue or one particular region, what if we wanted to look at everybody who was doing everything around a particular problem? And so this is the concept of the map of the movement where you can use, again, all of the data, all of the metadata that we’ve been capturing, look at it from this global perspective on the map. You can drill down.
You can see, you can filter on the side based on network affiliation, based on other organizational affiliations, based on target populations, even based on – you know, you can work from the bottom up and say who’s using a specific measure to capture performance measures or outcome measures.
But we can present it using, obviously, a map interface. And this gives us I think a huge opportunity for discovery for learning who’s doing what, where, who’s maybe being unusually successful in a community, and that feeds back into the beginning of the process, which is how do I then go and drive new change? What do I do differently? How do I improve upon the successes that I’ve already created?
So hopefully that’s illustrative, and that’s a very, very quick overview. I see a lot of squinty eyes and sort of wrinkled brows as to how this actually works. So I’m happy to go into more detail. And, again, we’ve actually built a fair chunk of this and are moving very quickly to build the rest of this.
One of the things that I want to come back to that I think is really worth reemphasizing is all of this is on open data and open technology, so anybody can build applications on top of this. We want to provide the data that’s useful to folks and provide some shared learning about how some of this can be done, but this is not – and the thinking behind this is not the big “P” platform, it’s not the single application that you have to jump on. In fact, each one of these applications can and probably should be developed by a different organization that has a unique perspective on the people that they’re trying to serve and what the utility is, what the value is, that’s going to be most useful for those people.
So thinking about many applications and how that can all be tied together, I think this has actually been, from an operational standpoint, a really successful sort of pilot project demonstration to realize that, yes, actually we can do something like this where you get the benefits of the single application, the single sort of resource for data where it makes sense to have a single resource while at the same time allowing individual data or organizational data to be kept in a separate place when that makes sense for privacy concerns.
So Monte is giving me the eye.
MR. ROULIER: Yes. Go ahead.
DR. REICH: Back to this, right data, right people, right time, right context, right decision. So that’s what we’ve been trying to do.
MR. ROULIER: Thank you, Mike. And I probably didn’t do a very good job of kind of setting up you all a little bit in some ways, but I think some of these capabilities, again thinking about this ecosystem approach when we’re thinking about data technology, I’m part of the Commons team as you all, as many of you know.
We have this beautiful thing that Cara and others have built, but the technology has changed so much that we actually, despite growth and usage, are really rethinking how we’re doing it. We’re changing our core technology and continue not to think about being a single platform.
So all this metrics, the Metrics That Matter that Soma’s team is building, and Brita and Carly, and that whole team, that capacity could be transferred or leveraged with these domains, it could be leveraged in a lot of different ways. And so I think we just wanted to kind of offer up there, there might be some real possibilities of how groups access metrics, how they build applications off that, how we share some of the learning across different fields.
And so we’ve kind of hit time, but I just want to maybe give a chance to see other – we’re going to pivot here in a little bit – thoughts on kind of technology or other things that you want to offer up in addition to what Bob shared earlier on.
DR. PERLA: I think just a quick observation to kind of tie together what I think Kevin and Soma were talking about earlier, and the idea that data and technology are important, but they’re actually in service to the questions that we ask and the answers that we want, and I think we often lose sight of that. And, heck, you could even go further and say that those questions are built on certain assumptions and commitments and world views.
And so I have a health system perspective because my organization has been working with health systems, and today, as we all sit here talking, there is no single health system in the United States that screens all of its patients for social needs.
Everyone who comes in will have a heart rate done, they’ll have their O2 sats done. I asked my primary care doc, “When is the last time you really made an important decision on a heart rate in a patient?” and he said, “Never,” but every single time it gets asked.
So the view of what population health is and can mean I think is critical, and if we don’t endorse this broader view, those questions won’t be asked. The same thing for health plans. There is no health plan that I’m aware of that screens every patient through their health risk appraisal on any sort of social need: food and security, utility support.
And it’s about the questions that we ask. And the Federal Government can play an important role there. And where this plays out in really bizarre places is, how many of you have got experience with ICD-10?
(Show of hands)
DR. PERLA: Yes, almost everybody. There are codes in ICD-10 that will document that you’ve been sucked into a jet engine or pecked by a turkey, but there is nowhere in there that we can document that a patient actually has food and security.
So getting back to the questions that are driving this, I think that’s a really important consideration.
MR. ROULIER: Yes. Thank you for that there.
Bob, and then I have a couple more.
DR. PHILLIPS: That’s perfect. That’s a great setup. So the American Board of Family Medicine has launched a national primary care registry, and we’re part of a 13-specialty registry platform, and we’re actually building out a population health assessment tool that takes the clinical data and uses the ecological social determinants measures from that patient’s address to create a community vital sign for them.
And the idea is that right next to blood pressure and BMI is this measure of risk based on where you live, and the goal of that is to start a conversation. What does risk really mean for this patient? So we want something that appears in the chart every time that just gets a conversation started about, do you really have this risk that is based on where you live?
And then with certification, we have to do quality improvement activities that are actually now a requirement of MIPS so you can get paid for it, too. We want you, if you elect to, can you use this PCAM tool to actually – it’s a validated tool for estimating a patient’s personal risk that can lead you down a more specific set of questions about this person’s personal sociodemographics, how they feel stressed about those things, and so it creates an even more important level of discernment for that patient.
But we want the community vital sign to be sitting there so that they are thinking about it and might think about doing it just like they do now increasingly around depression with the PHQ 2. I didn’t do those on a regular basis until the PHQ 2 was embedded in my EHR, and right next to that measure that my MA did about whether this patient was distressed today.
And what I really wanted to say then is, can you link that into the rich data resources, that we just had a wonderful conversation about, to start to get a better sense of thinking not just about this patient, but who else is like this patient in my geographic catchment area? Where are there hotspots using ICD-10s of disease, and where are there hotspots of that outcome just based on my quality measures?
So can I start to think on a community level of who I can partner with that can help me with a community-level intervention? Can we start to lead them to that using their clinical data and the rich population data that come out of our agencies or out of our communities?
MR. ROULIER: Yes. Great question. Great example.
How about just a couple – we have a little flexibility on time.
DR. LEADBEATER: Richard Leadbeater. I’m a data workgroup – a consultant to that group. As somebody who works for technology or for a company that’s bread and butter is technology, I want to leave this discussion encouraging everybody not to think about technology because those issues are solvable, and there are lots of geeks down the road that can solve them for you. So don’t stymie your creativity of what you need, what you want, what you have to do with how does it happen.
MR. ROULIER: Yes.
DR. LEADBEATER: Because technology is at such a state that there are a lot of things that can happen, it’s just a case of having the reason to do it.
MR. ROULIER: Yes.
DR. LEADBEATER: So don’t – I often get scared of technology conversations because then they get into the nooks and crannies and – stay out of that marsh.
MR. ROULIER: Amen. Yes. And I think that that’s one of the reasons I was putting out there, this is changing so rapidly, there are lots of possibilities, let’s not lock ourselves in.
DR. BEATLEY: So I am going to be devil’s advocate for one second because I love the idea that all this information is available. My question is, how easily is it manipulated? If people can decide what they’re going to use and how they’re going to report it, can they report it in a negative way? One. But you’ve also just said, in a community, what are the characteristics that – what’s your measure of risk? How is that going to – I mean, if I live in a low-income community, I’m at high risk for a boatload of things, but I may not be able to do anything about it. So my fear, it’s a good and a bad, it’s like, how can some of this be manipulated against people instead of being used for the better good?
MR. ROULIER: Good question. Soma and Brita, I both saw your hands up. I don’t know if you want to touch on that briefly and whatever you were going at.
DR. ROY: I guess I kind of wanted to move the conversation back to a lot of what we were talking about before, is that data is useful, especially when multiple people can view that same data and they all feel that they have a stake in the game and they all feel like they have a role in changing it and shaping it.
And so I think one benefit to the platform that Mike and Roxanne shared with us is that there are so many different types of data that people can add to it. So if you’re working in a multi-stakeholder group, you can incorporate things that matter to each person and show how those different measures change over time together or maybe separate from each other.
And by monitoring things over time, then we can really see what our impact is of interventions. Because as we have seen today, there is a lot of data out there, there is a lot of information out there, we can map many, many things, but that just gets us as far as we’ve always known.
Yes, there are a lot of places that have had huge wicked problems that are tied to poverty and incarceration and no access to some basic needs and basic rights for a long time. Most people in these communities actually don’t need these maps to tell you where those places are, right? What they need is a space and a process by which they can actually engage the right people and the right team to start to be able to change those things.
And so that’s what I think we need to move the conversation to, is not just the information availability, but the collaborative tools by which we can actually change things.
MR. ROULIER: Yes. So I’m going to put you and maybe Soma on the spot. Like how does this link back? I know, Soma, you were part of this whole morning to kind of the interests that are surfacing here, I think they were implied in your comments, Brita.
DR. STOUT: So I’m going to actually build on what Brita just said. So I think the important thing to realize is that everything we’re building in 100 Million, this is all a tiny part of a much larger effort to create real change in communities, to create the kind of unprecedented collaboration of people all the way from the federal level to the ground level that can lead to real outcomes. That’s the whole premise behind 100 Million, is that you set an audacious goal and then you figure out how to create the kind of unprecedented collaboration that will get us there, with a big part of that collaboration being the people who are most affected by something actually being major drivers of those changes.
So these tools are in their hands for them to use. While they can have access to things, they really need to choose measures that matter to them. And there’s a whole support process to help those communities actually think about, “What could we move given the assets that we have? given the possibilities that are here? given what matters to us to move?” and about engaging people who have power at different levels to be thinking about how they can use their leverage to create change.
While this is just a glimpse into something, I would just go back to our overall question. What is it, as a group of national leaders, that we’re trying to accomplish?
Technology and data both can be oceans in which we get lost if we don’t have a clear clarifying purpose behind what that is. Is it that we’re trying to better address the social determinants of health in places of poverty? Is it that we are trying to improve outcomes for chronic disease? Is it that we’re learning how to take a place-based approach to population health? What is it that we might wish to try to accomplish together where at the end of, say, by 2020 we could say someone’s life got better because we were here?
But this effort to come together to share this information isn’t just about everyone creating cool tools or about creating great datasets that answer a small segment of people’s questions, but, rather, about showing us that we’re able to make real progress in the things that not only devastate communities right now, but actually threaten our national economy as well as our health outcomes.
If we want to have Healthy People 2020 be a success, what would that look like? And how could we learn our way by combining our efforts to something better? That’s the fundamental ask in 100 Million. Let’s not work separately, let’s work together, but let’s work together toward real change for real people in real communities. And I just invite anybody who is interested in thinking about that together, we would love to talk with you.
MR. ROULIER: So, Soma, thank you for that, and I would like to maybe use that as a pivot.
But, Vickie, go ahead. I didn’t think you were – you had your hand up earlier.
DR. MAYS: I guess I want to build on that because I think part of the question, especially when we get into technology, because I agree with Richard, that we could just be lost, but to go back and say – and this gets to your point of, what is it we’re trying to accomplish? Data can be used either way, it can be used for the 80 percent that have a problem or the 20 percent that don’t, and trying to take the 20 percent and use that as a solution that you want to drive and build on.
I think the biggest issue that we’re kind of struggling with and that we need to kind of focus in on are solutions, and the solutions also coming from the community.
See, I want to tell you, one of the things that community often says about some of the data, I can make data very available to them. Here’s the problem: we often have to collect data to show what’s bad in order to get funding. The community comes back and they say, “Well, can you show us what’s good? You know, I think so-and-so’s program is great, but I want to know before I try and do it whether or not it’s going to really work. So can you do anything?”
So for me, in my role as a researcher, I do what I know I have to do to keep my bread and butter, which is I have to publish papers, and the biggest papers are the ones that say, “This bad thing happened,” but then at the same time, I try and use narratives, I try and use all these other things that I think the community will be able to benefit from.
So I think there should be a bit of a sea change to really think about also resiliency, the positive aspects. We have frameworks and stuff, but it usually is documentation, and it really is, if you ask the community what they really want, it really is, “What’s good? Can I package it? And who’s going to fund it? And will you not leave? Help us do the evaluation.”
MR. ROULIER: So I think it was Elizabeth this morning that kind of even put out this question, what is it that we’re really either trying to accomplish or what’s kind of our premise around what we’re trying to do together?
And I know that some that have been at this for a while have kind of come at this around multi-sector collaboratives. There is this notion of health and well-being and really setting up these multi-sector collaboratives to drive towards change. And that’s at a 40-, 50,000 level.
But there are some other stakeholders in this ecosystem that might have some other interests given particular federal agencies. And, you know, Tyler was here earlier on, our colleague Tyler Norris, and he talked a little bit about this notion of divergent goals and convergent strategies.
So we are thinking about an ecosystem here. There are some divergent goals, and as some folks pointed out, there are some different languages and some different frameworks, we’re not trying to replace any of those, I think as a group we’re trying to figure how we start to accelerate where there are some real commonalities around health and well-being, and more equitable and more just communities, even if we’re using slightly different frames, providing communities with what is really a missing ingredient. It’s the data, the learning, the technology, and the capacity. And I think we’re not alone in doing that.
But I would like to kind of say we could kind of get overwhelmed with how big this is. And what would be some meaningful worthwhile next steps around how we could do that? We know that this committee has some unique influence and is charged with making recommendations to HHS.
And some of the stuff I heard in 3.0 is really fascinating. But there may be some other public-private collaboratives that might either leverage technology, leverage some other pieces, that we could at least put out some next steps, possibilities to explore, and that’s what I think we need to spend the last kind of bit of our time doing, whatever brain energy we have left, if you all are up for it.
Bruce, did you have something you wanted to add?
DR. COHEN: Yes. It’s sort of building on what you were saying, Monte. I totally agree with the world view. People make changes, not data, and we really need to fundamentally keep that in mind. As you know, I’m on the National Committee and helped organize this session. Our goals were more modest. We have a chance to influence, I hope, federal directions around how the richness of federal agencies can help support local community initiatives to make change by providing data and obviously providing tools to use those data and helping communities understand how those data are just one input into decisions they need to make about whatever their priorities are.
So I applaud the notion of the ultimate goal, and we do need to keep our eyes on that prize, but for us, moving into the next actionable phase of this effort, it’s wonderful that we have people with so many perspectives and experience and knowledge. I hope we can pull it all together so we can begin taking the next steps.
MR. ROULIER: Thanks, Bruce.
DR. STEAD: Thanks Bruce. As we get ready to put you back into small groups, I want to put on the table some straw person ideas of next steps that several of us have heard over the course of the day, just as a possible starting point. You can go where you want to go.
So a few things to remember. First, this is from the frame of, what can we say to the Secretary that would be useful for HHS? And we think that one step would be to evolve Framework Version 3 into Framework Version 4 to reflect some of the input we have received today.
Another idea is that we need to probably clarify our language, that the framework, by being broad and being unbiased, is actually designed to accommodate both a parsimonious set of core metrics and a broader base of metrics, community drivers, that will connect locally.
So you might have a parsimonious set of core metrics that would be used to drive federal and state policy, that would be used to compare and rank communities, that would be used for things such as Healthy People 2030. We want those measurable at the sub-county level so that they can be used both to drive state and federal policy and to support a part of the ground-up work that we’re also talking about.
Sitting side-by-side with those metrics and within the framework would be a rich array of metrics or indicators that are helpful tactically to drive progress. They are likely to be much less summative than the core indicators or metrics. And so as the communities work their tactics and drivers and track their progress at that level, it should show up as progress as measured in the more summative core metrics.
So we see the framework as being a common landing pad that could allow these two ideas to work together. And we think it might overcome some of the barrier to parsimoniousness. One of the reasons people are a little bit afraid of that is they have to give up their favorite metric.
Well, their favorite metric can still show up in the framework without being part of this core set that we’re trying to drive uniformly. So think about this as maybe we can address attention by ANDing these two ideas together with the framework as a common landing pad.
The next thing is we need a way of aligning the federal work across departments. There is clearly a lot of very good work going on, but it is not hooked together, and so we need to know what’s the best way to do that. Is this the time to form some form of interdepartmental working group that could begin to coordinate these activities?
We also think that there may be, as we’ve said through the day, an opportunity for new collaboration across the public-private spectrum, both to support learning and to actually begin to get this work done. Are we ready to identify a convening point that will carry this work forward and begin to drive – it may be that point that would drive Version 4 of the framework instead of the Population Health Subcommittee of NCVHS, and it would certainly be that type of convening point that would provide the platforms and so forth that are necessary to bring this to action and would work out where we need additional indicators, et cetera.
So those are ideas we just feed into the discussion.
Bruce, do you want to add color commentary?
DR. COHEN: Sure. My dream job would have been a sports color commentator, so thank you for giving me that opportunity.
When I think about how the Federal Government can support local data initiatives, recognizing that data is just one little piece, I think about four buckets. You might have more. The feds could actually collect the data. A perfect example of that is the effort now to do life expectancy at the census tract level. The feds could help develop and support and disseminate technology to help local folks generate their own data, you know, BRFSS, YRBS, coming up with standard ways for communities to do surveys that would be consistent across jurisdiction would be fantastic.
The third area that I think feds could really be helpful is making small area estimates and promoting small area estimation technology. We can’t do an NHIS in every neighborhood, but there is enough data out there that are being collected where small area estimation might be valuable in small communities.
And the fourth area that I think was really mentioned is the development and support for analytic tools. None of the data make any sense if folks aren’t trained and able to use them in very straightforward and easy ways.
The Federal Government has enormous bandwidth, and, rightly, the priority has been developing data on data support for national policy development. I think now we would like to figure out ways to turn the ship to focus not only on national policy development, but how the federal data enterprise can support community-level initiatives as well.
So in your small groups, think of strategies and recommendations for us to consider. There might be other opportunities, but I’m just putting that out as four possible areas.
MR. ROULIER: Thanks, Bruce.
So let’s do just that. First of all, let’s thank Roxanne and Bob and Mike for giving us some context and possibilities.
MR. ROULIER: And these are kind of the broad buckets of moving forward. I think you’ve heard it in a slightly different way, so I’m not going to add one more version of it. But the next 20 minutes, not a laundry list, how might we move some of this work forward, the evolution of the framework? Let’s not solve it, but what would it tactically to kind of start to populate looking at these balance of measures, this kind of learning?
Maybe there’s an inner group again to cross some of the federal departments to do smaller estimates. What might be some ways to move that forward? And then you heard, again, it might be around the technology, the analytics, around kind of some of the learning networks that we’ve already heard today. How might we leverage those assets in some new and fresh ways?
So I want to hear specific kind of recommendations, even offers, if you will. And I’m going to invite those who are around the corner. I know it’s that time of day, but let’s kind of focus our energy for the next 20 minutes, hear some ideas, and then go enjoy the last bit of sunshine. Sound good?
So please capture the ideas, and we can hand some of those out.
PARTICIPANT: We don’t want to lose all of your good thinking, so we’ve got these, but we would like to collect that as well. Thank you.
MR. ROULIER: Specific ideas and recommendations to move us forward. About 20 minutes of your best thinking.
MR. ROULIER: So how is everybody doing? A little energy before the last part here. The one thing that we’re not going to do is a report-out of everything that you wrote for 10 tables. So I promised we were not going to do that. And there are a lot of really great ideas up here. And I think we have the mind space to really take and build on your best ideas.
And so I want to surface a couple of ideas and see if there are other ways that you would layer onto them so that we’re not kind of duplicating them.
And then I guess another piece is there are probably some ideas that are worthy of consideration but maybe don’t make the cut for the amount of mind space that we have together. So we’ll find some other ways for you either to kind of flag those so that we can get those back to the committee. I’m going to pass this to Bruce and Bill afterwards to kind of talk about next steps.
DR. STEAD: Also, put the table number on your butcher board because we will in fact pick them up.
MR. ROULIER: Thank you. Bill is more logistical than me and more specific. That’s good.
So I’m just going to see if we have a couple of ideas that are around kind of advancing the framework, how we might do that together. That I think is inclusive of a number of things. What are some specific kind of big ideas that you have to help move us forward potentially on that front?
DR. SUAREZ: Yes. I will tell you, we had a really detailed discussion of ideas. We came up with about eight or nine different next steps, organized this into four different audiences of next steps. One is NCVHS. What are the next steps for NCVHS? What are the next steps for the Federal Government? What are the next steps for community organizations? Some recommendations about actions. And then private sector entities. So those were four categories of potential next steps.
So for NCVHS, we consider certainly developing the next framework, very important to go into defining now the next layer, which is indicators, an inside each indicator the metrics, because that at the end is what’s going to be helpful for communities. Make it –
MR. ROULIER: Can I?
DR. SUAREZ: Yes. Well, anyway –
MR. ROULIER: I just want to – yes, on a couple of these chunks, I want others to add in. So this whole idea around moving the framework forward, capturing the indicators and the metrics.
DR. SUAREZ: The indicators. And then for the Federal Government, we like Bruce’s four, what we call Bruce’s four, plus adopting – asking HHS to formally adopt and embed this framework across all operating divisions inside HHS, and filter it through all the programs and activities, funding, grant funding, program operations, utilize this framework so that there is consistency across sectors – across departments in the agency.
MR. ROULIER: So it’s a potential recommendation for the Federal Government.
DR. SUAREZ: Yes.
MR. ROULIER: Consistent application.
So others that had something similar either to kind of how to fill out and kind of populate with the metrics, with the framework? Any other ideas around that?
DR. STOUT: So we totally agree with what you just said. And we felt it was very important to not only look at core measures and community indicators, but to really see that we need a period of learning to understand what actually moves those measures and to take it from that learning perspective, number one. And number two, we felt it was extremely important that subjective measures of well-being, which are well validated to actually be predictive, be integrated, whereas many of the data elements in the framework are already existing in different places and can be pulled in.
And number three, we love the idea of the Federal Government supporting a core set of measures to be measured annually at the very least with small area analysis while allowing a robust flexibility, as well as the tools and infrastructure that would be needed, both in supporting people for data use types of agreements as well as the kinds of infrastructure like what the Commons is building that helps people at the community level to learn how to do things and to track how that impacts the overall well-being measures.
I think if we do that over time, we can begin to get a much better understanding of what moves well-being. And until we do that, it’s a chicken or the egg. We don’t have the measures, so then we don’t know what moves those measures, so we have to learn together.
MR. ROULIER: Yes. So it’s helpful. Just so we’re tracking, it seems that there is a consensus around what would be a core set of measures and to identify those that provide both benefit to multiple levels of geography, but also provide some space, you said not only subjective measures, but some flexibility. So there is both/and that I’ve been hearing all day, right?
Are we all tracking that, that it’s helpful to have some core, it’s helpful to have a lot of flexibility, that there needs to be a learning process by which this gets populated, it can’t just be an abstract with – you know, we’ve done a lot with literature. I don’t know what you all were thinking in terms of a learning process, but there would be some sort of workgroup or something?
Did anybody have ideas around who would help kind of shepherd that process? Did you?
DR. NORMAN: Sure. One of the things we spent a lot of time talking about was – and it ties onto this – but just getting local engagement on this process and thinking through now, that this data is going to be incredibly resource-intensive to collect, to have quality control, and to disseminate.
And the question of core versus flexible came up, but it also came up in terms of, what’s the real demand for this? And how do we know the answer to that question until we get true local engagement? So we suggested doing local pilots ASAP.
We also suggested focusing on tools that are in existence and data that’s already being collected and being very clear about what’s missing in these toolsets and how we can add to them and build on them.
So, for example, Community Commons and the 500 Cities initiatives are phenomenal, and it could be potentially possible for the Federal Government to have a cooperative agreement to build on those resources rather than to try to collect everything in addition or have a whole toolset. And we don’t want to get too down deep into exactly how it happens, but to make sure that we’re being conscious of the resources involved, the demand, and also the effectiveness.
I think we also talked a lot about making sure we thought about this framework. Have we really truly captured what matters? Because I think a lot of the conversation wanted to drive towards, “How are we going to make a difference in our communities?” because that’s really what we care about. But at the same time, that’s not necessarily what this is. This is a set of data and tools out there that communities can use.
And so to really figure out if we’re doing it right, we should ask the question, are we measuring all that matters to produce good change? Because if we leave things off that really matter, we’re being negligent. That’s where we’ve made the mistake around what produces outcomes for people.
So a couple things that came up were residential segregation as an issue that was not a subcategory. And that residential leadership, we have a lot of discussion about community vitality, but the subcategories, we really seem to be passive in terms of how folks engage.
So if we really think about what matters for making a difference and how this data reflects it, we need to think about the subcategories as well.
MR. ROULIER: Good. So one observation today is that our conversation this morning, again, we were saying this framework is directionally correct, it is inclusive and comprehensive in some ways. At a high level, there is some real value to it and the premise behind how it might support particularly some deeper federal collaboration across a number of fronts, it could be useful. But there is this deeper learning actually at a community level, testing, piloting, as we start creating metrics in what really matters and what are some of those process measures that groups could leverage to create change. That’s some of the heavy lifting that’s still to be done.
DR. STEAD: Let’s pause and just really try to put a point on Monte’s point and see if there is agreement about that. So I think it would be very helpful to know the degree to which people agree that the framework, which are the domains and subdomains, not metrics or indicators, that the domains and subdomains are directionally correct and a good starting point, presuming they will continue to iterate, but that that part is relatively okay.
The place we need work is the effort to figure out which indicators and metrics need to be linked in and how to harness the learning and practice with that at these two very different scales of core metrics and broad-based community.
But I just want to get a fine point on, are we, in fact, in agreement that the domains and subdomains are roughly correct and that where we need to be working is in this other space?
MR. ROULIER: Do you want to just kind of shake a limb if that’s generally true for you?
PARTICIPANT: As long as you add subjective well-being to it, which is missing.
PARTICIPANT: Subjective well-being is a subdomain.
PARTICIPANT: In what area?
PARTICIPANT: I would think that would be a metric. Wouldn’t it? I would think that would be a metric. Am I misunderstanding?
PARTICIPANT: Or an indicator.
PARTICIPANT: An indicator.
MR. ROULIER: Self-reported. I mean, it’s still the scope of this work, I mean with this framework. I hear your –
PARTICIPANT: I just don’t see an overall.
MR. ROULIER: You would like to make sure that that’s anchored in there.
PARTICIPANT: I stand up for that.
PARTICIPANT: So we’ll put it in our outcomes.
PARTICIPANT: Oh, I’m sorry.
DR. HOMER: To answer the question, do I think this is in the right general direction? Yes, but I want to acknowledge the limits of my own perspective on this as a health-oriented person, a public health-oriented person.
So I thought the last panel was really – the federal panel was interesting because there are examples where others in these sectors we’re trying to work with are already working on these things. And so if we’re really trying to model this cross-sector thing, I think we can have a framework, we can have a set of domains.
I think to the extent we start to drill down into the specifics within a domain, I’m not sure that we should assume we are the only ones that are the relevant parties to that. So if there is a transportation effort already underway that’s looking at health and transportation, we ought to figure out how to buy into that or have them buy into this so that we can have a framework of our own for what we think of as health related, but when we get into somebody else’s sector and we want them to participate in providing data and providing examples of use, I think we ought to pause and not assume our usual posture of defining for everybody else what data they collect.
MR. ROULIER: Okay.
DR. COHEN: And I think you’re absolutely correct. Part of the framework development, particularly around the domains, was to pay attention to all of the secretariats and agencies in an effort to reflect the work that they have done and are doing and are continuing to do. So that was sort of embedded in the domains and subdomains an effort to capture all of that work and not just from HHS.
DR. MAYS: Bruce, this is Vickie. I still don’t think – and I’ve said this from the beginning, so I’m willing to be a broken record on it – that we have the racial inequality piece in here well enough to reflect well-being and emotional kind of well-being that we want to see. So things like discrimination, things like missed opportunity, inequities in things. I just think that those things need to be in there if we’re helping, you know, it’s health for all kind of orientation.
MR. ROULIER: We’ve got a couple more and then I’m going to call a question again. So maybe back here, right here, and here.
PARTICIPANT: We had a lot of discussion on this last point about filling in – whether the domains are close enough. And so I think we had a lot of conversation around the table that says the domains are probably close enough and we ought to actively be working toward indicator development.
MR. ROULIER: Yes. And that’s what I want to float again, your comment there. So a couple others so we don’t get too lost in the indicator rabbit hole.
DR. WANG: This is Claire. And just one comment on the subdomain of health care infrastructure. I think public health infrastructure needs to be in there as well. So whether it’s health care and public health infrastructure, so things like accredited well-resourced health departments, things like that.
MR. ROULIER: Two more, and then I want to ask a question.
DR. LEADBEATER: I think one of the main tenets of the framework needs to be that we don’t know what we don’t know, and that if we try to put too fine a point on finishing it, we’re done. And then it will be crippled, it will always be taken apart. So this has to be a moving target.
Yes, we have to address HHS, yes, now. They’re going to fund this going forward, but we have to be inclusive of other federal agencies, like Transportation, Census, and their additive attributes that they eventually will apply, but it has to be an explicit tenet that this is a moving developing framework.
MR. ROULIER: Yes.
MS. HINES: Can I ask a follow-up question? So how does that happen?
DR. LEADBEATER: Do you want to answer that one?
DR. NORRIS: So, in fact, periodic review is the typical way to do that. And to give an example, the county health rankings actually change – has an annual review. Basically, they have a framework in which they have specific items and indicators that are weight supplied to individual indicators and subdomains which then generate the overall rankings.
But they basically review it every year to see if there is now new evidence related to, for example, a particular indicator related to physical activity that might actually work better than the one they had before, or, in fact, should environment now not contribute 10 percent towards the total weight, but should it be in fact 15 percent?
So I say simply some sort of a periodic review and ability to, in fact, as you say, it can evolve as is needed. That’s, to me, the sensible thing to do. It may not – annually might be a little too frequently for this kind of thing, but that would be my suggestion, that that be built in somehow.
MR. ROULIER: So I’ve heard two ways of evolving this, what will invariably evolve. Whether it’s a moving target or not, I think we’re lost in kind of abstract semantics at this point, maybe trying to put too fine a point on it, but is it a good starting point?
Is it as we start to get into metrics and subdomains and start to see what other groups are doing, will we learn other things and will we have to adapt based on it? I think we’re all saying yes.
And I’m thinking you’re saying there is some learning with communities in a pilot mode that is a little different but complementary to more traditional methodology with a periodic kind of review of this, much like County Health Rankings is doing.
So, again, please, Bob.
DR. PHILLIPS: I want to pick up one thing that Gib was talking about. So we have an ingredients list. We’re not even sure what we’re baking. And so the temptation is to keep adding to the list so you don’t go to the grocery store and forget something. There is still some work to be done on what you’re going to bake, and that’s not just about what ingredients you need, but how much of each. So there is a lot of work that has to happen on, how do you weight these? how do you use them? what for? So at some point, we have to stop adding to the ingredient list.
MR. ROULIER: So, Bob, I think the question that Rebecca was starting to ask because I think it’s – what would be a process by which we would start to answer a number of these questions? To what end? What’s the purpose of this? What’s the utility value? What needs to be added to that? How does this adapt and interact with other domains or other frameworks?
Did anybody have a really great idea around what a process or a convening capacity could be?
Do you have?
DR. BEATLEY: I kind of have a question and process, because – so you’ve finally gotten to the idea that there should be maybe a core instead of being entirely flexible. But if that’s true, if you want this to be something communities can use, shouldn’t they be able to say how they want to weight different things depending upon their kind of objectives?
MR. ROULIER: And I think that’s a really good question that I don’t think that we have to answer here. I think it’s a question that gets to, what’s the utility value of this to various types of multi-sector collaboratives doing it?
DR. STOUT: So building on both these points, I think we need going forward a process that brings together people working at the federal interagency level and people working in communities to be doing this together. I think that that can help actually evolve both what the common core is, and I think we need to think hard about what’s the delta between what we already have in our measurement systems where most of this can be pulled from all of these great tools that we’re already using, and what’s the delta? And that’s really why I brought up the subjective piece that we don’t currently collect that could be integrated into federal data and measurement systems.
But I think it’s just going to be just as important to have a process of review to learn, what are the things that communities really value as well as tools that help them propose what those might be that allow us to learn what’s in our “don’t know what we don’t know” box. And it’s actually our ability to collaborate across levels that’s going to help us learn what we need to, to get there.
And if it’s helpful, we’re certainly happy to offer 100 Million, which just brings together hundreds of communities as well as some of the people thinking about creating change, many of whom are in the room, as one of the collaborators or supports in doing that process.
MR. ROULIER: Yes. So I want to come back to your offer there and Gib and here and then back over to Vickie.
DR. NORRIS: Yes. Just a quick comment. People have here talked about weightings, and just to be clear, typically you weight these if you want to come up with some sort of a summary measure. For example, either you want an index or a ranking, then you typically take the components and you weight them. But one of the basic premises of this particular effort, at least as I understand it, was that these would not necessarily be used to develop some sort of a summary measure. So there was no weighting or ranking necessarily planned for this.
So I just want to clarify that. If people really think it should be some sort of a rating or an index, which I think that’s probably not a good idea, we should speak up and clarify that. But that’s an important issue because if you start talking about weighting, then it’s a whole nother can of worms coming in there.
MR. ROULIER: Yes. I think you have a lot of agreement on that front. So we have a little bit of an offer that I don’t want to lose because – to see if we can think about how we want to move forward.
Please, some we haven’t heard.
DR. LIAW: Hey, I’m Winston Liaw, from the American Academy of Family Physicians. And just thinking about how to prioritize some of the measures and thinking about how to get end user feedback, given a common platform of using the measures, you can sort of look at what have communities used in a sort of a ranking sort of fashion. Like what are the measures and the tools that people in the communities have found to be most useful?
There can be also a voting process. The end users can say these are the ones that I like the most, and in the iterative process, they can also provide feedback. They can provide comments and say, “I like this measure, but the question I don’t really like is actually hard to administer in the field. I would propose this other measure.”
So I think it can be a statistical research type of methodological review of how do we update these, but it also can be a lot more granular in getting feedback from the end users.
MR. ROULIER: Yes. Thank you for that.
MS. MITVALSKY: Hi. My name is Laura Mitvalsky. I’m from the U.S. Army Public Health Center, and I really appreciate the opportunity to be here.
Just a quick comment. I would be remiss if I left here and said, “Man, I should have said that,” so I’m just going to say it. I was at the – we hosted – or I didn’t host it, Captain Elenberg hosted the National Prevention Strategy Council meeting at the Pentagon last year, and I attended it.
And what I think is missing here, if I may say, is that when I look at this list – and I love the framework – is that there are people who are responsible for each of these. The Department of Transportation, which I love to hear, that they were talking up front, or there was talk about Department of Education, and so forth.
And when they were sitting around the table, it just seems that when I think about in the Army, we say, “Who’s responsible for this?” and they need to be at the table because everyone on this list, there is an agency responsible for that.
So if you think about way ahead, those are the people that we want to buy into this because they’re already supposed to be doing it in some capacity. So I just wanted to put that out there.
DR. COHEN: Actually, that kind of leads me to a possible collaborative suggestion. Perhaps developing –
MR. ROULIER: Bruce, I did promise Vickie just one quick question.
DR. COHEN: Oh, I’m sorry.
MR. ROULIER: But I think let’s go right to you right after that.
DR. MAYS: Well, I was going to say, it’s getting to Rebecca’s process issue. And I think it would be great if before we left, people that have groups in which they can embed this and would be willing to do it, like you made an offer, there may be ACOs, there may be California Endowment. I think we can come up with a way in which, you know, after we get all the federal agencies kind of a sense of who’s responsible for what, that there are some really quick ways to be able to get feedback and to have some adopters to see, are you willing to do this?
If on your way out, you can go up there where Rebecca is, and if you’re going to volunteer, give her the group that you think about. We won’t use your name and say you committed them, but let us at least – no, it would give the committee the opportunity to explore using that group, and I think that will be very helpful to us.
MR. ROULIER: Okay. So we’ll definitely want to kind of capture something.
Bruce, did you want to get –
DR. COHEN: Yes. Vickie reminded me of one thing, and my comment was, given the richness of federal representation here, certainly the notion of developing an interagency workgroup to continue coordinating across domains and across datasets and across tools and aligning these efforts I think is a logical recommendation not only for the Secretary of Health and Human Services, but for all of us who are involved in federal data enterprises. So I think that’s a definite learning and insight today.
I just want to add one thing. When Vickie said – she reminded me, now that we have your names, you will not be strangers to us, you will be able to opt out if you wish, but whatever we develop from today’s workshop in terms of a draft report, we will certainly send and circulate to everyone for their input and their feedback. And however you want to continue to be involved will be very helpful for us. So I just wanted to remind folks.
MR. ROULIER: So this interagency workgroup, as it related with us, is that the one that you and Walter were saying let’s think about this as a recommendation?
DR. COHEN: Yes.
MR. ROULIER: Okay. Right here and here, and then I want to kind of call this question and how we move the metrics framework forward.
DR. WAXMAN: Yes. Our table also talked about interagency workgroup, and I just wanted to add one thing, which is it was suggested that actually having OMB charter that was a way to actually make it happen in an effective manner and may be really important. And that also suggests that getting it elevated to the transition team of the next administration is important because you need White House backing for that.
MR. ROULIER: Good suggestion.
PARTICIPANT: Yes. Great.
MR. ROULIER: Alison, we haven’t heard much from you today.
DR. REIN: Well, the table has.
MR. ROULIER: Good.
DR. REIN: That’s right, spicy table. So I love that suggestion because I’ve been struggling with that myself over here. But my question is actually – and I can’t believe I forgot it, but is there something on here about the IRS and the tax base in the community? And where does that fit? Because I feel like it’s so fundamental as an enabler, and I just didn’t know if it was in here or if it fits in somewhere else and I’m just not sure where it goes.
MR. ROULIER: Is there a quick on that because –
MS. HINES: Could you say a little bit more what’s missing or what are you looking for in terms of an enabler or an indicator?
DR. REIN: So I just feel like the tax base of a community and the distribution of that and sort of the sources of taxable income within a community is an important – and I don’t know that it’s a domain or whether it’s a subdomain of a particular domain, if it’s in economy. But income and wealth doesn’t – like that seemed to me individual, income and wealth, and that’s different from the systematic tax base that you have at a community level.
PARTICIPANT: The spiral of concentrating.
DR. REIN: Right. And the IRS has a lot of data, too.
MR. ROULIER: Yes.
DR. REIN: So it was actually, Laura, your question about who’s not at the table that triggered that thought.
MR. ROULIER: Okay. Quick thought.
DR. HUNTER: Good afternoon. My name is Mildred Hunter, and I’m the regional coordinator for the Office of Minority Health, but I’m in the regional office. And we, too, supported Bruce’s recommendation. However, we took it to the next level. Given the number of interagency workgroups, that we recommended that there be a hub and that this hub, that there would be representatives from each of the current existing interagency workgroups to work on this whole effort and that we would use the framework to identify – no, not me, not us, but anyway, that the group would identify gaps as well as duplications and then build on the framework.
PARTICIPANT: Great. A federal hub or home for this activity.
DR. HUNTER: Yes.
MR. ROULIER: So a little bit more definition for that.
Wayne, go ahead.
DR. GILES: So I guess one of the things we talked about in our group, because there are a lot of federal agency working groups, and, you know, their effectiveness varies substantially.
But I will say one group that has been relatively effective has been the National Prevention Strategy, and that’s been mentioned earlier today. And so I think one of the things you should think about would be, could you potentially have something under the National Prevention Strategy where you’ve got all the cabinet offices, and do it as part of that work?
MR. ROULIER: Yes.
PARTICIPANT: I just have to say that 75 percent of the end use indicators are helpful, and that was by design. So just know that that’s the case.
MR. ROULIER: So that’s a really interesting possibility, OMB, there are some different ways, some comments that we could kind of consider, but it sounds like again there is generally agreement that some sort of cross-agency group is really critical that would need to be well defined.
Also, from a process perspective, there have been a lot of different comments around the process. Soma has a network basically including All In and DASH and some of these other groups. And I think one of the things I’m wondering is that the group that’s been doing a lot of this work around the framework might actually kind of think about, what are the literature review possibilities? What is the work at Thriving Cities at UVA? What are the assets we already have? What are the metrics groups are already doing? What might be some pilots or pilots that are already underway that we could kind of be learning from?
And, actually, I don’t think we can do it in this group, but to create a process that is iterative that starts to kind of figure out what it would take to kind of evolve and operationalize a framework that would work for several different purposes.
So, again, I don’t want to force that, but I think I’m just trying to build on Soma’s offer a little bit around being a little bit of a convening support, and I know you all have done that with opioid with the Surgeon General as kind of playing that convening role.
DR. STOUT: Sure. So in 100 Million, we actually have actually multiple federal agencies working with multiple local agencies and community connectors at every level. So the whole idea is to come together around a piece of work that we need to get done and bring people together across public-private partnerships to try to get that piece of work done.
MR. ROULIER: Yes. And, again, I know that resources are a little bit of an issue, so, I mean, we don’t know how big, how heavy a lift that is, but I’m just wondering if we might want to connect that with a group that’s been doing that. Or how do you all feel about it? I see Wayne’s head shaking, but, Bruce or Bill?
DR. COHEN: Yes. One of the beauties of today is the people who are in this room, the representation of so many different groups and perspectives is invaluable. So not only I think figuring out a strategy for the Federal Government to align its needs, but the Federal Government working more closely with data intermediaries, foundations, and communities is something that hopefully can happen in the growth of this process. So I think that’s a key for us as well.
DR. STEAD: From my perch, NCVHS, in advising HHS, can play a role in catalyzing the cross-departmental work and in bringing the public part of the partnership to the table. It’s out of our purview, I believe, to bring the private part together.
So I think that if we have a critical mass here that could propose and actually begin to come together as a collaborative, we could then connect the two. But we can’t create the collaborative. So I think that’s – if we can get the twin engine of us getting the right recommendations and working to get the right federal collaborations in place at the same time that the private part of this is assembling an active point of collaboration, then we could bring them together. But that would be just a sweet spot, I believe.
MR. ROULIER: Yes. That makes sense, and that was suggesting, by the way, that I think there needs to be a little bit of a pass-out given how heavy a lift you all have been doing to bring it to this point so we could figure out what would be a sane and reasonable process for doing that.
Vickie, you’ve had your hand up.
DR. MAYS: Yes. I think one of the things at this point we should consider is now turning to the foundations. Foundations fund a lot of these programs. The foundations have big commitments. I guess what I’m getting a little concerned about is that we not just advantage who’s in the room and disadvantage who’s not in the room because many of us know there’s a ton of different things that are going on.
We remove ourselves one step from conflicts of interest by probably asking the foundations to do this convening for us next, and then making sure that they represent a great deal of diversity, more so than I think what we have in the room today.
MR. ROULIER: Sure. A point very well taken, and I guess I’m assuming that the missing perspectives, that community level voice, a lived experience, a lot of those pieces, would be involved in this process, and that we don’t actually have the right group. And I don’t know if any foundation leaders want to add to that, but I just think there is only so far we can go in defining a process right now, and actually we need a group to kind of at least get the convening going.
So, Vickie. Vickie just left, right?
MR. ROULIER: Oh, okay. So can we again, just in the interest of time and mind space, I don’t want to force something that feels contrived, but as kind of a next step, some of the critical mass of networks that are doing this work that have an affinity for the same work could at least talk and figure out what would be a process that would be reasonable, that would be a handoff from NCVHS and kind of –
DR. STEAD: Right. I mean, I think, from my perch, it would be worth its weight in gold if – since we’re all people in the room that are able to begin to drive the collaboration, would begin to put together that process and to bring these other foundations, et cetera, into that process, into whatever structure they think would be the best way to align the private side of this while we’re working on the federal side of it, I think then we could land the airplane. I think otherwise we’re in a chicken-and-egg story if in essence we’re supposed to figure out how to make that happen. It’s not in our scope.
So I understand that we don’t have the right diversity in the room, but I think we have people that are fully capable of assembling that diversity and putting in place a structure that would work across the right pieces of the private effort.
MR. ROULIER: Thank you.
Did you want to add?
DR. STOUT: I was just going to say we would be happy to work with a small group of people from within this room who are interested in how that convening might be structured to help organize that process, if that’s helpful.
DR. STEAD: Great. Thank you.
DR. SUAREZ: And that’s exactly what I was going to suggest because I think in order for the plane to fly, we need a few pilots around. And we need someone that can facilitate that convening, and that means someone leading a small group, and so I think that idea is a really great one to have at least a few people from the group come together voluntarily, and certainly I think that the committee can help foster that type of a small group to come together, and then begin to expand and develop some of the next steps.
MR. ROULIER: Great.
DR. STOUT: And it would be helpful if part of the recommendation of NCVHS actually empowered that group to lead meaningful change as they convene. I don’t mean the small group, but the group that they convene to move the recommendations forward.
DR. COHEN: Great. I want to be cognizant of the time. Let me just summarize what I hear as the three major themes that have emerged. If there are other major themes that we should be discussing, please add them.
Certainly, the first one is moving to the framework 4.0 and developing the – Version 4.
PARTICIPANT: Version 4.
DR. COHEN: Okay, Version 4. And developing the indicators and metrics to certainly include subjective indicators, well-being measures, and community resiliency if they’re not already in there. There were other specific suggestions about indicators and subdomains that we need to consider, but in general, the structure of the framework seems to be fairly solid. And certainly these measures, as they’re developed, need to be vetted and go through the lens of community, greater community, input, and need to be measures that really matter in terms of ultimately helping communities in their efforts for change.
The second major theme is the feds need to figure out a way to act more collaboratively in this community data space. There are lots of efforts in a variety of different agencies. They need to be coordinated in a much more strategic and apparent way.
And the third major learning for today is we need to develop a public-private partnership to hand off this effort to, and Soma has graciously offered to help coordinate a small workgroup to do that.
Are there other buckets or large learnings from the day that we want to add to this list?
MR. ROULIER: Do those sound right? Are they on target?
PARTICIPANT: They are.
MR. ROULIER: There is some overlap – right? – between one and three?
DR. COHEN: There is overlap amongst them all.
MR. ROULIER: Yes.
What were other big pieces that came out of your group, if there are any other ones that you feel like we really need to put out there?
DR. SUAREZ: One part of number two is really this suggestion of recommending convening an interagency group, and even part of going to, say, to be set by OMB, or even have a presidential executive order maybe. But I mean, seriously, it would be very critical to formally structure something like that in order to pursue the second bucket.
PARTICIPANT: Yes. Thank you.
MR. ROULIER: Thanks for that.
DR. MCKEOWN: Okay. So just a couple of things that we didn’t touch on in our discussion, but that we discussed, was, one, making data easier to access within the government to other federal agencies. As the gentleman from DOT was saying, he had to like file a proposal to get data from an RDC from another federal agency. It seems like maybe we could set an example in the Federal Government in data sharing. It was a suggestion. I am voicing it for the table.
And, oh, another thing we wanted to say was that we support Wayne’s team’s efforts in small area estimation, and, in particular, for smaller health departments that don’t have epidemiologists on staff, having these tools, such as 500 Cities is fabulous because, in theory, anyone can read them and make decisions. You don’t need someone with an advanced degree in epidemiology.
And to that point, even though this workshop is focused on sub-county, to take it to small area estimation, we wanted to say to focus on rural counties as well because rural disparities are important. They’re getting a lot of attention now with the opioid epidemic. And so to not forget those people because if we don’t have data, it just exacerbates disparities.
So thanks, Wayne. We don’t have any money to offer you, but we wanted to say thanks and keep up the good work.
MR. ROULIER: A couple quickies.
DR. LAURENT: One of the things that might fall under the bucket number two, across federal group, is that there might be some work with Congress that might need to happen with legislative authorities and funding streams that might be creating silos that don’t need to be there. So it kind of hits on, for example, the data sharing piece might be part of that, or even collecting data. So just to keep in mind that there might need to be some work with Congress to remove some of those barriers.
MR. ROULIER: Great. Thank you for adding that.
DR. BARNETT: One thing just to reinforce the earlier comment, one thing we’re clearly learning in looking at the intersection between health care, health and community development sector, is the good news, it takes you down to the small area that we’re talking about.
The challenge is you have to also think at the regional level and you have to look at priorities and funding streams. So you have to have that broader context if you’re going to be successful and if you’re going to scale that. So I think we have to keep that in mind.
And I can’t help but think – I was here in town maybe 4 years ago for a meeting, it was at the fed, and we had a panel of fed representatives, and at the time, there was an existing mandate from the administration to do exactly what we’re recommending that they do, and it didn’t happen. And we need to understand why it didn’t happen.
There are some philosophical binds that we’re grappling with, and one of those is to the degree that we talk – we’re talking about a comprehensive ecological model of health improvement, and that necessarily challenges agencies that rely on categorical funds. And if we start mixing that up, the fear is that the public will cut funds for those programs.
So there’s an inherent bind that we have to acknowledge, and challenge, and one of the advantages is this better data will enable us to break out those dimensions at the same time we’re looking at ways in which to proceed with that integration.
So it’s not just acknowledging that there are challenges that our federal partners have grappled with in trying to move in this direction, and this is one of those significant challenges.
MR. ROULIER: I think that’s well said. I see Wayne shaking his head, and many others that would agree with that. The area is not insurmountable, but if we’re not realistic about it.
So we have about 9 minutes, and we said we would honor folks’ time. I want to pass it back over to the co-chairs. It sounds like we have some agreement, at least on some ways to take these ideas to some next steps. And I think when we’re in a room like this and we actually are trying to collaborate, and we don’t have an organization so much, it’s a little bit challenging to figure out what that looks like.
So I just want to applaud you all for just kind of hanging in there with some ambiguity until we figure out, what would be some next steps that would kind of help move some collective interest?
Soma, did you?
DR. STOUT: Yes. I have this feeling of anxiety now. So I have a request. So can I see a show of hands of who would be willing to be part of that small working group, figuring out the right convener, and people who have sort of the federal as well as private place?
(Show of hands)
DR. STOUT: Great. Thank you. Can you all please give me your cards before you leave?
MR. ROULIER: Good. I was going in the same direction around seeing if we could get people –
DR. STOUT: I feel much better now. Thank you.
(off mic comment about sending out request through email.)
DR. STOUT: That would be terrific.
PARTICIPANT: We have this entire email list.
DR. STOUT: Thanks, Bob.
MR. ROULIER: And I think to Vickie’s point, I mean, this is a really great representative, a lot of perspectives in this room, and what we need to do is create a process that’s going to be inclusive, there will be different ways for people to participate in that.
So in any event, I just want to applaud you all for actually, one, kind of moving through a day like this where we’re dealing with enormous complexities. And I think I find in doing collaborative work, I know there are a lot of us that do work in these areas, that if we can start to create some areas of coherence, there’s a snowball effect, there are some pieces that we can build on, and I think that we have some nuggets here or some opportunities for coherence that could kind of snowball off of.
And I want to just thank many of you who kind of got us to this point with NCVHS, and the heavy lifting on not just the framework, but the convening is quite extraordinary. And I’m going to, on that note, pass it back to the two that have kind of been shouldering a lot of this for any final thoughts that you have.
DR. COHEN: The sun is shining, and go out and enjoy the rest of the day. And thank you so, so much for honoring us with your time.
DR. STEAD: I’ll second the thanks. I also want to call out special thanks to Kate, Brett, and to Rebecca Hines. We have run since last February essentially in a forced march to get from Version 1 through the environmental scan to Version 2 to do the vetting to get to Version 3. And it’s been a remarkable journey that has taken real dedication far and above any form of normal work hours. So I want to just thank them.
DR. STEAD: And also the team from NCVHS that you see, our staff, that you see around the room. They’ve all been extremely helpful.
And thank you for your help today.
MR. ROULIER: I think we’re officially adjourned.
(Whereupon, the meeting was adjourned at 5:00 p.m.)