Department of Health and Human Services
National Committee on
Vital and Health Statistics
Full Committee Meeting
February 7, 2019
Hubert H. Humphrey Building
200 Independence Ave., SW
P R O C E E D I N G S (8:30 a.m.)
Agenda Item: Welcome
STEAD: Welcome back. Hope everybody had a restful evening, feeling confident in the fact that you had really done a lot of work yesterday. So let’s start with roll call. I am Bill Stead, chair of the Full Committee from Vanderbilt University, no conflicts.
MAYS: Vickie Mays, University of California Los Angeles, no conflicts.
LANDEN: Rich Landen, member of the Full Committee, Standard Subcommittee, no conflicts.
CORNELIUS: Lee Cornelius, University of Georgia, no conflicts.
COHEN: Bruce Cohen, member of the Full Committee, co-chair of Population Health Subcommittee, Massachusetts, no conflicts.
PHILLIPS: Bob Phillips, American Board of Family Medicine, member of the full Committee, cochair of the Population Health Subcommittee, no conflicts.
KLOSS: Linda Kloss, Florida. I’m a member of the Full Committee, co-chair of Privacy, Confidentiality and Security Subcommittee, member of the Standards Subcommittee, and no conflicts.
STRICKLAND: Debra Strickland, member of the Full Committee, member of the Standards Subcommittee and the Population Health Subcommittee and no conflicts.
COUSSOULE: Nick Coussoule, BlueCross BlueShield of Tennessee, cochair of the Standards Subcommittee, member of the Privacy, Confidentiality and Security Subcommittee, and I have no conflicts.
STEAD: Members on the phone?
GOSS: Good morning. This is the Alix Goss with Imprado, member of the full Committee, cochair of the Standards and Review Subcommittee and no conflicts.
MONSON: Good morning, Jacki Monson, Sutter Health, member of the full Committee, member of the Privacy, Confidentiality and Information Security Subcommittee and no conflicts.
STEAD: Rebecca, would you start with staff?
HINES: Yes, thank you and good morning everybody. We have a quorum so thank you for those who are here in the room. I know Denise Love is also here and she can check herself in when she gets back to the table. I’m Rebecca Hines with CDC, National Center for Health Statistics, executive secretary, and welcome today too. I know we have staff on the phone. Did you want to say good morning.
DOO: Yes, this is Lorraine Doo with Centers for Medicare and Medicaid Services, lead staff to the Standards Subcommittee.
LINCOLN: Mike Lincoln from Department of Veterans Affairs and University of Utah.
HINES: Thank you, Mike, and then we’ll just go around the room.
NIEMEYER: Suzanne Niemeyer, Ketchum.
KANAAN: Susan Kanaan, writer for the Committee.
KOCHER: Gail Kocher, BlueCross BlueShield Association.
HINES: And we’ll introduce our speakers when it’s time.
STEAD: Let’s just briefly look at the agenda. I think it will work quite nicely. We’re going to this morning as planned with the two blocks on Population Health, starting with the update on the implementation of and evolution of the measurement framework on health and wellbeing. And then hearing from Healthy People 2030. We hope these conversations may connect with each other. And then Steve Schwartz for the discussion of Vital Statistics.
Then after that block Jon White is going to join us to continue the discussion for collaboration between NCVHS at ONC and the HITAC Advisory Committee. Then after lunch we have a block for action items from day one final revisions which fortunately we do not need to use for that purpose so we’re going to bring back the work plan block, and in particular, the ICD scoping discussion that we didn’t get to at the end of day one and then roll in to health information privacy and security beyond HIPAA.
Linda, I hope in that block we will be able to shave the fuzz off the peach of the language we want to use for the largely unregulated space of the patchwork. Because if we can land that language for the working session document, we will then be able to pop it back in to the report to Congress and make those light up. So it would be nice to land that while we’re all together.
KLOSS: I have a proposed fuzz off the peach landing that Rachel and Maya worked on yesterday.
STEAD: Outstanding. We are looking forward to it. Then we have public comment and then we’re done with day two. So is everybody good with the agenda as revised. Then let’s proceed with the committee updates.
Agenda Item: Committee Updates
PHILLIPS: So, we have a guest with us today to be talking about the Population Health Measurement Framework, but before we do that I just wanted to give a little bit of update about the Board of Scientific Counselors for the National Center for Health Statistics. I sent you all, from that meeting, a summary so I won’t belabor a lot of it. I’m glad that Carter Blakely is with us today to update us about Healthy People which is one of the conversations that we had there.
Just very briefly, the good news about NCHS, it’s good news in most years, is that their budget is flat. It wasn’t reduced as might have been proposed. They have started hiring again which is very good news.
The Federal Data Strategy was probably the biggest news that I shared with you all in that it has moved incredibly fast. In fact, this last month they were supposed to get through with their draft first year plan. It’s not up on the website yet but there is a very helpful slide set there that kind of outlines the whole process.
They did a lot of use case testing at the end of last year. I let them know that this process was happening and that it was built on use cases and it would be awfully nice to try and harmonize some of those so that when we came through that data strategy, we had these measures as a potential product. And invited them, in fact, to come today to join us and sit in on this very important presentation.
That Federal Data Strategy process for the generation of the first-year plan will culminate in April and we have a little window of opportunity here where they are taking the draft plan and getting input on what they should be doing the first year. So we have a very narrow window if we want to engage that process.
There was a very helpful presentation about the NCHS data systems by Carol DeFrances and Pete Meyers. So the National Hospital Ambulatory Medical Care Survey or NHAMCS, those are all visits that happen to clinics or emergency departments affiliated with a hospital, is being rolled into the National Hospital Care Survey so NHAMCS is an annual event. The hospital survey is not. In fact, it hasn’t been done since 2010.
But they are rolling them together so that they have a collective data stream about everything happening in hospitals. And in fact they got PCOR funding to do some trialing with about 48 EHRs across a number of hospitals trying to do direct data extraction. And the hope is that the health care electronic health record infrastructure that they are starting to bring online will start to merge with other products, like the ability to link with CMS with NCH and NDI data.
And this is part of the general effort under the Federal Data Strategy to increase security and access and the vision for that is to bring them into about 30 research data centers across the country so expanding from the two that they have currently where researchers and others can come and use the data with those linkages. And I think I’ve expressed before one of the concerns about that is it makes those data less available to the general public and other users if they don’t have a way to get to an RDC or don’t have a partner, even though it makes the opportunities more robust.
Also, when you start to connect with Census data, it brings the data security requirements under Census’s separate statutes which are much more restrictive than most of the National Health Survey data security. So they are still trying to work that out.
The other thing that Pete Meyers did that was interesting is that this has become so complex that there’s a concern about the loss of data ownership from the health side and so they are actually looking at Oakridge and the Department of Energy Health Data Institute to be a solution for managing both the security side and the access side of data.
So it doesn’t sound like at that time it was fully baked and we only met for one day in the fall because of President Bush’s memorial. And it wasn’t discussed again on a follow-up phone call that we had in December. I just want to make sure that you all know that this was happening.
Carter Blakely is here to talk about Healthy People 2030, so I won’t go back over my summary to you all about what I learned at that meeting.
And there is a focus on – in December the one thing that I wanted to bring your attention to was there was a focus on vital statistics, and particularly about the capture of medications. And particularly about opioid prescribing. It’s that linkage to understanding medication-related deaths and opioids, in particular.
There I introduced them to all of our efforts around vital statistics and Next Generation and the need to kind of bring those ideas together so that we have one effort going forward.
And I think that’s it for that summary.
STEAD: Thank you, Bob. Are there questions for Bob? Then Alix, do you have an update for us?
GOSS: I do. Thank you, Bill. As a part of our Predictability Roadmap effort providing the industry with more than 60 days of opportunity to review our 22 recommendations and narrative report, we were pleased to support a request to deliver, in person or webinar presentations on the materials, enabling us to support questions and answers and help the industry as they prepared their respective testimonies.
I did several different presentations including an in-person session to the Workgroup for Electronic Data Interchange at their fall event in DC. I also supported the National Uniform Claims Committee, the National Uniform Billing Committee, and CAQH CORE’s webinar. I think that Deb Strickland also has an item to share on that one and I encourage any of the other subcommittee members to share their updates related to the Predictability Roadmap webinars.
STRICKLAND: So I also did a WEDI webinar on the Predictability Roadmap for the Workgroup for Electronic Data Interchange.
MAYS: Previously we had talked about the changes that were going to go on to the Census and we were trying to see if we had some bandwidth to be able to do things. I talked to, since that, the Statistics Committee that’s with the National Academy of Medicine. When we have our meeting, I think it’s in June, I think they are having potentially, a hearing at the same time that we’re going to be here. So they may want to talk with us to see if there’s any way that we can participate as experts coming over to talk with them. So it’s still in the works but it’s kind of on the drawing board.
STEAD: Very good. Any other updates? Okay, then Bruce, would you like to introduce the next block?
Agenda Item: Population Health
COHEN: Thanks, Bill. I am proud to introduce Soma Stout, who is Vice President in the Institute for Health Care Improvement and executively, for 100 Million Healthier Lives. The inspiration behind the collaborative that is taken our Measurement Framework and built on it and expanded it to create what they are calling the WIN framework, Well-being in the Nation framework. The collaborative I think is the only group I’ve been involved with that works just as hard as this group.
So over the last year the accomplishments have been outstanding and they have developed an almost final report that was just passed out today to the full Committee and hopefully over the next several weeks we will have a chance to review this report, finalize our feedback and then endorse this project as it moves forward. So thank you so much Soma, the floor is yours.
Agenda Item: NCVHS Measurement Framework for Community Health and Well-Being: Update on Progress Toward Implementation
STOUT: It is a pleasure to be here. Thank you so much for having me and I just – as you hear sort of the process we followed, I’d love for you to really think about does this meet the approach and the principles that you set out when you created the initial report. And what ways do you think you might guide us as into implementation and we had a couple of key decision points that we really need your input on before we sort of finalize and bake it and say all right this is tied in a bow for now.
And I say tied in a bow for now because we really see this as the measures behind this framework as a living library of measures that will continually be revised and updated as they are used in the field and that offer a guide for what it means – to help us learn what it takes to actually improve well-being, but also gives us an opportunity as we learn to be able to seamlessly update and to be able to adapt as things get implemented in the field.
And hopefully as we hear of some of the stories of how even before – it’s just because of the way it’s been tested over time – the number of groups that have adopted before it’s actually been released formally is actually pretty extraordinary. Which means that we will have a really rich array of places to learn from as we go into this next phase of our work.
So for our time together, I’d love for us to review the process we’ve taken so far, to discuss how the framework has evolved, and to look at some key decision points to make sure that you feel these are consistent with the approach and principles that you laid out.
So for the work today, of course, version four of the NCVHS Framework was released in January 2017 with a letter to the Secretary in May of 2017, and then you passed it off to a group that could lead a federal/nonfederal process for shared implementation.
And 100 Million Healthier Lives, for those who don’t know it, is convened by the Institute for Health Care Improvement, but really represents a partnership, what we call an unprecedented collaboration across thousands of organizations to transform the way we improve health well-being and equity in the world. We say that working on equity is the price of admission. I wanted to share some of our core principles today because they’ve really reflected how we’ve approached this process.
So for us, the idea of nothing about us without us is critical in implementation so we partner pretty deeply with whoever we think are going to be the intended end users of something all along the way, not just at the end when we’ve come up with something. And that you’ll see reflected throughout the process.
The ability to sort of what we say, promote equity, justice, strength and wholeness and to have a humble posture of learning and improvement. That’s really been applied here in that as we’ve gone it’s been an iterative process of developing this framework, bringing it back out to the field, seeing how people respond and saying, oh, for this group of users they are really going to need to see this in order to see themselves in it and be able to use this. And so that’s really shaped how the framework has evolved over time.
And then our ask to you is to help us with four and five. Four I think is how can we together remove the barriers from people on the ground who are trying to learn how to improve health well-being and equity in communities. I really think in the way the collaboration with the NCVHS committees, especially the Population Health Sub group, is an example of making this an example of what’s possible and how we work together. I’ve been so grateful for the support and guidance.
So as we thought about this, our goal here was to arrive at measure sets aligned with the NCVHS framework domains to create a measurement ecosystem that both had a parsimonious set of core measures – this is what was asked for when the framework was handed to us – as well as an expandable what we call, branching menu of measures that could actually go on to a fairly large set of branches to find what people needed.
To achieve that balance of standard, widely-used measures, along with developmental measures that show great promise that there isn’t yet good evidence for sort of widespread use. So how might we balance that to create sort of a flexible set that was more a set of developmental measures.
And then how do we include a breadth of experts, both across sectors, people who think about measurement in different levels, as well as community residents, thinking about those who have lived experience as their own kind of experts in the process.
So the process to date – we started in the summer of 2017 by convening a Metrics Development and Measurement Implementation group. The two groups together represented close to 100 organizations, and many of you have served on it at one point or the other, established a Stewardship Group to oversee the process, built on a landscape analysis that had been created by the Population Health Subcommittee, and then expanded on as others said, oh, you should also look at this and this and this. Co-developed and adapted criteria that had been developed in the National Quality Forum and then adapted for this purpose.
And then based on that landscape analysis, compiled hundreds of measures and then rated them based on the criteria which included validity, availability of data, usability, feasibility and value to stakeholders at the national and community level. And we asked people to rate those separately. You can see just a set of measures that were added to what was already created by the initial landscape analysis.
We then underwent a Delphi process that included over 100 organizations, first for people to add what was missing from the initial set, a second for people to prioritize and then third for people to evaluate. And then we brought people back together to say, across the development and implementation groups, to say how does this resonate, what’s missing and where do we see opportunities that we saw as quality improvement opportunities in the process in terms of filling out for instance, some domains that just didn’t have enough measures expertise.
So we then decided to do a fourth Delphi round where we reached out to experts in some of the domains or subdomains that didn’t have strong votes in the Delphi process, so we could really get that expert validation.
But the other things that emerged were people strongly felt that well-being, like the well-being of people and equity, needed to be there and deserved to be their own separate domains. That health wasn’t – if we’re really moving toward a frame where social determinants are understood as important, that we needed to help see how people valued their well-being as a good in and of itself.
And then equity just emerged as core. Then people felt strongly that it shouldn’t just be a crosscutting thing within domains, but should be something that is its own standalone set of measures so that we really adopted the idea of it being both.
From that we provided an update to the NCVHS Committee at that point, and then this fall, have been drafting core measures, what we call leading indicators in a flexible expanded set, and then have done some terminology changes. So I’ll talk a little bit about how those things have evolved and why.
So the Initial Framework – this is where I say the framework evolution in a slide. On the left-hand side, are the original 10 domains, and then their associated 30 subdomains of the NCVHS framework. We added those two domains of well-being of people and equity and then as we went out to try to talk about this, we struggled. We found that people liked it, but they had a hard time actually talking about what it was beyond a framework that was about multisector measures. They got lost in 12 domains and 30 subdomains.
And so there was a need to sort of synthesize it back into what we thought of as – when we really looked at like what people cared about, there were a group of people where their primary lens was the well-being of people and they cared about people in communities but that was what they did. That was like business and health care and social services and some community-based organizations. That’s what they thought about.
Another group of people, their lens of the world was the well-being of places. They were thinking about the built environment and structural changes and environments, and that was sort of a second big group of people. And the third were – these were of course crosscutting with those who thought about equity a lot and that was really their lens and driver for creating change.
And by creating these themes around the well-being of people, the well-being of places and equity – and we’ll talk a little bit about how we actually put that into the framework – it was like this collective sigh of relief sort of went out into the field and that’s the best way I can describe it. Suddenly people went from, oh, this is nice, to galvanizing into action. I remember sharing this to the Well Being Legacy group that had said, hey, we really want to see, could you show us a draft of where you are now. And there was a legislator in the room who literally, by the next morning, had drafted legislation for her state to adopt this framework and to look at it at least in terms of measures.
And that’s a kind of mobilization to action that we found and that was across the board. And what’s been interesting is I’m now finding out after the fact that people have adopted in some ways. So what’s interesting is how much adoption has just begun to happen naturally.
So if it’s helpful I can go over sort of why the well-being of people is a strong priority, as I mentioned, and for key sectors it really mattered. When we looked at what measures were available, it turned out that there were some very simple, powerful, very evidence-based measures that could be used and hadn’t been tested. For instance, the Gallup Well-Being Index had two Cantril’s ladder questions that have been administered over 27 million times around the world.
And in this period of time, like OECD has also actually said people-reported outcome measures of well-being are actually how they recommend measuring population health. And so that thread of other countries going in this direction, alongside, what we found is a meaningful set of evidence-based singly validated set of measures were useful.
And they resonated as common measures because people at every level could see themselves in a simple ladder, for instance, where you see your worst possible life and your best possible life. Clinicians could use it to do goal setting with their patients. In fact, there are some pilots already that are beginning, combining that with a health lead, social lead screening, et cetera, in a couple of different places. But it could roll up to the population level to give, in a validated way, a percent of people thriving and a percent of people suffering using just two simple questions, and others.
And then similarly, when we have hooked back – we actually – and this accompanying document sort of goes over some of how the framework has evolved in two different lenses. It actually reflected a lot of – it was actually we found something that was there at the very beginning of the framework that had gotten lost along the way, and during the testing process it just emerged as important again. So we really saw it as going back to the beginning of the whole process in the first place.
Now, I want to be clear. We didn’t lose any of these. It’s not that we reorganized the whole framework into these. What we did was we said this is for communication. This is how we can talk about the core measures. But the well-being of places we really left as the initial NCVHS framework as it was, is sort of in that entire theme of the well-being of places. And that just made it easy for us to be able to preserve the framework as it was, but to enhance it with these ways in which we could more easily communicate about it as well as get people to adopt it.
And then people who were in the know knew what NCVHS meant but others didn’t necessarily, in the community didn’t. So we thought a simple – we find that acronyms that are easy to say and remember work and Well-being in the Nation – there had actually been an initiative called Well-being in the Nation that had actually had as its recommendations to have measurement around well-being in the country that had then been – we participated and they transitioned to 200 Million Lives to support. We said, well, that was a really great name. People liked that name and then we tested it out in the field again, it just received great response for people.
And so we talked a little bit about this. I think I don’t need to go into why equity matters to this group but let’s just say both of these recommended by the stewardship group. And we talked a little bit about why we made some of the changes to the framework.
So the way the framework is now organized is around these core measures, around these themes of the wellbeing of people, the wellbeing of places and equity which is around – wellbeing of people is based on people-reported wellbeing measures and then life expectancy. The wellbeing of places, which is the initial framework domains and subdomains, is based on a Healthy Communities Index which we’ve said the US News and World Report and County Health Rankings and Roadmap sort of capture well the general principles and domains of this.
What we found though was some groups had a hard time with just an index and so child poverty kept emerging as an indicator measures. We basically then went around to different groups that were big measurement groups and said if you could pick one measure for a healthy community what would that measure be.
And without just asking an open-ended question multiple ones of them said child poverty because they said it’s an indicator measure that indicates a whole bunch of other things and correlates with a whole bunch of other things in their measurements. And that included every City Health Dashboard, the C.A.R.E.S. group, US News and World Report, County Health Rankings.
These were the groups that think about these measures all the time in implementation who recommended that as a single measure. And again, what we found was for those groups that weren’t ready to think about an index, they loved child poverty. It was motivating. It made sense to them. And they could adopt it – a health system could adopt it and say this is what we’re going to do for our efforts.
And then equity was a difference in the wellbeing of people. We thought about just saying the difference in life expectancy but from a prevention perspective, we found that for public health, years of life gained was much more compelling and for public health professionals to be able to make the case for prevention. And that really – that will of course have difference in life expectancy based on demographic factors but the thought of actually also helping people easily get to years of life gained would be useful for those in public health who are trying to make the case for prevention.
And then other things around income inequality, graduation rate, and then of course differences in demographic factors by race, gender, language, place were things that emerged.
And then beyond that those initial – remember the 10 domains – we call those the indicators related to those 10 domains, the leading indicators. And these were the prioritized measures that came out of the Delphi process. And then behind those leading indicators is what we call a full expanded set that has many more of those developmental measures. So the leading indicators are ones where you really – you can see this in the appendices that are here in your document. These are ones that emerged in the Delphi process where through multiple rounds, they were the victors of the process. They were the ones people said these you’ve got to have these for me to see myself in this.
In addition to that, the full expanded set has many more of those actually really interesting developmental measures, some of which are not available, like perceptions of everyday racism and yet there’s early research to say this actually is going to be an important measure as it relates to life expectancy for people. And so as we allow that research actually giving us space for it, will we think allow researchers and others to be able to have a common set of measures that they are testing out in their work or communities as they are seeking to create improvement to be able to know does this make sense.
Another great example is the American Heart Association decided to embed the NCVHS approach into its setting of the 2030 goal which is what the entire Heart Association organizes itself around. And that was a really exciting development and they loved the framework. And as they adopted it, they said, okay, but we’re going to do healthy life expectancy using the Global Burden of Disease and they are working with IHME to look at local level analyses, like at a county-level analysis of that. Now that isn’t available now so it can’t be a leading indicator because that was one of the cuts we said, is things have to be available. But it’s in the developmental set so that if that becomes available and as the Heart Association is testing it, it becomes useful, we could promote it to be a leading indicator. There are other reasons we wouldn’t right now. Life expectancy at birth is available at the subcounty level so anything that is available at the subcounty level and as an equity breakdown is something we prioritized and that’s what life expectancy at birth represented. So we wouldn’t elevate it to a core measure but that’s the kind of decision point around the principles of what makes it to a leading indicator but what allows us to actually have a rich array of developmental measures that can be added to in much more real time tested and then promoted as needed. And things can be demoted as well if we find that things actually nobody adopted this great measure that we all loved but nobody in the field actually thought it was useful to them.
And so that’s really that branching set of measures in each domain and subdomain and living menu is what’s there. This just shows you some of the implementation partners who have already said they are going to adopt this framework and integrate it into what they’re doing. We were lucky that US News and World Report of course adopted the initial framework and now they are looking at the adaptations to add some of these questions.
We did do analysis of hey, for the things that came out of the Delphi process was the feasibility and if there was something else that was already being used by US News or City Health Dashboard or County Health Rankings we correlated it to those to make feasibility easier so that we’d have the least number of new measures that needed to be collected or developed.
But American Heart Association, National Council on Aging here which represents employers, health systems have already begun to adopt and integrate. States have been early adopters, so are Delaware, New York, California, Massachusetts was the other one that there’s legislation on the table now.
Federal agencies have integrated some of these into their surveys or funded testing of these measures with their groups such as Administration for Community Living with the National Councils on Aging, which is integrating these measures into 1000 senior centers for testing in the next year. And we have already tested with 20,000 people in Baltimore and that’s what I mean by people get motivated and actually get busy doing the work.
And then public health agencies such as ASTHO, funders, Well-being Legacy Partners and then technology groups like Community Commons, LiveStories are building tools to make it easier. And if we have time I’d love to share one of those early tools. And then, of course, we’re adopting in 100 Million Healthier Lives along with the 1000s of partners we represent.
These are some use cases. But I actually wanted to pause – this was just how are different groups using it and they will be available in your slides. But we’d love your help in answering does this framework evolution feel consistent? Are you comfortable with the structure and the new name?
Should we include measures for all domains and subdomains? And for this question, Vickie did a wonderful job actually pointing out, hey, here are these domains. Why are all these subdomains not showing up? And what we did was in the Delphi process, we sort of kept it to which ones really emerged and all domains were covered but not all subdomains and in yellow are the ones that are the subdomains that didn’t emerge in the Delphi process.
But they are all in the expanded set across all domains and subdomains. However, if you guide us to say actually even though they didn’t make it in the Delphi process, we want every domain and subdomain represented in the leading indicators, we would be more than happy to just take them out of the developmental set into the leading indicator set. And so that’s a point of guidance that we would love to have from you.
And then if you look over to this side, in framework four, there was a lovely table that showed the evolution of the framework over time and this just sort of shows you what’s happened. And if you notice, in this first column, the well-being of people and life expectancy was there. We’ve just moved it back in now as part of the wellbeing of people.
The wellbeing of equity is there and we’ve sort of indicated the measures related to that which are actually in some cases part of the demographics in others and then the wellbeing of places is a theme name but it’s really the original framework that was handed to us in framework four in its form. So we didn’t get rid of any of those framework elements, we just brought back something from the past and then put that pin on equity.
So questions? Comments? Thoughts?
CORNELIUS: Thank you. Wonderful, wonderful, hard work. Very robust framework. I have a practical question. So this is an iterative and dynamic framework. And so here we have an iteration that’s in our hands today and it’s going to build up and build up. So the practical question for like the larger public out there is when would something like this appear up there on your website and then as I hear the energy – people are using it – so would this be version X and then there will be version Y and so on and is there like a timeline tied to that?
STOUT: That’s a great question. Let me show you. So this is a website that LiveStories is building to support this because in reality what you need is something that anybody anywhere can use and you can see it’s just visual and engaging. Right now, they’ve only populated it for the demographics. We’re waiting for some of these decisions to be made so that we can then get the actual measures for each one.
But if we can load this up what it allows people to do is drill down and look at the measures at the national level. It explains what the measures are but allows them to drill down with different lenses at the county level, at the state level, the county level but also a metropolitan service areas and others. So this is an example of describing – and what’s nice is it’s a tool that makes it easy for people to sort of know what different things are. And you can see where there are placeholders along the way. But when you hover on something you see what it is.
The adoption didn’t come because of the platform. This hasn’t been shown to any – you’re the first group that’s actually seeing the platform. But this is what we’re beginning to build out.
So what we would do at the end of this – this just gives you sort of for Veterans, for people who are service connected, et cetera. As you can see, you can choose a state and build out.
So our goal would be to keep the core measures and the leading indicators stable at least for a year and probably not change as much of the core measures to the extent that we humanly can. And what I mean by that is County Health Rankings and Roadmaps, US News, and they revise too. That is appropriate based on their learning. It will evolve as those evolve.
Annually, we will have a process of review and elevating measures up or demoting measures I suppose and then that will then update here. What’s nice is the development measures will be updated live so that can get adopted at any time. There’s this great report that’s coming out so some workgroups now are adapting this for children so we have lovely life course measures already. That was something we really paid attention to as we did this. And a child workgroup and an older adult workgroup is making recommendations.
And we knew the Cargill Foundation and the SCAN Foundation had measurement scanning efforts already. We just learned yesterday that Cargill actually decided to recommend our approach as their tier one approach so that makes it a little bit easier but there are other measures around social isolation and social connection that we know will get added because of that. So those can be added live and be available to people live as we do it.
And then that other platform this will get connect to is one on the Community Commons called Measure What Matters so that someone can take a measure, pull it right in, look at what their background data is and begin to record data for improvement and have all the tools necessary to be able to drive improvement and to be able to show that improvement in a place-based way across sectors.
COHEN: Soma, could you go back to the slides and the questions for the Committee. I think if folks want to respond to these now that would be wonderful or if folks need a little more time to read the report and then think about a response, that would –
STEAD: From just a process point of view, I think we might see if one of the tents are up are questions or comments and then come back to this slide and what you’re trying to next if that’s okay. But let’s see if we can deal with the questions over the next five or so minutes so that we’d have enough time to really come back to this slide with the last seven to 10 minutes of this block. Vickie, then Denise.
MAYS: I have to agree with Lee here in terms of the hard-working people. I’ve been to some of the meetings and I thought we worked hard but they do. So in terms of the amount of input that you get, it’s absolutely wonderful.
I want to ask two questions. One of the questions is about one of the things that communities often don’t have access to, because of how we do articles and blurbs and all that, is the actual information that would guide them in using this. So I’m going to give you an example. There is what we call response sets, different groups will answer questions in different ways. And so if you were to take some of the well-being, which what you’re talking about is actually subjective well-being versus objective well-being, a community might not know that it’s measuring it but that their group has a tendency to answer better than their actual health statuses.
So I’m wondering where there is evidence of you should modify or you should think about this, whether or not you’re going to have those somewhere so that the community can say, ah, if I’m doing this in a predominately Latino community and I’m worried about how people are answering, I should then kind of modify this in some way. So there is literature about several of the questions that you’re asking that say there are certain modifications that are needed.
STOUT: Thank you for bringing that point. So one of the partners in this effort is Kaiser Permanente. They did a nationwide survey across all of their markets looking at specifically this question. And they are in the process of looking at are there adjustments for different racial and ethnic groups, for instance, or cultural groups that need to be made.
But the other piece to it is actually that there’s been extensive testing of those people-reported questions in different communities where really the intent is for those communities to be looking at their own data and to be balancing subjective well-being responses with things like life expectancy that are quite objective so you have the balance of the subjective and the objective together in that domain. And I think that is really helpful for people to begin to come together to have a conversation and to identify places for improvement.
So it turns out that when we look at that, and people are in a community-engaged process, it’s been an incredibly useful tool to say, even though they may seem like they are responding well, if the average response on the ladder is seven and our response is five, that’s actually – it may be better than you would think based on where things are but it’s actually substantially less than baseline. So helping people to be able to see where they are in a benchmark way as well as looking at what might be adjustments that are made for groups is exactly the kind of research we’re doing but it needs to be done with testing in communities.
MAYS: Well, some of this we have some research data. So I guess I’m asking are there going to be guidance on the website that is actually the research results that say – not just a – but here are five things broken down for people to be able to use to guide them.
STOUT: Yes. You saw on the website, each of the website blocks has text as well as links to resources, so that’s the kind of resource that we would want to put on there. And we would love any recommendations you have about specific pieces that you think would be important to include. Along with that we’ll create implementation guides around what we’re seeing emerging as common use cases already from the early implementation. So that we can actually give different groups who are using it in a particular way guidance about what to do and how to do it.
LOVE: Thank you for work. This might be a solution for all those communities who are out there scavenger hunting for measures that are relevant for their communities so I commend you for the work. And you may have addressed some of this but my processing is slow this morning so I apologize, but on the granularity of these measures, I mean some of these cannot be drilled down at the community level, right?
STOUT: Such a huge number of them can.
LOVE: Okay. Is that part of the criteria for selection?
STOUT: Yes, very much.
LOVE: And then for developmental measures, how do those come about? Because I just noticed because I know a little – I know a lot about a little and one of them is health. But I notice for admissions for hospital for preventable, it’s CMS but most communities I work with don’t really – the elderly is not the big problem. And so we have 48 states with hospital data that are just like those CMS datasets. So how would they get into a community measure set for preventable hospitalizations and ED visits?
STOUT: So this framework is being used at multiple levels. And I think that’s part of what you see, is the diversity of some of this. So states are using this in some cases to look – Medicaid, for instance, to look at how things are going.
LOVE: So it’s pretty dataset independent then. They could take the measure and then do their own scan and populate it. And so my job is to let people know that these datasets exist because some community initiatives that I work with don’t know.
STOUT: And this would make it easy for them to aid – instead of like making up measures which is what they do because they don’t know. We had initially created Measure What Matters as a way of at least creating – like helping them identify validated measures and use it, but this would allow them to actually get to those datasets.
LOVE: So in the website, I mean in your implementation guides, maybe give some pointers of who to contact for these types of datasets and there’s others but I’m just talking about hospital at this time. That’s good to know. I’ll watch that and be in touch. Thank you.
STEAD: Nick and then we’ll go back to Bruce.
COUSSOULE: Just a quick one, and you may have answered a little bit of it before. But kind of being in operations guy, one of the challenges I see is lots of really good stuff happening. How do you communicate not only the value of the framework but actual implementations and what people have done to move the needle on that to get more and more people to recognize the value of where they start from? I think you talked about doing some either use cases or publishing some stories. I’m just curious how you can tie in this community has now taken these kind of actions to improve this kind of measure.
STOUT: I feel like you guys keep setting me up. Some of this has to do with – so in 100 Million Lives we have monthly calls that go out to communities and communications that go out to the communities. There’s a change library that communities populate with stories and data that show how they are creating change.
And so part of what we’re beginning to do is build some of that narrative of how are communities using it and what are the changes that are happening as a result of this. So that’s something that we probably put out on a weekly basis in some form, and on a monthly basis where there’s a webinar for people to join. Frankly, there’s probably at least one webinar a week if not more, for people to be able to join and share stories about how they are making these things work.
And so the thing to know is this is part of collected ecosystem and there’s – beyond 100 Million are integration into other data platforms like Community Commons that’s really signed on so that this just becomes part of their core base of what they have on their front page or County Health Rankings. US News and World Report that’s writing stories about this. It’s really creating what we call that measurement and improvement ecosystem for well-being.
And so as part of our next steps, in addition to seeing how we might integrate with other aligned initiatives, like Healthy People 2030, NASDOH, SIREN and others and assuring that data availability, access and use were convening stakeholders that can serve as these nodes who can help lots of others in the field. Because no matter what, a network only reaches a particular group of people but when you connect networks together and begin to amplify the stories but also do what we’re doing which is a federated search across all our platforms so that no matter where you are, you don’t have to know to go here, you can actually search and find those no matter who has created them. That’s where we begin to really amplify not just the use but the stories of how you make this work that then I think supports further adoption.
PHILLIPS: Soma, the face validity of this and the effort that went into it to make sure it had face validity is incredible. I am struggling because as the Federal Data Strategy and the evidence-based policymaking law come together, there’s no clear receptor site for this. The fact that you’re getting this into some state legislation or efforts is incredible and that may be where this needs to move next.
But a number of the items in this list are dependent on those federal data sources and I just wonder where you see – how do we convince that Federal Data Strategy group to help populate this. How do we get enough call from states or from stakeholders to say, look, we really need you to have these as primary outputs for this purpose even if you don’t own the process of how this gets maintained?
STOUT: Like with any other engagement strategy that we do, we would first want to meet with that group to understand deeply where they’re coming from and what they are trying to accomplish, how this might be helpful to them and who they see as our core stakeholders that they need to hear from. And then we would mobilize those stakeholders with them in a conversation.
KLOSS: This may be something that Pop Health has contemplated but is there an opportunity to use our Committee status as a federal advisory committee to more formally present and argue for why this needs to be considered as part of the evolving federal – is it time to go back to the well with another recommendation.
COHEN: I’m happy to address my feelings about this. We wanted to see where this process was at this point in time so that we could essentially engage in this discussion and if we felt there was an opportunity for the Population Health Subcommittee as part of the full national Committee to get more involved in something that’s in our bailiwick to help promote these activities, we could build that in to our future work plan.
So you’re absolutely right, Linda, having seen this presentation and as Bob said, the emerging Federal Data Strategy and other kinds of activities happening at the federal government and in HHS, I think there are clear convergences of interest that would make us think about what we could do from our perch to move forward with some of this work. Bill, you might want to address that as well.
STEAD: I, for one, as does the rest of the Committee, love what is happening and particularly love the viral update and the real validity behind which this is being developed. I probably should just be quiet and not try to have perfect get in the way of the good. The question I raised on the PopHealth call I still think is important and I’d love to figure out a way to solve for something that works in that space. And it really is this distinction between well-being and equity as in essence summative outcomes.
And you’ve done an extraordinarily good job of relaying the story of why people need to see them in the domains so that they can see the things they care about are there. And I get that. I do think there’s importance in, if you will, the structure of how things relate in branching logic.
And so in much the same way as we tried to have domains which we frankly did, we wanted the domains so that every cabinet level agency could see themselves in those domains with subdomains they recognize as in their current goal set so that we in essence could help them begin to see how these could come together for collective impact.
And the collective impact is the well-being – so I don’t know – I begin to see some of this language in the pieces. So I think it would be helpful if, without confusing the responsiveness to what people want, because that’s the most important thing. My caring about informatic structure is, at the end of the day, less important except as its reflected in the wiring diagram.
But I would love a way to help people see that – I’m wondering outcomes, maybe these three outcomes, then domains and then subdomains. I could almost see that, everybody could still see they can work at an outcome level, at a main level or subdomain level with their own set. So if it’s not destructive, I still would love to see that clarified in some way but I don’t want to be destructive in any way of the progress you’re making.
STOUT: I don’t think that is destructive. Actually, we think very similarly about it and have done some that mapping and would be happy to – so we can think about how it’s communicated and we can think about what’s an overall sort of index approach, if that makes sense. I think on the well-being of places we already have that. But for the well-being of people you could potentially use the well-being adjusted life here or value as something that’s added to that in some way that then leads to something more and that has an equity differentiation with demographics.
COHEN: Just a second before we get to you, Vickie. I think if you look at how WIN has adapted the framework, they still keep the domain structure consistent with our intention of having the domains being recognized by the different executive secretariats. What they’ve done is built on top of that these three overarching – the beauty of this approach is communities, the federal government, agencies, programs can identify the particular domains and subdomains of interest without having to use the entire framework structure if that’s really their priority and focus.
So I don’t see a problem in the expansion. I just see that as additional opportunities and I think it’s kind of been proven out because the VA has already adopted looking at this framework what’s relevant to them. So I don’t have that issue that I think – I don’t think we’re trying to shove people into those three core indicators but the leading indicators allow the kind of activity that we initially had envisioned at the federal parsimonious indicator level. So that’s how I would respond. Sorry, I didn’t mean to intervene, Vickie.
- MAYS: I think in terms of Linda’s suggestion, the thing to think about is kind of where we would want to put this. I think at the federal level it may not be the right level for us to do the advocating. It’s too bad Susan Queen isn’t here, because the bottom’s very high in terms of anything you put into a survey, anything you put into an agency. Susan can really probably attest to this in terms of Healthy People Initiatives, having done one of them for the leading indicators for 2010. We struggled with well-being. We could not use the Gallup Poll, for example, because of the lack of actual evidence-based data that could support it.
There’s actually other measures that we could use so I think thinking about the states are very different. I think thinking about I think it’s called a national state legislator group. There are a couple of state groups and there’s a state survey – NCS, yes. And then there are state surveys.
I think that we would probably be better off working at those levels where they have a different bar because our bar at the federal level has just risen higher because of the law that was passed about the evidence-based policy group. So I think it would be harder, for example, to tell Healthy People that they had to use some of this. They just won’t be able to.
STEAD: We will bring this to a close in about five minutes.
COHEN: Before I express my opinion, I’d love to hear from some of the other committee members how they would respond to these questions unless you just really haven’t had enough time to think about that. And we can have that discussion as feedback from the Population Health Subcommittee or the Executive Committee.
LANDEN: I’ll kick off. Might as well answer the questions in the order they are there. Does the framework evolution feel consistent? Yes, I think it’s right on target to do any kind of work that we envision when we have the resources to develop the framework.
Are you comfortable with the structure and the new name? I think it’s courteous of you to ask but we’ve kind of turned the product over to you. It’s not our call anymore.
Third, should we include measures for all domains and subdomains? I’ll defer to the PopHealth Subcommittee colleagues on that one.
What recommendations do you have to guide us as we advance into implementation? On that I’d like to be sure that both you and – that built in to the process is the collecting of the feedback from communities who use the framework to design and to launch improvement initiatives that we were talking about a few minutes ago. What did those initiatives look like? Were they successful, unsuccessful? What were the results? Did the role of the framework components actually contribute well or not so well to the initiative and its design’s success or failure?
And then importantly, please come back to us with any – as you learn – come back to us with any requests for how we might be able to remove barrier that you identify, provided those barriers are within the scope of our mission.
STEAD: Other comments or answers? That was a fairly summative answer.
HINES: Quick one, Bill. So Soma, you and I have been talking and I know you are knee deep in some pretty serious implementation work and I just want to make sure you leave here with what you need so that – I think there’s been a clear message there’s no interest in slowing things down on the Committee’s part.
So I just want to make sure you are going to leave here, if there’s anything else you need to raise so that as you are designing the database that’s going to drive that website, do you have what you need from us to move forward? And the PopHealth Subcommittee will be meeting in the next few weeks and we will obviously continue to coordinate with you, but just in terms of today, what else do you need from us to keep the train moving?
STOUT: So definitely a decision about should all measures that for domains and subdomains be included in the leading indicators? That’s crucial to how we do the website implementation. We’ll be including evidence updates on the Gallup and other studies so those are things that will work out over time.
But I think for us the endorsement of this full Committee matters because although you handed it to us, we really see this as something we are doing in service to you. And so that matters at that level in addition to just what it means to support and to know that this group has blessed it, I think will actually support implementation and adoption in a wide variety of other ways even in and of itself.
STEAD: What I would suggest from my perch just as one person – since we’ve got this concept of a parsimonious set and an expandable menu, it would seem to me that if there is something that’s valid for the subdomains that didn’t come out prioritized with the Delphi process, I would have them there. If they’re not valid, I would have them in the developmental piece. Either way they would be part of the ecosystem and people could use them or work them wherever they want to and they could be promoted or demoted.
I think the way you are envisioning that in totally different space is actually what we’re envisioning in the Predictability Roadmap about how the standards would work – up from use cases, and promotion, not down from regulation. And so I think – does anybody disagree with that general statement because I think that would give you enough to run with without us being prescriptive. Is that right?
STOUT: That is. Thank you.
STEAD: Is everybody comfortable with that?
COHEN: A couple more before we close. Again, as we mentioned, I like the Population Health Subcommittee to review the final draft and get feedback to you so that we can perhaps finalize our input and give it to the Executive Committee who might want to make a statement of full Committee endorsement of this particular product. I think you’ve got the sense that there’s general consensus that we certainly fully endorse the process and for me, it’s incredibly humbling that this framework that we started out with has emerged and developed into this robust measurement system with the incredible input from so many people and organizations and such rapid uptake already. In my wildest dreams when we began our framework activity, I had hopes but certainly no expectations that it would evolve and that our – we get 400 hits a year. This project and the tool that emerged from our initial activity will get 100,000s. So I couldn’t be personally or professionally happier in the work that you’ve done here.
My final thought is I agree with Bill. I don’t care where the subdomains are placed, whether they are in the leading indicators or the developmental just so folks can easily find them and hook into them.
PHILLIPS: Just very briefly, Bill, is a query. To Soma’s suggestion about meeting with the Federal Data Strategy folks, given that we’re not meeting until June – this full group – under our authorities what’s our options of either convening that or for participating in that? Of meeting with them with Soma to talk about the WIN.
HINES: With the Federal Data Strategy people? I think there is no problem with that at all. We are supposed to engage with stakeholders. It’s really just a question of the outreach. As you noted, I don’t think there’s a lot of HHS presence on the Federal Data Strategy leadership so we would need to talk to ASPE about how to negotiate this. But I don’t see a problem at all. It’s what our role is. We’re very stakeholder driven so we have a subcommittee call coming up and I think that should be at the top of the agenda and then we can move forward.
STEAD: Thank you. We’ll bring this particular conversation –
HINES: Can Soma just put up one last slide about an event on April 10th.
HINES: I just want to make sure everybody’s aware.
STOUT: You’re all invited. On April 10th, at the Wellbeing Trust in Oakland will be hosting us for a meeting of key stakeholders that want to advance this into implementation across federal and nonfederal agencies. There will be a virtual option to join for those who are on the East Coast. All our meetings so far have been held in DC so we thought we’ll have one on the West Coast to make things easier for people.
But it’s really going to be focused on we assure – like, what’s the strategy for assuring data availability and access for people, including commitments for funding for that data collection if additional measures, for instance, are needed, whether that’s at the federal level or whether that’s funded through other means, such as foundations making a 25-year commitment as part of their grantmaking commitments.
How we make adoption easier. And then really creating a set of stewards that are going to be part of that learning journey for the long-run that are going to commit to supporting this framework, learning from it and then evolving it over time. So that’s the goal. We have some great people already committed to coming and we’d love to have you to be there.
KLOSS: Just a quick question. Could you send out the way we might link in virtually when you get it?
COHEN: And I had just one final thought. I neglected to acknowledge and recognize Kate Brett, who is not here, and who was lead staff for the Population Health Subcommittee who put in an incredible amount of work in helping converge our framework with WIN and she’s primarily responsible for the summary table and other activities.
STOUT: And I would be remiss if I didn’t acknowledge Marianne McPherson and Julia Nagy who have done lion’s share work behind the scenes in facilitating the process. And Rebecca who has also been such a guide.
STEAD: Thank you. Will you introduce –
HINES: Yes. So on the same topic but a different lens, I’m really delighted to introduce Carter Blakey. Carter, thank you so much. You have been involved with Healthy People for decades now I think, maybe two. And is with the HHS Office of the Assistant Secretary Office of Disease Prevention and Health Promotion. I know you have several titles which you’ve helpfully put on your slide, Acting Deputy Assistant Secretary for Health of Regional Operations and Deputy Director of ODPHP. So without further ado, Carter, thank you so much for joining the conversation this morning.
Agenda Item: Healthy People 2030: Federal Role in Supporting State, Local, and Community Measurement
- BLAKEY: Thank you for letting me come to give you an update on Healthy People 2030. I think the work that Soma presented is really exciting and hopefully there’s a way that we can work together at some point.
So I imagine most of you are familiar with Healthy People. It’s been around since the 70s when, I guess it was Julius Richmond first put out a report on Healthy People which grew into over the decades a set of objectives, national objectives with 10-year targets that the Department of Health and Human Services puts out every 10 years. We’re currently in the fourth iteration, Healthy People 2020 and working on the fifth iteration. One of the nice things about Health People is that it has spanned the decades and it has spanned administrations. So no matter who happens to be at the helm, we’ve always had a lot of support to move forward.
So this decade, if you’re really familiar with Healthy People you’ll know that we started out in 1980 with our first set of objectives that had a grand total of about 200, a little more objectives. And then if you fast-forward to where we are in this fourth decade, we have more than 1200 objectives which has proved to be unwieldy in many ways. For our stakeholders who would like someone in the federal government to do the job of calling those measures and say here are the ones that are most important. This is what you should focus on.
And then also resources, just the expense of maintaining something that’s that large. The federal data systems underpinned everything we do. Data collection is expensive. I think every time you add a question to a survey it can come close to – not far from a million dollars just for one question so it’s not cheap, as you all know.
We have heard from our stakeholders that while they want us to shrink in size, they also want to make sure that they can see themselves, and that’s always proven to be a challenge. We tried to cut last decade. We thought we had a plan in place but we kind of dropped the ball in terms of communicating before we went out with our plan and we hadn’t paved the way to help people understand the importance of getting smaller.
So as a result we had our advisory committee last decade actually recommend to the Secretary that we not reduce because they were concerned about breaking data collection trends and also who else at the federal government has the resources to maintain that number of objectives and measures and provide the data but it’s just proven too much to maintain.
And we have had full support of the past administration who actually had us start the development of Healthy People 2030 a year early. So I think it was Karen DeSalvo, who was the Acting Assistant Secretary for Health at the time, led the charge. We convened our advisory committee which is still in place and the charge to the advisory committee was to provide recommendations to the Secretary of HHS on how to reduce the number of objectives in Healthy People by at least half of what is in Healthy People 2020. So we’re well on our way to that.
So the development of Healthy People 2030 mirrors what we’ve done in the past. We have an advisory committee which is made up of outside experts. It’s co-chaired by Nico Pronk from Health Partners and Dushanka Kleinman from the University of Maryland. There’s a total of 13 members on that Committee that have met I think 11 times since they were sworn in in December 2016.
Then within the federal government we have two groups. We have the federal interagency workgroup that’s made up of representatives from across the federal government, not just HHS. And then we have what we call the topic area co-leads. Healthy People 2020 has 42 topic areas and there’s a federal workgroup of representatives from across the department that manage all the objectives in Healthy People.
And then public input is always key to what we do. I think that’s one reason why it takes us so long to get the process completed is that every step of the way we have public comment opportunities. We have an open mailbox for public comment throughout the whole development process as well as explicit public comment opportunities for specific elements of Healthy People. So we really feel like we’re stakeholder-driven, not only from outside the federal government but also within the federal agencies as well.
So development of Healthy People 2030 has been proceeding in phases. Our first phase was develop the framework and that was completed in June of 2018 when Secretary Azar signed off on our framework for Healthy People 2030. And we’ve initiated and are deep into the process of objective development. I’ll talk more about that in a minute.
And we plan to launch, in the first quarter of 2020, so sometime between January and March it is our goal to release Healthy People 2030 which is actually almost a year ahead of what we did for 2020. It was the end of the 2010 year that we launched last decade.
So this is our framework. Our vision is similar to what we’ve had in the past – a society in which all people can achieve their full potential for health and well-being across the lifespan. And our mission, we really culled this decade with the help of our advisory committee. Our mission has been a long-winded series of bullets in the past but we have a succinct statement – to promote, strengthen and evaluate the nation’s health efforts to improve the health and well-being of all people.
And then our overarching goals are something that we really focus on with Healthy People. And these have grown in number across the decades. Numbers I’d say – well, the first one, it’s basically achieve health, increase the life expectancy and healthy life for all people and we’ve always focused on disparities.
At the beginning of the Healthy People decades it was simply to reduce disparities but over the decades we’ve changed that to eliminate disparities and even added achieving health equity. And this decade for 2030 the advisory committee wanted to underscore the importance of health literacy, ensuring that people actually understood the opportunities that they have to stay healthy.
Number three addresses the social determinants of health and that was something that was new for the 2020 decade and for the 2030 decade its expanded even more to go beyond just social and physical environments but also the economic environments that people live in.
And of course, number four, looking at across the life stages with healthy development, behaviors and well-being across all stages of life.
And number five is totally new and this was something that the advisory committee felt strong about adding. That was to not simply stop with the lofty goals but also make sure that we have an overarching goal to make sure that the action is taken on Healthy People. So it elevates the need to actually implement and engage folks.
Phase two is looking at the development objectives. This decade for 2030 we’ve broken objectives into three parts. We have what we call the core objectives and those are the objectives similar to what we’ve always had. They have their baseline data and promise of data collection throughout the decade.
We’ll continue to have developmental objectives and those are objectives that are of critical importance and they have the assurance in the very near-term of having data so that we can establish a baseline and then set a target.
And the new category of objectives is what’s called research objectives. And this again was a recommendation from our advisory committee that we put on the table important health issues that may seem logical. We may all think well, this is a key component of improving health and staying healthy but we just don’t have the data, at least not the data that we can use at the national level within the federal government.
And so we want to encourage research in this area and encourage data collection. So these are – the bar is very high to be a core objective, not quite as high to be a developmental objective and then we’re taking just about anything for a research objective too because we don’t want to inhibit further advancements.
So we’ve always had selection criteria for Healthy People objectives. We’ve been very good and we list usually a series of at least nine criteria to be included. But we have not been very good about actually enforcing the use of those criteria. And that has something to do with why we have 1200 measures right now. We tend to cave when you get proposals for objectives. You find you start signing off on a lot of them. But we realize that we really needed to stick to our guns and make sure that every single core objective in Healthy People has valid data behind it and we rely heavily on the National Center for Health Statistics to ensure the data validity of anything in Healthy People.
So one thing that’s different – we’ve always had the data needs. You have to have your valid nationally represented data. We’re requiring at the start of the decade more than just one follow-up data point beyond the baseline but we are requiring at least two which is a difference. We’ve also with any proposal for an objective, we’ve required a description of the national importance. Why is this objective or measure important on a national level?
And we’re also requiring evidence base, that you can actually make a difference in this area. There are a lot of areas where that may be important but we don’t have evidence-based practices to make a difference.
And then of course we look at – we’ve required an assessment of the health-related disparities and health equity in any proposal. So an explanation of how this objective could impact health disparities and health equity.
So the way we got down from 1200 measures to where – actually we are right now at 355. That went out for public comment. We are so happy. I could pinch myself. I cannot believe we actually did it. But it was not an easy process. We started last spring going out to the topic area workgroups and saying here are the criteria that we expect you to adhere to. Tell us which objectives will meet those criteria.
And they came back with the first assessment about little over 700 objectives kind of self-selecting. And then we said, well, that’s still not good enough. We have to get down below that. So we asked them to do another assessment. An alternative would be that we would have to cut for them, someone else would be making the decisions for them. So they really did a good job of looking if you had 70 objectives, what were the top ones that you really needed to focus on to make a difference in your area.
And with that assessment they came back with the 500 and after that we said, fine, you have 500 objectives. They are not all going to make it but we asked them to complete a lengthy memo that touched on all the criteria and come present before the Federal Interagency Workgroup over a series of about two and a half months. And we met that Interagency Workgroup of about at its max about 40 people, met probably 12 times over the summer, sometimes all day, to go through each of the proposals.
And when the workgroups actually presented to us they only came up with about – it was a little over 400 measures that they presented. And some of those were cut back from the 500 to the 400 something, just with the discussions with us, is there a way you can make composite measures. There’s some duplication here, do you really need both of them.
So the Federal Interagency Workgroup accepted 411 objectives by September and then we convened a smaller group of about six representatives from across the government who were not as entrenched in Healthy People as we are that could have a higher-level look at the objectives. So they could say does this balance, what’s missing.
And that group – we told them when we recruited them – and we actually went through the Data Council to try to get good representatives for that group – that they would meet maybe four times between October and November but they actually – our group then again actually met I think we counted 11 times. And we counted up the hours, how many hours they had to spend. But again they rigorously went through each of the memos, each of the objectives and they cut it down 355. They pointed out duplication.
We went back to the workgroups who feel strong ownership of their objectives and said here are the results. And we had very little pushback with them. They understood the reasons for the cutbacks. So with those 355 objectives we went out for public comment in December and that public comment period ended in – it was January 17th and we had 4700-something comments on the objectives.
And some of those – I think about just 600 are proposals to add new objectives but we did – it’s getting back to where we started – require that anyone proposing objective meet the same criteria that the workgroups did. And some of those objectives are nice ideas. They are people and people come up with indicators or measures, they don’t think about is there data to support it. So we have a lot of good ideas but there’s no data to support it.
So where we are right now, we just sent out to all the workgroups last week, all the public comments and they’ll be culling through those and then presenting to the Interagency Workgroup the disposition of each one of those comments which we have to do for FOIA purposes in case someone – there’s a Freedom of Information Act request wanting to see did we actually look at their comments.
So we plan to be done reviewing the public comments. I imagine that will grow a little bit because I suspect that some of the perhaps the workgroups have stakeholders who could advocate for them and suggest that objectives that weren’t included be inserted again. So we have left room – we’re thinking we might accept about 25 proposed new objectives. We have to make sure that when we add an objective to one part it doesn’t upset the balance in other ways. So there’s a lot of thought that has to go in to adding objectives.
Then we’ll have to go up for departmental clearance and that’s another opportunity to grow again because the agencies feel strongly about their measures and that’s really where there’s the chance to grow the most, is during departmental clearance. So that could take quite a while, a couple months, to work through that process. And after that we’ll launch.
To give you a glimpse of where some of the cuts have occurred. So these are the topic areas in Healthy People 2020 that have 50 or more objectives and you can see how the workgroups have cut back to reduce. So we will see if that holds as we go forward. It’s not an easy process.
Then we turned toward implementation and thinking about how to move forward. Our advisory committee has a workgroup or a subcommittee that’s looking strictly at that.
Early on in the advisory committee’s work they did submit recommendations to the Secretary. It was like over a year ago and that had to do with community level data. That was a discussion that they had the very first day that they met, trying to make Healthy People relevant at the community level which I think Soma’s already talked about that, and Vickie as well, is there’s a real struggle between the national level data and what we can use at the national level and what’s available in the community level but we’d like to work on ways to find some comparability in that. So we do hope to have a stronger focus on community level data wherever possible.
What we’re talking about now – there are all sorts of issues that have come up with respect to community level data. So like you all, we want to build partnerships with folks at the state and local levels and help the state and local level folks understand data standards. If there is somehow we can provide guidance to them on what’s high quality data in our eyes, that we would feel like we could link to.
Then also we recognize we need to have a repository for data which we actually do. We have what we call a Healthy People Data 2020 and that’s all the measures for Healthy People 2020 and we will drill down to the state level when we have that data.
Then in terms of engagement we’re looking at how do we engage with the local level. Again, I think working with Soma that might be a good way because you’ve already done a lot of work there. So if there’s a way that we can build on the work that’s already been done that would be a good idea.
And then encouraging the development and adoption of local objectives based on local data and circumstances to support the national objectives. So I guess one difference from what Soma is doing and what we’re doing, we’re looking at the community and local data as a way to bolster the national objectives. So we would be looking down at the local level, what data do you have that we could roll back up to ours rather than – they might have a lot of data and a lot of measures but they might not all be relevant to what we would want to do.
And then communication is a huge effort. Making sure that we communicate what we have, communicate what we’re trying to do and also provide resources to communities so that they can be comparable or work with Healthy People. We would love to reach a point where we could actually go beyond the state level data and even link down to community level data explaining the caveats in comparability.
So the next steps. As I mentioned, we’re going through the public comments. We’re right now establishing targets. What went out for public comment did not have the targets. Target setting is key to Healthy People. There are a lot of indicator sets but not everybody is daring enough to set targets. So we’re working on those 10-year targets. And we’ll go through departmental clearance.
The website will be different for next decade. It sounds not unlike what you’re trying to do and we have, for management purposes, what we call our topic areas and right now it looks like we might have 42 or 43 because we added something around opioids. But that will be primarily visible to the folks who have to manage. So what we’re trying to do in building out the 2030 website is build a website that users can come to and they can develop their topic areas.
If there’s a particular interest they can do a search and pull out whatever objectives that are relevant to them because a lot of the tagging of objectives and the way we manage might be subjective. What’s related to health to one person is not the same thing as what’s relevant to another. So that’s something that we’re working on.
We’ve also engaged the National Academy of Sciences to develop the leading health indicators for Healthy People 2030. That’s something that’s just barely getting going. I think they put out for public comment is the list of panel members that they’ve selected. I think the Academy’s put that out for public viewing. So those meetings should be starting up fairly soon in the spring.
And then just the launch of Healthy People so that’s where we are. If you have any questions, I’m happy to answer them. I think there’s lots of opportunities to work together.
- MAYS: I am so excited. I can’t believe it. You’re going to be the Mother Theresa of Data in terms of Healthy People because that’s an incredible task that you’ve got it to 355.
What I’m really excited about is the research category that you all are using now because it’s really designed, as I understand it, to get everybody thinking about what kind of data do we need. And you’ve said this is an important issue but we don’t have the right data for it.
So I guess I’m wondering a couple of things. Is whether or not there’s any partnership or the department is going to talk with like NIH, AHRQ, agencies that can take this on as kind of maybe an agenda to start to fund that kind of research. Particularly within that also to fund – and this is something that’s come up with this committee several times – is expertise in terms of either small area data or small populations because you have a population issue as well.
And then my last question would be the data as I understand it that you’re going to use for racial and ethnic minorities is still going to be the aggregated data so it will say blacks, it will say Latinos, and that I think, given what we know now about foreign-born, US-born nativity differences within groups seems like it just doesn’t fit with all the innovation that you’re doing.
BLAKEY: I’ll address the first one. Anyone who has worked with us knows that we have no money so we’re really good at partnering and passing the hat to get support from other agencies and that’s how a lot of our budget is – that’s how we survive.
And an impetus or incentive for agencies to want to pursue these research objectives is that there’s a real desire to be a core objective and if they don’t have the data, they don’t have the evidence, they are going to need to get that if they want to be elevated to a core objective. So that’s just something that we do. We do always try to partner and find funds elsewhere.
And then in terms of the demographic data, the breakout data for the objectives, it will be still split in the same way we always have unless the NCHS can come up with a different way. Typically we follow the OMB guidance for all of that. The advisory committee has – one thing I didn’t mention – the advisory committee did at the same time that group of six looked at the objectives, they actually have had a subcommittee looking at all the objectives.
They haven’t presented the results. They will on February 26th, so we can see what they want us to work on and I know that equity is one of the areas. They’d like to see that raised up more. They were supposed to present January 30th but because of the furlough they couldn’t and so we weren’t allowed to publish our Federal Register Notice. We weren’t furloughed but the Federal Register was.
So we could never get the public notice out. So we had to cancel that meeting. So we kind of wish we had the information right now but I know there’s been a lot of concern on their part that just by having a focus on disparities with the breakout, the demographic data, they don’t feel is enough.
MAYS: So are you saying a comment from us is helpful when you are saying something –
BLAKEY: Yes, that’s fine. We have the public comment on the specific objectives closed but, we do have – the advisory committee has an open mailbox. We have an open mailbox – Healthy People is firstname.lastname@example.org – and we take comments on an ongoing basis there.
STEAD: Linda, then Bruce and we’ll go to break.
KLOSS: Thank you very much for the update. I just went quickly to the proposed objectives list and I just had a question about what shape a target might take. A number of the objectives seem like increase or decrease. Will we kind of know what the desirable target is or the starting point or how will that –
BLAKEY: Each objective of the core objectives will have the baseline. Those 355 each has a baseline but we didn’t put those out. Each of the 355 already has a baseline and then we have different formulas, methodologies for the workgroups to use to set their targets. Sometimes there’s existing policy – there’s a law – so you follow what the policy is. You have to accept that.
There’s trend analysis and modeling if they can project out and get a sense of where they will be in 10 years and we ask them to go a little bit above that. So we want targets to be – a stretch target but not totally unrealistic. We also – we are required to minimally statistically significant when there’s a lack of the trend data. So there will be a stretch. But in some cases it will be just maintain. Like with the vaccinations, sometimes it’s maintain.
COHEN: Thanks. Great presentation. I am excited about your desire to get drilled down into at the state and community level and I wondered if you’ve developed an organizational strategy to connect with community data collectors and what that process will look like and how folks and organizations could get more involved.
BLAKEY: No, we have not done that yet. But that’s on our list with all the issues that we’ve been considering. And I think, Soma, we might be working with you.
STEAD: Thank you. Rebecca, one final comment and then we’ll go to break.
HINES: I just want to clarify – Susan Queen of the National Center for Health Statistics has been listening in and there are representatives following up on the comment after Soma’s presentation of HHS on the Federal Data Strategy Workgroups. So obviously they’ve parsed out their work into workgroups and there are representatives, officials from NIH, CDC, the Office of Minority Health and the HHS Office of the Assistant Secretary for Preparedness and Response. So in the actual implementation there are HHS representatives. So I just wanted to correct the record on that and I will follow up.
STEAD: Thank you both, and we’ll go to break. We really appreciate the great work and updates.
Agenda Item: Essential Role of Vital Registration Data and Statistics in Community Measurement
STEAD: It is time to rock and roll. Bruce, will you introduce our next speaker.
COHEN: I’m happy to. My old friend and colleague, Steve Schwartz – excuse me, my experienced friend, Steve Schwartz. Steve and I go back to NAPHSIS days when we were both working at states. And we’ve moved on to other jobs and I’m so excited that Steve is now the head of DVS following the incomparable Del Mackinson and we asked Steve to come here today to update us on where DVS is and to talk briefly about his ideas of how the National Committee can help support his efforts at NCHS. Steve.
SCHWARTZ: Thank you for the introduction. And thank you for inviting me here today. I’m delighted to be here in my new role as the new Division Director of Vital Statistics at NCHS.
I’ll give you a little background. I am new to DVS and NCHS but not new to vitals. I started out as the registrar from the great state of New York City. So New York City is one of the 57 vital registration jurisdictions and some would say one of the important ones. We actually are all important and we’re going to be talking about state data, how we get it and how well we are actually doing these days in the National Vital Statistics System.
In my old life in New York City when somebody asked me what I did for a living, at first I would say vital statistics and people would yawn. So instead I realized that what I should say was something like I’m responsible for life and death in New York City and that I sign the birth and death certificates. So that was too delicious to let go of and then sometimes, if I just said I sign the certificates, one time somebody asked me, do you have anybody to help you? And I said, no, the city is so backwards and it’s like I have to take work home with me and it’s even worse on weekends because the work piles up and on holidays I’m still having to sign certificates. So I say that because there are – another way people ask me what I did for a living and sometimes I just said I’m in retail.
So what does that mean in this context? In this context, there are two sides to vital statistics and they are really important in the United States. The retail side is serving customers, issuing certified copies, collecting revenue, charging for certificates, adding a child’s name, correcting a spelling of a name, adding a father’s name. So that retail side in every vital statistics office in the United States, that’s actually the bulk of the staff. That may be a little scary so the other – and I say that because it takes so many essentially clerical staff to be processing the records so since I know in New York City where we had 125,000 births a year we were automatically issuing 125,000, mailing out 125,000 certificates. We used to lick stamps and we got over that part.
It’s important to understand that because the revenue and actually tracking revenue and being careful about how you handle either cash sales or credit cards, I think all vital records offices in the US now use the internet for procurement of certificates. I think New York City had something like 110 or 120,000 internet orders a year. So it’s big business and there’s a lot of focus on that. So that’s what I call the retail side.
The other side of the business is the wholesale side. The wholesale side is public health data. And actually when I first – and I may spend a few moments on this because when I first started working in vital statistics in New York City, vital records, the retail side and vital statistics, the wholesale side, were separated. And they had put vital records certificate issuing part in with the environmental health permit bureau because they handled money. And so what they didn’t realize was that they were separating the source of the data, the certificates, from the public health data. So that was one of the first things that I did when I got to New York City and as a registrar there, was we combined them again.
But it is a challenge, and it’s a challenge because throughout the US there is this – it’s not necessarily attention but most of the staffing resources, and they are the lower paid staff, they are the ones who are actually processing the records. And of course, the data, all the data in the US come from states and the state data come from data providers and the data providers are the hospitals, the funeral directors, medical examiners and coroners.
And the system is only as good as how well everybody is working well together. So that working well together is how well does NCHS communicate to the states and the states communicating to their data providers, working with hospitals and funeral directors. Because say for funeral directors, they are the ones who provide the race and ancestry data on death certificate. So the retail side and the wholesale side are incredibly important.
And, by the way, I wanted to – at each of your places we gave you one of a party favor, three actually. This is the major cause that we have – the major causes of death in the US in 2016, birth data and also death data by race and ancestry. And these are 2016 which is relatively recent. In a couple of months we’re going to be putting out the 2017 ones and that’s an example of just how well the system is operating right now.
And we’re going to go through in this presentation what we’re doing to improve the National Vital Statistics System, how we’re doing it, the resources that we’ve been able to use and how quickly the National Vital Statistics System has changed from used to be a pretty sleepy system to a really vibrant National Vital Statistics System that is producing data for all of us in near real time.
So going back just about seven years we only had 10 percent of the records coming in within 10 days of the date of death. And now, we’re up to 60 percent. And that’s only in seven years. It’s a huge change and it creates a lot of opportunities for using data in real time.
So here’s another look at that. Percent of mortality records received within 10 days of the date of death by number of jurisdictions. We have 80 percent of deaths records received within 10 days of the date of death by number of jurisdictions. So states are getting the data in faster, getting the data in faster means everybody can use the data faster.
It is becoming more of a surveillance system. So for those of you who have been looking at vital statistics data for a while, this might come as a surprise to people that we were able to put out final 2017 birth data within essentially a few months after 2017 by August of 2018. And for mortality data, the 2017 final mortality data came out in November of 2018, so less than a year later.
Here are some examples – I won’t go through them but selected birth reports that we put out in 2018 with very recent data and mortality data in 2018. We’ve developed reports that we call Vital Statistics Rapid Release Reports. So we used to be squeamish about sending provisional data out and now we’re getting it out there and we’re getting it out quickly. So quarterly provisional estimates on the natality.
We have those currently available through the third quarter in 2018. Mortality is a little slower. Here’s another example of the mean percentage of death certificate records available for analysis in 13, 26 and 39 weeks with thresholds for selected causes of death. You’ll see that, for example, that the ones that come in slowest are those that are essentially medical examiner coroner cases because they require more investigation. Medical examiner coroners are a challenge throughout the US because there are medical examiners, there are coroners, there are elected sheriffs who certify. There are coroner’s physicians.
It is a real challenge to be able to get death data in quickly and it’s something that we’re working on with funding from AHRQ and PCOR to be able to get data in faster and helping to push medical examiners and coroners to get the data in faster. And we’ll show you some of those results.
So here’s monthly provisional drug overdose death counts and these are quite recent, going through July of 2018. So these are all helping us to track the opioid epidemic. We are also using a lot of tools to actually get the quality of the data and also the details of the data from medical examiners and coroners much faster and with more accuracy.
So what are some of the enhancements? Improving systems to automate more of the report production. We are now using machine learning for coding causes of death and we’re doing it in real time. In the old days, every state had nosologists and were doing hand coding of death certificates. Now, most of the nosologists have moved on or they have simply retired. So almost all of the death certificates now in the US are coded very rapidly by NCHS staff and generally are reported back to the states within half an hour of receipt of the data file. Nobody would have dreamed of that years ago. So that enables us to get the data in faster, analyze it faster and get the data out to users faster.
The automated coding systems that we used to have had a – one coding system had a throughput of about 80 percent. We’re now pushing 90 to almost 95 percent throughput with assisted machine learning automated processing. It enables us to get more details faster, getting the provisional indicators out faster, more geographic detail, and additional demographic information.
As I mentioned, medical examiners and coroners, this has been a huge efforts supported by AHRQ to enable us to learn more about how to work with medical examiners and coroners, the National Association of Medical Examiners to get information more quickly and understand how the data come in and how these systems can be used more efficiently. In many jurisdictions the medical examiners and the coroners are not using the same systems and some of the challenges are some states don’t even have – say they will have –
Well, let’s say a state that rhymes with Kansas. So Kansas does not even have an Association of Medical Examiners or Coroners. So they essentially act independently. So there’s a challenge there of how – and many of them are elected – so how does do you get a system that is more 100 years old moving faster than that.
Well, we’re working on that, not by ourselves but along with a lot of partners, to work with states and with medical examiners and coroners and we are also on a steep learning curve ourselves on this, to figure out how we can get this information in quickly and how systems that are otherwise unconnected. How do we get that information in quickly and using new techniques for interoperability for getting not only the medical examiners and coroner’s data in but how do we even get data in from state systems into our systems more quickly and sharing data faster when each system was essentially hand-built and each has their own requirements.
And that also goes back to the retail and wholesale side of the business because a lot of the work in that retail side of the business is the complexities of how you just process all of the either how they come in, how the records are handled and the difficulties of using the data that come in on the retail side. Because most states have their own unique systems or they are required, for example, by states that they will have certain laws saying you have to process your revenue in a certain way and the systems that vital records offices may be procuring don’t work with that accounting system.
So that tension will always be there but the challenge is how do we have systems that can interoperate together so that we can get that information into all of our systems more quickly and in a standardized way. And that’s where interoperability is so very important so that these either homegrown systems or systems that were built by maybe a dozen different vendors around the US, they are all using different systems or modifications of a system to meet their needs.
So we’re working with the National Association of Medical Examiners. We’re working on creating training materials. We’re using all the resources we can to obtain the data as fast as we can and as completely as we can and also using training tools to help physicians understand why they are even completing death certificates. For those of you who are physicians in the room, you’ll know that nobody goes to medical school to learn how to fill out a death certificate. I don’t know why that is.
But in fact, that’s a huge challenge. And of course the data are only as good as what the physician writes and generally, as I’ve learned in working with the most important powerful people in the world, I’ve learned those are the clerks and those clerks, especially in hospitals, will tell physicians, especially the young physicians, doctor, this is how you have to fill this out and if you don’t fill it out the Board of Health will reject it. How many of you have experienced that?
So that’s a huge challenge. One of the things that we did when we used to have paper death certificates we thought we should have instructions there when there was a paper death certificate. And it made perfect sense. So at one point when we were trying to – in New York City – trying to do a better job of getting higher quality death data on the death certificate.
We brought in a focus group of physicians and we worked with them and we showed them the death certificate which on the front was – it was 100 percent cotton paper because in those days we were saving the paper forever and so the hospitals were really careful about who actually got that expensive 100 percent cotton paper. So on the back of the New York City death certificate, we printed in multiple colors the instructions on how to fill it out. So when we showed this to our focus group of doctors, they said, we’ve never seen this.
And what had happened was that the clerks didn’t want to waste the paper so what they were doing is they were xeroxing the expensive paper and just giving the doctor the front and not showing them the instructions on the back. And you only learn that when you get in the weeds of the data and figure out how things really work.
And when you are dealing with tens of thousands of physicians and all of the states, how do you handle this sort of thing to actually get to getting high quality data. And that is a challenge we will always have and all we can do, for example, on a death certificate is code what we see.
What we have to get to is how do we get doctors even to understand that this an important piece of paper and that it’s not just a paper chase or just another form. And that’s still what we’re facing today and how do we do a better job of getting physicians to truly understand why they are even doing this and there has to be some way of training them.
NCHS now has an online tool for doing that and there are also apps that can be used to improve the reporting by prompting how to fill out a death certificate. It sounds so simple but to the – especially in death registration – everything has to move superfast.
So what do we mean by that? We want it to move fast but the ones who really drive death certificates moving really quickly are the funeral directors. They are the full-time professionals and they are the ones who, whether it was a paper system or an electronic system, they were the ones who were picking up that piece of paper and they were – and getting it as quickly as possible to get a certified copy of the death certificate. And the reason they needed a certified copy is that nobody leaves the room until the death certificate has been completed.
So those are real-world challenges and how do we get there. How do we get from – how do we get better data. It’s still going into the weeds to understand exactly how you get that information that we need and not simply getting a clerk to say if you just write cardiopulmonary arrest nobody will say anything.
So I also wanted to show you a little about this, if you’ve not seen this, the US Small-area Life Expectancy Estimates Project (USALEEP). NCHS was heavily involved in that as were all of the states. This was a Robert Wood Johnson Foundation funded project. It cost about $2 million dollars and our own estimate in NCHS for actually doing our portion of the work was four-person years to do this, to get the data for this project with some tremendous results.
So it enables us – the increasing body of research is recognizing the importance of measuring mortality outcomes at the local level to the identification of health disparities within a population. So life expectancy at birth was essentially calculated and presented throughout the US. Two states were not able to complete that but otherwise it’s a treasure trove of data and we are expecting to be doing this in some form every 10 years surrounding a decennial census year.
So this would provide almost real time data for life expectancy at the Census track level. So it’s a huge effort and a huge opportunity to learn a lot more about disparities across the US. Here’s one example in Eugene, Oregon, of what that looks like.
So with that, I’ll stop there and offer to answer any questions you may have.
COHEN: I’ll start, Steve, a couple of things. Thank you so much. It’s great that you’re at DVS now. My questions are where do you see the vulnerabilities in this system? And what kind of role could you envision for us, the National Committee for Vital and Health Statistics, to be supporting your efforts. And finally, I’d like you to talk a little bit about access to the data. In several venues, there have been concerns about a decrease in access, particularly to small area data and I’d like to know whether there are anything’s going on around vitals data.
SCHWARTZ: To your first question on vulnerabilities, I have mentioned a lot of the challenges that we have at the state level. The data are only as good as the states are doing their jobs well. So one of the challenges is, for example, on electronic death registration, we have four states that are still not on electronic death registration. And this is a technology that has been used by some states for over 20 years.
So then you have four states – I won’t use their real names – but North Carolina, Connecticut, Rhode Island, West Virginia rhyme with those. That’s not to shame them in any. It’s a challenge for states. They have other priorities.
The advantage of an electronic death registration system is huge. The data can be transmitted within minutes and processed that quickly. So one of our challenges, and hopefully, if we are able to get some additional funding, to get those four remaining states up so all the states have it on an electronic death registration system.
We can’t forget about the US territories either. So there are five US territories that do not have electronic systems. They are US citizens too. So we have to figure out how to do that. We have to bring everybody up to a minimal level including the territories.
So with the additional funding that we have been getting, we have been able to move the needle a lot. So we’re not perfect yet but the data are really coming in faster. They are coming in cleaner. We are also – we have the best working relationships ever with our state partners and the state partners with the hospitals and the data providers. We have to really build on those relationships and how can we facilitate our data providers to understand what they need to do, what their responsibilities are, not to just fill out a form really quickly but to pay attention to it.
And the only way we can do that is – NCHS can’t do that. We have to work with the states and the states have to work with their data providers and that’s a big challenge. We’re moving in that direction. Electronic systems help but they don’t solve nearly the need. How do you get people to fully understand that they need to do this really important job – when a sleepy doc at 3:00 in the morning just does not want to fill out a death certificate of a patient he or she doesn’t even know.
So we’re even looking at who are the biggest – who are the power users in the death registration system? We’re looking at that now and one of them – one really big one, hospice care. About a third of the deaths are certified by hospice. So just think about that. If we can work with hospice providers, how can we help educate them and create systems for them to do a better job of understanding why they are filling out a death certificate. It should be easy. It’s not just that they have about a third of the business. It’s that they should know what the patient died of because they should have the record. They should have that chart. So we’re looking at those kinds of things, those kinds of opportunities thinking outside, dare I say, the box.
STEAD: Vickie, then Nick, then Linda.
MAYS: Welcome and thank you for spending some time with us. Since you know Bruce, probably Bruce has shared with you the kinds of activities that we’ve undertaken on the Committee. I guess I have a couple of questions. One is do you have priorities that you’re trying to address that we might be helpful with.
And two, are there a set of specific questions that you have, either in terms of exploring technology and the use of the data system, data access and use, some of the things that we kind of focus on in these other areas because we have the ability to be able to explore, to bring people in, to talk with people. So it would be great to hear if you’ve kind of thought about areas that overlap with us so that we can think about those. And then if you have priorities.
SCHWARTZ: One priority on the medical examiner coroner side, external causes of death get 90 percent of the medical examiner coroner cases in to us within 90 days. Huge goal. But we’re moving towards that.
In terms of the data themselves, we are getting the – we also run a National Death Index. So the National Death Index is a huge resource. We are getting those data out more quickly so that users have access to that, researchers have access to it, so much faster. It used to be three or more years before the file is completed. Our current file is up now. I think that’s 2017. So that’s huge.
We are working with research at Research Data Centers so for researchers who need more data in a form that is otherwise not accessible because of personally identifiable information. We are working to make it easier trying to make it more accessible so that RDCs are – either RDCs or the virtual RDCs can be providing those sensitive data more quickly to researchers.
COUSSOULE: Thanks again for the time. I just have a slightly different question that may be not quite in our purview but it’s more of a curiosity. How have you or have you been able to demonstrate on kind of the consumer side, not necessarily the data creators but the consumer side what the value of having this faster would be? And I guess I just think about it from the standpoint of it is almost always faster is better and it’s almost everything you do. The question is how do we either convince or sell the value proposition on the frontend to make sure that the people that are actually providing the data are making the changes at the state level or within the states recognize that value when they are being asked to change or encouraged to change.
SCHWARTZ: Thank you for asking that question. Here is one of the solutions. Many of you probably know about public health accreditation and how state and local health departments are being accredited. Within the last month or so, there was a special unique public health accreditation category created just for vital records and vital statistics. So that was just approved by the Public Health Accreditation Board. We’re very excited about that. We are hoping to fund states to start doing that.
And the PHAB board has said this is the only time they are ever going to create a special category within public health accreditation and it is for vital records and vital statistics because it is so unique. So public health accreditation, it’s already been proven, and now this will help – there are standards already written. The standards have been approved.
So working with Public Health Accreditation Board and through the Professional Association, NAPHSIS and NCHS, this is a real opportunity to show to the state health officers and to others how important this is and that there are rigorous standards that the state health offices have to follow. So that’s a really huge one and we’ll be seeing results hopefully this year.
KLOSS: That does sound like a big step forward. One of our jobs at NCVHS is to advise the Secretary on code sets. And we know that in May of 2019, ICD-11 is likely to be adopted by the World Health Organization. And trying to get ahead of this, the Committee is doing some project scoping to think about how we evaluate the merits and benefits and, as you know, with ICD-10 cause of death reporting using ICD-10 began in 1999 in the morbidity uses brought up the rear 15 years later.
So I just wonder – I think we’ll most likely be calling on you and working with you on some form of evaluation of ICD-11 and wondered if you’ve taken any look at that and what your preliminary thoughts are?
SCHWARTZ: So, I haven’t but we have colleagues who have been working with our –
KLOSS: Donna Pickett and her team –
SCHWARTZ: Yes. And Bob Anderson, Chief of our Mortality Statistics branch. So we are heavily involved in that, of course, and – I don’t have more to report on that but if the Committee is interested we can certainly –
KLOSS: It’s kind of the key lynchpin of what the data looks like so it will be very interesting to I think do some comparisons of what the benefits of 10 versus 11 for cause of death reporting will be.
SCHWARTZ: I remember living from 9 to 10. And in the comparability tables and figuring all of that out, it’s quite a thing.
KLOSS: It’s quite a thing. And a thing to be repeated. Thank you again for being with us.
STEAD: Let me add my thanks also. Go ahead.
PHILLIPS: Sorry for the late request. Steve, thanks so much for this presentation. As you may know, looking back to September of 2017 and coming forward, we’ve made several recommendations about NextGen Vital Statistics and one of the things I was struck by in the last six months is the success you have with getting PCOR funding for that particular project. Is there any opportunity to go back to that and make the case for some of the more general recommendations that came out of our process to improve vital statistics going forward? Is that a source of funding for us to draw on further?
SCHWARTZ: That’s a great question. The benefit of say the urgency of the opioid epidemic has awakened everybody and they are seeing the opportunities of getting the data in faster. So if you’ve had a chance to look at our website and see that we’re putting out monthly opioid death reports. Where the tragedy of the opioid epidemic presents huge opportunities for us that over these years NCHS has been getting additional, generally one-time funding, not permanent funding but one-time funding for doing a lot of special projects to improve the entire National Vital Statistics System so we are benefiting from it in many ways and also sharing funding with states so that the states can build their systems better. As I’ve said before, we’re only as good as the state data that they are sending us.
And I think another point about this is that with public health accreditation that we have to have feedback mechanisms that the federal government can only look so far or so deeply and we have to rely on the state health departments to do their jobs to be checking on – to pushing hospitals and funeral directors to do better jobs in their reporting, because that’s where the data come from. So we have to look to see that there are standards in place so that – why shouldn’t hospitals be audited by state health departments. They can do that. They can ask to see charts. They can look to see if this actually matches a record on file.
So we have to get down into the weeds by working through our partners in the states and the states to the data providers. And I’ve talked to funeral directors and asked them what do you put on race and ancestry on a death certificate? We know that that’s pretty important to us when we’re actually looking at the data. It’s the funeral director who fills that out. And or may ask it over the phone, how are they doing it.
So we are totally dependent on those local providers to give us the right information and I talked to funeral directors where they say oh, I’m not going to ask the family. That would be offensive, to ask the family what is the race or ancestry of the decedent. But how do you do that? Do we have a standardized form? Each state may – well, not even a state has.
There are large industries in what they call the death care industry and within that they set their own standards and say this is how you are – that big provider will say, this is the form you have to use. How do we get those private companies to change how they collect that information or even care enough to ask. I’ve heard of situations where someone would say, Dad never considered himself white. And how do you get at that level to get that information – but it’s what we need.
STEAD: Bob actually started where I was sort of trying to think about. When we looked at this, we looked at it pretty hard and with Gib Parrish’s help we got a really good fact base which took a lot of the onions off of our eyes. And we in the end came out with a relatively simple statement to the Secretary which is that the system is fragile and that federal leadership is essential.
We recognize it’s a federated system and we’re not suggesting it not be a federated system but that leadership is essential and that does not mean that we’ve not had leadership of the sort you’re describing and that that hasn’t made progress. It’s just that the trajectory of the progress we need to make given the essential nature of this system for everything from identity to, at the end of the day, national security, came through quite loudly and clearly to us – through both the hearing we had and the environmental scan.
So what would help us is to know – given we did what we sort of knew how to do given what we do with the good input that we had – what we now need to know is given, as you now have assumed the leadership in this space and are moving forward, given what your priorities are what are underway what we sort of put out there, where could we be most helpful in accelerating progress.
And what are some root causes because yes, the education is going to have to be local but in our privacy space, if we were dependent solely on the local work with hospitals to get privacy practices right, without the leadership from OCR, and making sure we got reusable materials that those people can leave, we wouldn’t have moved as fast as we moved. So we see it as really needing to be a partnership, which it is, but I think, from the outside looking in, we think that the federal role needs to be strengthened and any way we could help sort of to know what the next step might be that we could help with.
SCHWARTZ: We certainly welcome the help and the support. I am familiar with the report that was written and it’s – I think one of the –
You’re probably all familiar with what we call the Good to Great Committee. So really in just a minute or two – 10 years ago, the National Vital Statistics System was sleepy, terribly sleepy and had been. And there were pretty low expectations for ever changing that. And 10 years ago, I was part of a committee that was on the side of the states saying this is not working. And it was – at the time the committee was called the Cost Committee and it was all about how can NCHS either pay more money to the states or what do you want to cut.
When I first started 20 years ago in vital statistics, one of my first awakenings was that NCHS said we’re no longer going to collect induced termination of pregnancy data. And I remember at the time, 20 odd years ago, it was something like $75,000.
And over the years what happened was every time NCHS negotiated with the states, NAPHSIS the representative, NCHS would say, we’re sorry, we can’t pay for that. And the states would say, well, if you’re not going to pay for it, we’re going to remove marriage and divorce data. They did that. And then they took other parts of it, like induced terminations.
So over the years, what they were doing was they were cutting off the arms and legs of the National Vital Statistics System. And that had been going on for years. And at one point, we started meeting and instead of calling it – well, initially it was called the Cost Committee – but as we met, NCHS staff and I on the other side, on the state side, and we talked to each other and said, the thing is that we are all part owners of the National Vital Statistics System and I think that’s the most important message of all. It’s that we are all part owners and the only way to do this was to figure out to make the system better and by making it better, people will flock to it.
And that’s the process we’re in really now. So 10 years later, that committee became the Good to Great Committee and we called it that because it’s like well, we’re not perfect but we’re a lot better than we were. We said we’re getting from Good to Great.
And what we did was over that period of time that we said we have to do better, here’s how we can do better. And what happened was – and this was 10 years ago now – and you all know that Vitals is a federated system. It is the state system and the only reason states send the data to NCHS is because we get paid for it, or the states get paid for it. There’s no law that says vital statistics have to be collected in the US.
So the National Vital Statistics System is a creature of that. So what we did was 10 years ago when the states and NCHS negotiated, we said we are going to make a better system and we are not going to charge NCHS any more money for it. So when we had the contract, that breakthrough, that change was that the Vital Statistics Cooperative Program which costs about $20 odd million dollars a year, not much money, that that system was so important that it was better for the states to do a better job than to get more money doing it.
And that’s what was agreed to and actually – so in the last 10 years there’s been no increase in the funding to NCHS for vital statistics, in like a 10-year period. So that’s a problem. Because the states are building more expensive systems. We have to figure out how to be efficient at doing it and sharing resources so we have to work on those kinds of things so concepts like interoperability are really important because it can make things more efficient and we also get the data in faster.
So we’ve now proven that the National Vital Statistics System can do the job. That’s fabulous. What we have to do is build on what we’ve done so far. And for people to see how important the value-added is.
STEAD: Thank you. I think we are at the end of our time slot here.
SCHWARTZ: Thank you for inviting me.
Agenda Item: ONC and HITAC Update and Discussion
WHITE: Greetings, citizens of NCVHS from the land of ONC. Good to see you all. Thank you for coming here and as always thank you for the amazing work that you do, much of which I have here in front of me at the moment. So I was kind of prepping up for the meeting and I always think, God, I’m glad these folks are here. You guys do great work so I’m grateful for it. Thank you for having me here today. I am very excited to be able to share with you a couple of updates from ONC and to tease you about things yet to come.
We’ve been pretty busy at ONC, as you might imagine. And there’s some stuff – I don’t know how much you guys like pay super close attention to the goings on in the world of health IT, but some stuff you are probably well aware of, some stuff you may not be aware of but we’ll get to the discussion section and feel free to ask me about anything you want to.
You may or may not be eagerly anticipating the release of our proposed rulemaking to outline policies that were required of us from the 21st Century Cures Act. I am pleased to tell you that we are anticipating this very soon. Get on your boards, start paddling, you’re going to catch a wave pretty soon.
So the particular rulemaking that we’ve got in there relates to conditions and maintenance of certification for health IT. That’s our certification program, the ONC certification program.
We also have provisions in our rule related to information blocking which was a topic of discussion for the past couple years but specifically requested by the 21st Century Cures Act. It’s going to be interesting. Like I said, buckle up.
In particular, just so you’re aware, you probably know but ONC’s rulemaking is to specify what is not information blocking. The Office of the Inspector General takes enforcement actions on charges of information blocking. That’s how it’s outlined in the 21st Century Cures rule. But our job in our rulemaking is to propose and then ultimately finalize what is not information blocking, what might constitute a reasonable reason to not share information with folks.
So the implementation of these provisions, of course, are looking at advancing interoperability. That’s largely what the 21st Century Cures Act is about. And supporting the access, exchange and use of electronic health information, in particular through open application programming interfaces. And again, we can get wonky if you want to. And transparent and not inappropriately inhibited data sharing.
Those of you who are careful followers of OMB’s website may also be aware that there is a concurrent CMS rule on interoperability that has been under review as well. Stay-tuned.
We’re getting near the finish line for our next iteration of our draft Trusted Exchange Framework. This is relating to policy guidance that ONC has issued, not a regulation but policy guidance that ONC has issue on health information exchange and those organizations and how we establish basically better trust between these organizations to facilitate information sharing. And we are looking forward to your input there as well. We anticipate that a Notice of Funding Opportunity will be posted for a cooperative agreement to select a recognized coordinating entity and that should be released in the near future as well.
Back, way back in November, those were the days, we had our eighth Annual ONC Meeting. The theme was Highlighting Progress Driving Success. It was chockablock with distinguished speakers, panel discussions, breakout sessions. If you go to healthIT.gov you can dive deep should you wish, should you have nothing else to do like reviewing a several hundred-page rule.
Also in November, we released a draft Strategy on Reducing Regulatory and Administrative Burden Relating to the Use of Health IT and EHRs for public comment. This is a requirement also under the 21st Century Cures Act. It was done in partnership with CMS as required by the 21st Century Cures Act and we have gotten several hundred comments at this point, which is great, on this draft. And we are going through those now. In fact, my team is going through that literally as we speak. To allow physicians and other clinicians to provide effective care to their patients with a renewed sense of satisfaction for both them and their patients. So we’re looking to release a revised final version of that in the Spring of 2019 so Spring of this year – remember Spring goes until June 20th – based on input from stakeholders.
Additionally, under the 21st Century Cures Act news we concluded a request for information on the EHR reporting program. A total of 77 comments received from that, from developers of health IT provider organizations, payers and health plan organizations and various folks. We will soon be holding nationwide stakeholder engagement sessions to further inform and develop the program in related draft reporting criteria.
Last but not least, our advisory committee, the Health IT Advisory Committee or HITAC met last December where they continued to discuss draft recommendations related to standards for referral and care coordination among other things. The first meeting of this year, we had a meeting scheduled for January. In one of those curious twist of fate, we were not shut down. Mercifully, we had our appropriation literally before September 30th, literally for the first time since I was in medical school so we were not shut down. However, in order to hold, as you well know, an advisory committee meeting you have to have it published in the Federal Register which was not open for business. So we had to cancel the January meeting but the first meeting of the year is scheduled for February 20th and we are looking forward to sharing a timeline of activity. Just on a parenthetical but personal note, I am incredibly grateful for your renewed commitment in engagement for coordinating between the two advisory committees. We love our advisory committee but we love you as well and you all play very complimentary and much more effective roles when you are coordinated. So I genuinely appreciate you all working with them.
Should any of you be planning to attend HIMS next week I will be there along with several of my ONC colleagues. We have four educational sessions there and perhaps more other sessions who knows. We will be sharing more information on ONC policies and standards work.
I do want to mention, Rebecca, asked for just some brief comments on some of your excellent work, which I have liked. They are going to be broad comments so don’t look for me to be wordsmithing. I like your work, both on the letter to the Secretary with your recommendations. I think that this is extremely cogent advice in your five recommendations on improving the adoption of national standards. So I genuinely appreciate it. I don’t have specific feedback for you to change this or do that but what I will say is that this is going to be extremely useful for us should this be what ultimately moves forward to be able to take both to our internal processes as well as our other processes as well.
I will finally add that reviewing your 13th Report to Congress – I clearly saw the hand of Dr. Stead. Reading through this I’m like, I have seen reports like this before which is a very effective way – I have always loved Bill’s reports, whether they are National Research Council or National Academy of Medicine or what-have-you. So bravo. Thank you for clearly laying out the coordination with ONC and here I think everything you pointed to is spot on and I appreciate it.
So with that, I will stop blathering at you and let the good times roll, what should we discuss?
LOVE: NAHDO has been following closely the TEFCA, how much can you say or is it too early to talk about the TEFCA and the framework. Because what we were trying to do is figure out how the implications for the existing state initiatives and the huge variation of laws across the states, and how that might work.
WHITE: Quick word into the discussion we just had here previously with public health reporting and vital statistics report. So, I mean I can say some but just broadly. So I’ll briefly recap. So one of the provisions of the 21st Century Cures Act is for ONC to either adopt an existing or establish a Trusted Exchange Framework and Common Agreement that helps improve trusted exchange between organizations around the nation that are exchanging health information. And many of these are at the state level for very obvious and good reasons. And actually ONC has a history of rather robustly funding state health information exchange organizations.
So we issued a first iteration of the first part, the Trusted Exchange Framework and we did that last year in January and we got great comments back. Don Rucker, my boss, has said publicly that we are going to issue a second draft version that will be coming out again in the near future. Honestly, we’ve been piled up with a lot of different things as frankly have been our channels of clearance with whom we have to coordinate amongst the government. So there’s a lot of kind of policy guidance coming down the road.
So what I will say is that it is not languished. It has been actively worked on and you should see that next version soon. But it will be open for comments, it’s not going to be finalized. So that’s the Trusted Exchange Framework.
The Common Agreement piece is separate. We often say TEFCA and that unfortunately kind of mushes them all together. So the Trusted Exchange Framework is just that. It’s a framework. It’s meant to establish principles and it also has some specific technical guidance that’s associated that we added.
The second part, the Common Agreement, we are looking to have established actually by the private sector, and I mentioned the Notice of Funding Opportunity for Cooperative Agreement for a Recognized Coordinating Entity. And we’re looking for that coordinating entity to work with us to establish the Common Agreement to which health information exchange organizations will adhere, to that they will actually sign. So that agreement won’t be directly with us, although we will look to it but with the coordinating entity. Sorry, I know you know all that.
LOVE: Well, I forgot a lot of it. I read through it all and we commented but is the vision to be more expansive than just HIE to HIE but include other kinds of data that could be hospital discharge data or claims data that states include. I think we’re pushing for more expansive vision of just HIE to HIE.
WHITE: So it’s a great point. So I’ll give you a general comment which is that of course information exchange is not limited to HIEs or HIOs or whatever kind of designation you want to specify. I think what I would say is that guidance relating to that is going to be contained not just in the Trusted Exchange Framework but also in Proposed Rulemaking that you will see in the near future. So don’t worry. We’ll have something to keep you busy, if you are eagerly awaiting the TEFCA.
STEAD: I will pose a question as the generalist and then add others in the Standards Committee to correct it to the degree it’s incorrect or to flesh it out.
We want to figure out what we can do to get traction on what we know will be a long journey to convergence of the administrative and clinical standards. And it’s one possible point of intersection that we thought might be a use case that we might – we, ONC and HITAC actually might be able to explore from our different perspectives, if not together at a table somehow, that might inform what we’re all thinking about which would be the area of prior authorization. Because one thing we could do if we were not trying to understand conversions would be to look back at that again within the traditional lens we brought to it.
Another thing we could do is to try to look at it and if we wanted to do whatever we were doing there as a step toward convergence, what might that look like in a way that doesn’t get us all tied down in making something too complex but might, in fact, use a problem that we, I think both ONC from its perch and we from our perch, are thinking about dealing with. Is there a way that could be as a use case to advance the conversation of what convergence might look at and therefore what the path might be, while hopefully doing some thing useful in the space of prior auth as an early byproduct.
So if I pose it that way, did I hear Alix, do you want to chime in before someone replies to correct what I did wrong?
GOSS: Thank you, Bill. No, I think you framed it up well and I’m eager to hear Jon’s response.
WHITE: Yes. So the longer response is actually I think that would be an extraordinary useful particular example. We’ve all been swimming in these waters together for a long time. But yes, I actually think that’s a great example of a particular use case. Aside from the fact that I made a personal promise to a guy I know three years ago to fix prior auth and I haven’t done it yet.
But I actually do – it’s not quite that perfect storm but it’s a great like thing to seize on and at a minimum, pick at the threads. But ideally I would be amazed if we didn’t find some specific things that we could use to move that forward. If it’s right in the theme of – you’ve seen out of Don’s leadership, not just particular attention to the technology but also to the administrative requirements associated with practice and this pertains to all the discussion, collapsing one through five codes that we’ve been having over the past year, year and a half. So yes, the short answer is, I and ONC, would be enthusiastic about doing that.
STEAD: Thank you.
COUSSOULE: The only thing I would add to that is I think there is enough impetus for each of us to be undertaking this from the constituents that we work with regularly. It would be a shame if we didn’t try to come together on that because otherwise we may end up in a place that sounds better from one side or the other but doesn’t make the system better in total. So I do think this is a really good opportunity for us to do this. It’s not a trivial exercise by any stretch of the imagination and I think it’s really important to look at it holistically that way.
WHITE: Prior auth is a great example of a well-intended, potentially appropriately used, kind of regulation on the use of health care that has just gone insane, that’s a technical term.
STEAD: Other questions.
GOSS: If I can react. Jon, I think the next steps are as we work through our work plans thoughts to ensure that we have an appropriate path forward with HITAC and ONC on that prior auth use case. So I think we should cue up some further discussion and look for your guidance on how it’s best to do that. We’ve been coordinating quite closely with Rebecca and Lauren who have been helpful, and if that’s how you want it to remain that would be good to know or if you want to have some preliminary leadership discussions like we’ve done in the past.
WHITE: Okay, totally will.
STEAD: Rich and then Linda.
LANDEN: Jon, one of the things that took me a little bit by surprise, not a lot but that was in the course of the Standards Subcommittee’s look into the Predictability Roadmap and particularly in the final hearing we did a couple months back. The surprise was less about the industry consensus around their belief – and these were practitioners and users – that the convergence of administrative and clinical data is happening and there’s no longer any meaningful distinction or differentiation between the two.
What surprised me about that was the strength of that consensus. I mean I think we’ve all seen this trend coming and academically acknowledged it. But you see that very clearly in the reports and the recommendations that we were addressing yesterday and will be sending to the Secretary and Report to Congress. So essentially, why I’m saying all this is to gauge your feedback. Is that something that ONC and HITAC you believe are aware of and are actively thinking about in whatever you’re doing that applies to those two formerly separate worlds.
WHITE: We’re all beset by our implicit biases, right, that kind of haunt us through our daily for life. It’s an actual thing for us to continue to be like, oh well, that’s billing data. It’s long been my view that that’s not the way it is, data, our data.
Where it gets a little – so I’m in accordance with the strong consensus. Where it gets funky is who – and you start talking about federal regulation and different authorities – is who has control over what types of data. And that’s where it becomes interesting and challenging to navigate. While I am always amendable to stopping, taking a step back and revisiting how we have our systems set up, so as to recognize that that artificial distinction without a difference plays out and are there ways and things we can do to change it.
Unfortunately, as you all well know, it’s the reality of the world that we live in and it’s left to the dirty work of hand to hand trench combat of stuff like what you do here and what our HITAC does. You asked if ONC was aware of it. Yes, I know ONC is aware of it. Is the HITAC aware of it? Yes, sure, because it comes up in their discussions all the time when they are talking about their doing their own subcommittee work and getting into the things like, well, you know, that’s not ONC’s purview and they are like, but, that data is important, yes, but, which is why it’s important to stay coordinated.
KLOSS: I guess my question is sort of related. You commented on the Standards simplification letter but how do you view the work that we’re doing now with vocabularies and terminologies and kind of starting to advance thinking on ICD-11. And how do we look forward to coordinating that because that does get into clinical content. That may be an area for some coordination as well in a single sort of vision for a path forward.
WHITE: Anecdote time. I was at OMB when the final approval of the 2008 transition from ICD-9 to 10. When I was at ONC it was at ARQH at the time. But I was there and everyone was like yes, this is a great idea and then you all saw that show that happened afterwards.
KLOSS: We’re trying to avoid that.
WHITE: And actually – you know me, I’m a bubbly optimistic guy so I am optimistic in thinking that oh, we all went through this already and everyone is now going to like not completely panic when they hear that okay we’re upgrading to a new system. There will still be some of that that goes on. I do think it will be less than it was last time because there was such Y2K oh my gosh the sky is falling and it wound up being not.
Now, the lesson to be learned from that also is that we gave a fairly short lead time in that 2008 rule. I think adjusting the lead time period which will never be long enough, we all know that, but thinking about how we roll that out.
But to the broader issue that you’re asking about you all getting into the clinical – yes, you guys always have, that’s okay. As you should, which is why it’s so important to see it coordinated.
KLOSS: I think that’s just another area for coordination, the code set standard as well as the administrative standards.
STEAD: Be cautious that you know from what we gave you and we did approve it yesterday. We’re sending the letter to the Secretary with the criteria for adopting name, terminology and vocabulary standards. The first update in that since 2000. And with guidelines for curation which are new.
And from our perch at least, if we could make progress there it then increases the semantic harmony in the value set you are drawing from those terminologies for various of the value sets within the promoting interoperability guide path, et cetera. So we at least think it’s important to work on how we harmonize the set that we draw upon at the same time we’re working on the standards that in fact deploy those sets. I don’t know if that makes any sense.
WHITE: It makes great sense. I would just briefly say that having watched various things play out over the years gets fascinating to me. And when something gets broke, genuinely broke, like stop, all full stop broke, those things do tend to get fixed faster.
The pace at which we adopt new standards – the system has bent kind of a situation and therefore the urgency to do it is less because we’re worrying about broke things and stuff like that. But boy, is it frustrating and really, it’s like we don’t have to do it this way but there’s just not that kind of motivating crisis you do when something’s broke.
So like I said I thought it was a very cogent set of advice and again, having watched it play out several times over the years, sometimes it does get broken and you need to throw something in, a payment rule or something like that to get it fixed. But surely we can be smarter about this while at the same time trying to be respectful of the kind of did you process that we accord to standards development so we don’t just thrust new stuff upon – foist stuff upon people.
STEAD: Thank you. Other questions. Any question on the phone? Alix or Jacki? Thank you all. And thank you, Jon. We really appreciate.
SMITH: Absolutely my pleasure. Thanks for having me here.
STEAD: Before we break totally for lunch what I would like to do is get a picture taken of the Committee members and staff that are here since this will be our last chance to get a picture at a full Committee meeting with Linda and Bruce and we need to record this moment for the future. We will figure out a way to do that but of course we’re the ones that get to pick the images we photoshop in.
We will reconvene at 1:00 and we’re going to begin with the work plan discussion and in particular the ICD scoping discussion that we were not able to get through last night so that we can flow into Health Information Privacy and Security Beyond HIPAA on time at 1:30. Thanks all.
Agenda Item: Action Items from Day 1 – Final Revisions
STEAD: The two things we really have to land are the ICD scoping and Beyond HIPAA. We have the block for Beyond HIPAA. Let’s start with ICD scoping and see if we have time to get back to the broader work plan.
KLOSS: Do you want to work off the document? I had a couple of slides on it, but that is all right. We can work off the document. I do not know if everyone wants to look at the detail of the scoping document, the actual words.
STEAD: I do not know that we need to.
KLOSS: Greg, could you bring up the ICD or the terminology and vocabulary slide deck from yesterday and just scroll to slide number 9 please?
HINES: The HTV one. It said NCVHS Health T-V Overview.
KLOSS: As we discussed yesterday, we have kind of made the decision that we closed out the project scope that got us through the environmental scan, the roundtable, and the first two letters, this near term health terminology and vocabulary work.
And that we were next then moving to a focus on ICD-11, create and evaluate in the pathways to ICD-11 and created a new scoping document. That is in the work plan.
The first part of the work plan just lays out the facts that ICD-11 is intended for use for both cause of death reporting and morbidity reporting. That the timing for this particular work plan reflects that the May 19, the World Health Organization, will consider adoption of ICD-11 with an effective date of January 1, 2022. And that that release triggers a new cycle of evaluation by the National Center for Health Statistics as the US Collaborating Center for ICD and for NCVHS in its advisory role to the Secretary. That is the opening paragraph teeing up this project scoping document.
The goals for this new project would be to evaluate ICD-11 against updated principles for adoption and implementation that we approved yesterday.
Evaluate the cost of implementation against the benefit of moving to ICD-11 also addressed in yesterday’s letter to the Secretary where we requested resources to commence a project.
And evaluate the cost of implementation against the benefits of moving to ICD-11 for morbidity classification. Again, the paths for mortality will be different than the paths for morbidity because there is a second decision point with morbidity even though the statement of purpose by ICD is that ICD-11 was developed for morbidity classification, but the US will evaluate whether we need to do a US clinical modification to ICD-11. Again, for those that are not into this, the ICD-10-CM, the CM is a clinical modification for morbidity classification. The question is to what extent has the World Health Organization developed ICD-11 for morbidity classification. Is it robust enough to support the US needs or do we need a clinical modification because that certainly affects the complexity of development and the timeline for the development and it informs what kind of pathway we would recommend.
And then a fourth goal is to consider and I presume make recommendations to the Secretary on stakeholder engagement, industry communications, and as we discussed earlier we see a real benefit to doing that sooner rather than later. That is the goals.
And then what this scoping document does is just lay our four phases, two of which we have kind of largely tackled. We laid out the rationale for simplified regulatory adoption of code set version updates in our letter that was approved yesterday. We have through the work of NCHS, done a pick it and staff. We have a document that lays out a timeline for ICD-10, what the agency went through and what CMS went through in preparing PCS. We have a start on this.
And the next step then would be to identify cost benefit research questions because I think as we are thinking about this, this research on cost benefit is certainly something that would have to be contracted for, but that we would inform how to execute on our criteria with regard to cost benefit research for adoption of the next version of ICD for mortality and for morbidity because in some ways, they are different paths.
And then design a proposed pathway and timeline for adoption. In this scoping document, we laid out – we thought that work in Phase 3 could get done in 2019 and that 2020 would be when we would be ready to propose or get to a proposed pathway. We saw this as the work of the balance of this year and at least part of next year. Pretty targeted.
STEAD: I think from my perch, we did complete Phase I, which is to get the letter done. We have not yet gotten the research that Donna has been compiling in a form that we can share with the committee.
KLOSS: I actually do have a document from her, but we have not brought the group back together again to look at it and see where we need —
STEAD: That is partially — it seems to me that then compile existing research on impact of ICD-10 is work that we would need to do over the course of the spring I am assuming and then solicit feedback on how ICD-10 is working for cause of death reporting.
My middle model and correct me if I am wrong as we are trying to figure out how to land this airplane is that we might try to assemble that research, which is really capturing history. It is a little bit like a scan. Before the June meeting. Assemble the existing research. Bear with me before you frown. We have already calendared an August roundtable. It seems to me the purpose of the August roundtable might be to, one, share the research that we have an initial lens on with the Full Committee in June and to then vet. That would inform the vetting of the research questions for the evaluation of ICD-11 going forward.
It seems to me that is what we want to accomplish in the August roundtable, which says to me at the June meeting, we would want to be closing down the agenda for that roundtable and we would want to do probably an initial session at the Full Committee meeting. I know this will be beyond your perch – where we would share some of those findings and have an initial discussion of what the research questions might look like so that we inform the process going forward to flow into August. That is how I read this. That is a little more detail.
HINES: Bill, let me just interject because you are talking so fast and I want to make sure we are all with you because I am not quite with you. Just to reflect back that the idea is to develop the research questions for the evaluation of ICD-11 and the purpose of the meeting in August is to —
KLOSS: First, we do the look back at what is the status of 10. Obviously, it is still early. We are not going to have a full impact of ICD-10, but we know something. That should be pulled together.
STEAD: What you have from Donna that we are going to bring in is the summary of what we did to evaluate ICD-10.
KLOSS: And develop the CM.
HINES: What is the purpose of the — what are you honing in as what we are going to get accomplished in August?
STEAD: Convene a hearing of terminology experts, economists, and health researchers to design the research questions.
HINES: It is in that other document.
PARTICIPANT: I should have looked at the full document.
STEAD: Because I think that is what we are hoping to do.
HINES: Is to come out with the research questions and then move forward on the evaluation.
STEAD: Right. Then that meeting would be in late August and then we would –
HINES: It is actually early August.
STEAD: We would then be able to report at the September meeting, share where we are and we would be drafting a letter reflecting what we think the research need is informed by that, to the Secretary either in September or late in Q4. Is that pinning down what we are talking about?
PARTICIPANT: Yes, it is.
STEAD: The other thing we need – so we need to get Nick and Alix because this will be done under the auspices of the Standards Subcommittee as our current T&V work has been done.
The other thing that we have talked through with the Standards Subcommittee and I think the people involved is that the working group that is going to be moving this forward is initially you, me, Rich, I think, the ICD –
PARTICIPANT: Remember the calls we had, Rich, being voluntold.
STEAD: You have been doing it. I am just trying to connect the dots here. Mike Lincoln and Steve Brown and the Vivian or Suzy from NLM as the small work group – and Donna would be driving – it is the group that got us to where we were with these three letters. That is just trying to get all the fuzz off that peach.
STEAD: The document is written. It does not quite say all that. We will have to do a little bit of editing while it is fresh in our mind.
KLOSS: Well, because now it is clearer having been de-fuzzed.
STEAD: Bruce has a card up.
COHEN: I have not seen Donna’s review of the research done.
KLOSS: Hers is a timeline. We do not have the research that has been done.
COHEN: I might not be around for this input. Historically, when I have reviewed, examining, and evaluating the transition from ICD versions, there has been a real deficit in looking at comparability ratios, post-implementation and the impact that has on mortality trend analysis. In particular, there has been virtually nothing done in examining comparability ratios for population subgroups where there might be huge differences. I would put in a plug for forward thinking about looking at post-evaluation strategies to look at the impact of changes on mortality trend analysis. I think that is an area that has been neglected.
LOVE: Not only mortality comparability ratios, but there were no morbidity comparability ratios at all. A researcher, Pat Romano, had to go some private data sets and cobble together those ratios for morbidity as well. We need to think about that in addition to mortality.
KLOSS: And remember Patrick was at our meeting last August. We have dubbed him part of the team too.
STEAD: Are we good with this as the current – we will tweak the thing and then this updated scoping document can go to the website. And then one of my things I had in the note is we need the next call of that work group this month.
We have six minutes. Do we want to go back to the high-level work plan and talk through in particular the block at the top and maybe the stuff more in the – it seems to me the thing that might be worth talking through with the whole committee is the stuff up at top and the stuff at the bottom and not go through all the lines for each of the subcommittees.
HINES: Greg, first page, we can see as much of that list as possible. I will just quickly go through this. We are at the Full Committee meeting obviously. We are at the February one now. We have one for the first week in June and the first week in October.
We already got the Federal Register Notice through in case anything should happen in case it shuts down for the March 21 – we have heard about it twice already, two meetings being cancelled. We went ahead and thanks to the ASPE, it was quickly taken care of last week. It got published on Friday. Thank you, ASPE team. We are all set for that working meeting.
And the ICD evaluation project that we just discussed – if you want to hold another convening of a subcommittee, there are resources to do that. There was discussion about another use for those resources in terms of a report that Linda and Maya had been discussing. I think that is an open question for the Executive Subcommittee or this group at some point. I just wanted to go over here is where we are with resources for the committee through the end of this fiscal year September 30.
STEAD: From my perch, the candidates that are being discussed at the subcommittee level for the TBD slot are something around the DSMO because that is relatively time sensitive. The possibility we just mentioned of something around prior auth possibly in coordination with ONC or possibly simply on our own.
And I think the third possibility that we have discussed so far would be using the resources to get a writer to deal with the update of the late 2000s report that we are going to work in the Beyond HIPAA hearing. Right now, I think those are the three candidates for resources, two of which would involve a convening, one of which would not.
My view of the world is we need to make a decision at the March Executive Subcommittee. If we do not make the decision then it is too late to do anything in this fiscal year. I think each of the subcommittees needs to come into that meeting ready to speak, make their case and/or forever hold their peace.
HINES: I apologize. The version of the work plan up there is wrong. I am looking at the current one. I will email you what Bill just said. I had updated this and I sent the contractor the wrong version. I apologize for that. But what Bill just outlined is in another document version of the document.
STEAD: Does that compute with everybody or am I missing something?
HINES: It is March 21, 22, Thursday, Friday. It is going to be at the National Center for Health Statistics.
PARTICIPANT: March Executive.
HINES: March 4. Everybody on the Executive Subcommittee got the March 4 on their calendar at 3:30.
STEAD: One of the things we really have to do as soon as we get out of here is – I think set doodle polls. We have to come to closure on the April and May Executive Subcommittee –
HINES: Indeed. It is on the list.
STEAD: I think at high level we are about out of town. My sense is maybe we say we have done what we can do with the work plan. We will stay tuned until the Executive Subcommittee and we will now shift gears into Privacy and Security Beyond HIPAA.
HINES: I just want to be clear that Frank is an incoming member. We have his paperwork in hand and we will eventually get him in the HR system, but he is approved. It is just a matter of closing out some of these details. We are glad you could join us today.
BERNSTEIN: You should know that our executive director gave you a very flowery introduction yesterday when you were not here and noted that you would be here. A thorough introduction.
HINES: And we will send you the audio.
STEAD: And we will send a copy to your mother.
Agenda Item: Health Information Privacy and Security Beyond HIPAA
KLOSS: Welcome Frank. The first thing we are going to do is go back to the Report to Congress and look at what we suggest – we are going to ask people to pull up their own copies of the Report to Congress if they cannot remember where this is anchored.
The reference to largely unregulated first appears on page 11 or the introduction to actions, which is page 76 of the E-Book. My suggestion is that we do not change it there because we are not really digging into it. But I think maybe that we might want to look at that one more time to ascertain that we do not want to make any changes.
This wording comes on page 88 of the E-Book or page 23 of the Report to Congress. As you will recall, the discussion was that we were concerned by saying today there are two health information worlds. One is regulated by HIPAA. The other is largely unregulated. That we wanted some modification of that to reflect state and local regulation. That is really what the purpose of this red insert sentence is. Everybody with me?
STEAD: One clarification. In my understanding of what Bruce said was that use by federal agencies is also regulated. It is not just state. What is missing from that if I understood yesterday correctly is the fact that the laws – I am channeling Bruce.
COHEN: This is moving in the direction that I like, but again it is not just state and local, department of education. Substantial number of federal agencies has privacy, non-HIPAA-related privacy in their mandates. I would say other government agencies are federal and state.
COUSSOULE: Do we talk about things that might be in an agency mandate, but that are not codified in law or statute?
COHEN: It is a combination of both. The Department of Education – I think it is the FISA laws – FERPA – are very well codified and developed and extraordinarily robust and have actually been an impediment to getting access to health data by the way they actually protect the data. Some of them – I would not go that far.
The other quibble I have is these laws are largely unenforced. I do not know that we have data or proof that that is the case. I imagine the entities that generate these regulations like state health departments for protection of cancer registry and vitals and organizations that protect hospital discharge data and APC data would argue that they do not enforce their privacy regulations. I think that is sort of a gratuitous line that does not need to be there.
LOVE: I concur with Bruce. I was troubled a little bit yesterday that that perception would be baked in.
KLOSS: We did not have these words up there yesterday.
LOVE: Right. But it does not get at the federal issue.
SEEGER: Just a point of clarification. FERPA pertains to education records. HIPAA and FERPA – we do have guidance that we co-authored. OCR co-authored with Department of Education that clarifies that the health data that is held by educational institutions as it pertains to a HIPAA-covered entity is covered under HIPAA. The education records – there are federal – Bruce is absolutely correct. There are plenty of federal laws that pertain to the privacy of data like the Privacy Act, but these laws do not cover the health space in general. It is not specific to health data.
KLOSS: I think it would be helpful if we have look at this sentence and talk about how we want to modify it rather than debate the issue. Let’s talk words.
COHEN: My suggestion would be there is a matrix of federal and state laws that govern the use of data in a variety of departments and agencies. Then I would delete the next sentence however those laws are largely unenforced. That would be my recommendation.
LOVE: Would you be comfortable in saying unevenly enforced? Largely troubles me because I cannot —
COHEN: That is unevenly enforced. I could live with that, but again I think it is unnecessary to the point. And the real point is there is a lot of health-related data that are not covered by HIPAA. We do not need to go beyond that by disparaging the coverage of other sets of health-related data where there is privacy protection.
PHILLIPS: I wonder if it would be fair to say that there is a matrix of state laws that signal an attempt to fill gaps left by HIPAA, but they lack federal enforcement. No? They are just separate.
COHEN: They are unrelated.
STEAD: From my perch, Bruce’s last statement I think actually is critical. It plays into all the work we have done over the last two or three weeks to get the tone of this report right and not to overreach. I think this is a case where we want to make sure we are actually communicating the right thing, which may not be the detail of a law.
KLOSS: I think we have consensus that we are going to drop that sentence, however those laws are largely unenforced. That is not really germane to the point we are making here.
And then I think we have consensus that we are going to add to the – HIPAA sets a federal floor for privacy. We can all agree on that. There is a matrix of other federal and state laws that govern data. It still does not work.
PARTICIPANT: (off mic)
KLOSS: I think if we get agreement on the concepts, the next round of concepts.
LOVE: In addition to HIPAA or federal laws, there is a matrix of and then go on and on.
KLOSS: We could say in addition to HIPAA and other federal laws, there is a matrix.
STEAD: I think if you say HIPAA sets a federal law for health data privacy. Put in health data. And then I think what you are just doing may work.
BERNSTEIN: But it is incomplete is the point. I think the problem I am having with this and it is not meant to be pejorative. It is just really meant to be descriptive of what the environment is. I think matrix gives the – I might have been our words. It is not meant to be pejorative. It just means to say that matrix makes it sound like it is complete and fully covered, which it is not. The point is it is piecemeal and in some cases kind of random depending on how the rules –
STEAD: Why don’t you replace matrix with patchwork.
BERNSTEIN: And the idea is that group of federal, state, and local government law is incomplete. I think that is all we need to say. Talking about FERPA or talking about any particular law, there is a huge swath of stuff that is totally not – unless you count the Federal Trade Commission coming after it at some point. Consumer-generated data and data brokers and all kinds of other health data that is out in the atmosphere that is completely not addressed by any of these things. That is what we are trying to get at.
KLOSS: Our purpose in adding that was to acknowledge that there are some other laws.
COHEN: The point that I think is important to make is there are other government laws that protect the privacy of health-related data; however, they are incomplete. I do not think matrix is needed. I am comfortable with patchwork.
And most of the other protections are in my mind are government related protections. Government held health related data.
PARTICIPANT: Can I jump in just for a second because I want to get under the word part of this. I think it matters. I think the term patchwork is very appropriate because it is not planned. It is not a planned and definitive time. It is for very specific purposes, but not for the general protection of health data. I think that is a very good word to use that way. Sorry. I did not mean to interrupt.
PARTICIPANT: I also like patchwork. The other approach I might suggest is that in addition to largely unregulated, we may say the other is largely unregulated (that is, Beyond HIPAA), or poorly coordinated with HIPAA.
STRICKLAND: I was just thinking of the other unregulated non-HIPAA because it is everything else. It is just taking that it is non-HIPAA. It can be all of the things that you say it is. It can be patchwork. It can be everything else, but it is just not HIPAA.
KLOSS: We have a suggestion then that we modify the theme. Today there are health information worlds. One is regulated by HIPAA. The other is largely unregulated and uncoordinated.
STRICKLAND: And if you say by HIPAA then it is specific to HIPAA. It covers all their things that they want.
KLOSS: I kind of like the impact of a cleaner theme statement.
STEAD: (off mic)
KLOSS: I think where I am at right now is that HIPAA sets a federal floor for health information privacy. And then I would go on to say there is also a patchwork of state laws that govern data use and state agencies local and county health department. We have to tweak that a little bit. Two agencies. Take out the sentence about enforcement and then continue our Beyond HIPAA work focuses on health-related data that is often unprotected whether disclosed or originating outside of the regulated environment.
COHEN: Can I make one suggestion? In that middle sentence, there is a patchwork of laws of federal and state laws that govern the use of agency data. There is a patchwork of federal and state laws that control the use of government agency data or something like that because it is beyond – I just want to get federal in there because it is not just state so somehow you have to work that concept in. I like the word patchwork and I am fine with all the other changes if you can just get that flavor of federal laws with the state.
KLOSS: Does everyone agree that we have sufficient feedback and that we can – we will take it offline. You will see it in the final red line.
STEAD: Or this one may come back sooner than that. I think we have surfaced the fact that it is really important to get to where this communicates the right way, the message the right way. I think we need to take it offline. I am thinking we probably will want to bring it back as a very short email by itself as soon as we have it done and have a chance to make one more iteration if we have to before it goes into the red line. I just think this is a hot – this is really the only thing in the whole report that we are this far apart on. I think we have come a lot together. We need to get together.
BERNSTEIN: I think there is enough representation on the Executive Subcommittee that you will come to a good conclusion.
STEAD: We are operating within the confine of yesterday’s motion. We are working on clarifying what we are trying to say. On this particular case, I think it fits in. We would like to get this clarification run by the Full Committee as quickly as possible and then be rolled into the part that will come later.
KLOSS: Okay. That just kind of gets us into the gestalt that Beyond HIPAA is a very simple, straightforward topic. We are very close to complete consensus on what to do here.
Just to remind you, the purposes of the Beyond HIPAA Initiative were to build on our past work and the work of other government and private initiatives to consider a health data privacy and security framework for the 21st century information challenges, not a small lift. But our specific goals were to identify and describe the changing environment and the risks to privacy and security of confidential health information, highlight promising policies, practices and technology.
And then to lay out integrative models on how best to protect individual’s privacy and security health data outside of HIPAA. Outside the HIPAA protections and at the same time enabling data use. We did not want to strangle this down or lock down data. We want to have a reasonable framework.
Then formulate regulations to the Secretary, but also inform stewards and other stakeholders in the private sector of what we have learned through this work. We have had four goals. And our progress to date since we launched this in 2017 is we did the environmental scan. That was kind of wrapped up at the end of 2017, published early 2018.
Last year we dabbled with developing or beginning to frame – develop a framework. We did so by starting to look at some what we called exemplars, use cases that were at the intersection of the regulate and unregulated world. What happens, if you will, when covered entities hand off data? What happens when health data is generated outside the HIPAA world and then pass it back or held outside the HIPAA world?
We looked at three use cases and the next block shows those. We talked through and did some research on registry models. We talked about personal health devices and did some research and description of the issues there. And we looked at personal and other kinds of apps that handle health data.
Then we stopped and said let’s not drill too far into a single use case. Let’s try to come back up to our goal, which was laying out an integrative model. That is really where we are right now. We are looking in the gold box to try to lay out a framework that may serve as a contemporary vision of this.
We then struck on the idea of looking back to previous work of the Committee where the Committee had done pretty extensive work of enhancing protections for uses of health data, a stewardship framework. We looked at this past work from 2008 by the Committee. I was not on the Committee at the time, but I recall this work. What triggered it was the emergence of personal health records. Some of you will remember that where Apple and Google and Microsoft and everybody was going to jump into this personal health record world. We were worried and concerned because those were not going to be covered by HIPAA. I think that probably was the proximal trigger for this work.
When you read through this report, it does speak to the issue Beyond HIPAA, but it was a different world ten years ago and certainly the Beyond HIPAA challenges are far greater today. Still, we think that this would be useful to contemporize past work and roll it forward. I think we kind of settled on we are not deliberately updating this report, but we are using its outline and its framing of developing a document that can be summarized for policymakers and can also be useful to stakeholders in the private sector who after all – we are not going to cover all of this through laws. We are not even suggesting that that is a reasonable framework because we need to balance protection versus innovation. We do think some stepping up some variety of mechanisms is probably what we are going to capture here.
This diagram is kind of complicated, but it reflects where we are at today. I will just unpeel this one more time for you. I know the Committee has seen this, but bear with me please. Across the two red blocks, we see that this stewardship is really a continuum because data moves back and forth, in and out between a protected state, an unregulated state. When it is in the hands of the covered entities, we focus on compliance, compliance with the rules. When health data is not covered by HIPAA then we have issues relating how it is being used, who it is being disclosed to, re-disclosure, other kinds of risk mitigation. We have kind of a different threat picture or risk picture, different lens on, if you will.
What we looked at was a continuum from the far left, stepping up the protections that are based on HIPAA and are the domain of covered entities and business associates through both a change to the HIPAA laws, regulations, or sub-regulatory guidance and private sector initiatives to do a better stewardship job. But the far left, we are really looking at the world we know as HIPAA-covered entities and business use.
For example, mechanisms could be that HIPAA-covered entities require data sharing and data use agreements before releasing PHI. That is customarily the case, but it is not always the case because we identified that there are registries that get patient data that are not governed by business associate agreements. There were some things that we identified through our looking around that seemed like areas that could be tightened down using current protections.
That covered entities could strengthen their risk management practices particularly with regard to de-identification policies that it cannot any longer be, as we pointed out, in our de-id letter, release and forget, which is kind of often the case. De-identified release and forget. Does the covered entity and business associate – should they have an obligation, a stewardship obligation that goes beyond that?
We talked about covered entities could improve patient transparency regarding uses and disclosures. We certainly heard from Jackie and others that – and in our testimony that people just want to know what their data is being used for. We have not done a good job on that.
I think the whole initiative – something else I probably would add to this today is stepping up the responsiveness of getting data to patients when they ask for it and doing it right and doing it in a timely fashion so enforcing the requirements we now have.
But there is also a public answer here that federal expansion definitions of business associates could occur. There could be additional action by FDA with regard to approved devices, covering not just security, but also privacy protections. What we did this year was essentially identify a range of public and private actions that could be taken in three spaces by covered entities at the far right hand through going much further and going in the direction of enacting data protections like we see in California and under consideration by other states and through the European Union.
And then in this middle space where we have a lot of this Beyond HIPAA data living and being managed by data holders, an array of data holders with very uneven agreements and requirements.
Against this background, what we want to do then is take this old 2007/2008 document, create a 2019 version, a ten-year on version. It likely will be directionally consistent, but the issues have exploded in complexity. We want to address the regulated and unregulated world. Some process parallels though are major environmental scan really sheds new light on it and covers. I think the benefit of that environmental scan is it really does pull together all of the resources on these topics that were available as of the end of 2017. And be principle based. I think we have some characteristics, if you will, of the workload ahead.
That led us to the design of what are we going to do with this March meeting. We went through three design iterations for this to settle finally on a small group working to produce a very tangible product by the end of a day and a half. I started by thinking well we do a roundtable like we did with vocabularies and terminologies and we will bring in a host of people and we will come out of this with a model.
But we decided that we are far enough along so we ought to just be having a smaller working group roll up our sleeves and put some stakes in the ground. That what we do not need to do is surface 40 other perspectives. We need to pull our current thinking together into a working model.
Our agenda is pretty straightforward. We intend to outline principles for stewardship of health data in the environment described in the scan, identify the essential public and private levers to ensure appropriate governance, develop recommendations for a contemporary framework, identify key themes for communication and identify what seems to be the area where we can draft some recommendations for the Secretary. We are going to do that through a one-day working meeting with invited expert guests to work along with us and then another half day or two-thirds day with the Subcommittee, taking the output from Day 1 and coming out of it with an outline for this report.
We extended invitations to Barbara Evans, Leslie Francis, Melissa Goldstein, Sallie Milam, and Mark Rothstein. They have all accepted. We have our brain trust on deck to attend this meeting. They are going to roll up their sleeves and help us produce a report with output.
In that report, of course, as I said, I want to say that again. It is for stakeholders. It is for policymakers and it is for agencies.
What we want with the time remaining today is to pick your brain on some things, help us flesh out a little bit what we are going into this meeting with.
Against that background, this is what we came out of the environmental scan with. We have these two spheres, the existing and emergency policy frameworks, growth of HIT, this matrix of laws. I guess it is patchwork and changing consumer attitudes. Those were some of the conclusion themes from the environmental scan.
What we would like you to do is think about this draft statement of the current problem and helps us refine that or raise questions. Does that state where we want to be?
PHILLIPS: Linda, I know that explicitly in here have avoided Congress as a target audience. Given our being named in the actual HIPAA Act, I wonder, one, if that should be expressly stated and second if the input we get should be directed by some of the folks who either advise that or the 21st Century Cares Act or currently are advising Congress on data protections and security to make sure that that audience’s interest and their awareness of our charge are understood.
KLOSS: I think that is a really good question and I think that is one of the reasons why we feel a sense of urgency to pull this thinking together into a report that can affect some of the current thinking in the data protection and other privacy laws’ spheres. I think this is a time – I do not know.
I think we thought we needed a product and then we can dialogue with those policymakers that are working in this regard.
COUSSOULE: I think it will become important to think through and try to guide what we expect in a little more detail the outcome to be. Is it a very prescriptive or is it more of identifying and codifying risks? The question of how much do we frame up and is the report – is the output of this going to be an informative document or a direction-setting document? I think it is important to try that because that will help set the expectations of what we are going to end up with at the end of the day.
KLOSS: I would answer that as more direction because I think we have done the laying out of issues through the environmental scan and other work. I think we are planning to put a stake in the ground. This is a reasonable framework.
MAYS: I think there are a couple of things that are probably still informational. To me, the big one is this unregulated space and its relationship to HIPAA or not being related to HIPAA in some ways, but yet having implications for health. I think that could be both informational followed by it being directional. But I think that is a critical space right now.
Those of us who sit at the university, we have students who are suicidal, but because of FERPA, there are things that are very difficult for us to do. There is difficulty in getting data. It is an interesting concept in terms of some of the walls that exist around certain kinds of data that have health implications. I think really laying that out would be good.
And then I think the other, which can be more directional, is to begin to talk about the relationship between technology-like apps and the integration of that into health care and the kinds of protection. That people can realize what it is that they need, but consumers really want to do this more so than I think some of the care facilities do. I think getting a consumer perspective on how do we do that and what needs to be the direction to make it happen in a secure and protective way is really critical right now.
KLOSS: Bob makes such an interesting point about congressional. I think we are one step before that, but I certainly think that we could come out of this with a recommendation that we find a way to brief people who are working on setting policy because they may be, again, not appreciating the kind of nuance that we brought out in the environmental scan.
PHILLIPS: It would be interesting to understand appetite, whether there is a current interest in this in Congress, whether they know that we are advisory to them in this regard. Sometimes if you can create that relationship then when you deliver your final product, there is more receptivity to it without politicizing it.
KLOSS: That is why we laid some of the groundwork on that in this Report to Congress, but we are advisory to the Secretary. We kind of have to do it through our Report to Congress, not directly, I think. I do not know.
STEAD: I think if we can – one, we are obviously still struggling a bit with the final language with the Report to Congress, but we will and that airplane. We are going to try and get that out pretty quickly. Although we have toned down the privacy pieces and the call for action, we have toned them down to clarify that they will take some investigation. They are not something people just know how to do. I think that is how we begin the conversation with congressional staff and Congress to the degree we can get congressional staff interested.
I think at the other end of the spectrum if we can get the recommendations clear that we are ready to make after the hearing to the Secretary. We are working two very different lenses. That is two different levels, two different scales. That would be clear and ready to go.
What I am thinking I am understanding is that the rewrite of the narrative report would be the vehicle you are thinking would provide the next degree of guidance. I am not personally clear whether that will be a bridge too far from what we come out of March with or not. My gut is that it might be. I think if we really get the recommendations, some actual recommendations to the Secretary, that is, to me – if we can get that, if we can get more that would be good. I think it is that second thing that would play back in I believe to what Bob is suggesting around how we would then engage about it. But it is almost three things that build on each other. Does that compute, Bob, or am I in left field?
PHILLIPS: I do not think you are in left field. I am just concerned that they never see it or recognize it that we are advising them on it unless we pull them into the conversation about what are they looking for. What would be useful to them partly to give us guidance, but I think more importantly to make them aware. I think if we go back and look at the folks who sponsored the 21st Century Cures Act, we might start to hit on who are the key legislators and who are some of the key staff who have an interest area in this.
STEAD: Absolutely and personally I plan to make sure that Lamar Alexander’s staff understands the Report to Congress. I can do that as an individual. That was one of the interested parties in 21st Century Cures. I think as we all work at how we market or get uptake of the Report to Congress, we need to think which of those levers do we collectively have and how do we use them to have the useful conversations that may then give us guidance. We cannot ask them to come tell us what they need. I do not think that is within our charter.
LANDEN: One of the things I would like to see us take a look at and I am not sure it is technically part of the problem discussion, but certainly in our analysis, our learning curve and then ideally in the recommendations. We should take a look at what internationally is going on with specific reference to Canada. I know they were at some of our hearings, but both federally and provincially, they have a different approach to privacy than we do. And of course the other big knot is the European GDPR.
I think one of the ideal outcomes and our recommendation should be something where the US is again ideally compatible with requirements of Europe and Canada.
LOVE: I concur, Rich. Going to Congress, I do not know what we would ask. I am always afraid of the sledge hammer when maybe we have a more toolbox approach.
I have to go back to the 2007 report because I did read it, but I forget what is in there. But when I think of these issues, I really put buckets. There is a legal wish list solution in addition to alignment with international laws. But I have often thought there should be a uniform law across states. Some model language for uniform law and that might get at some of the unregulated data sets, but surely will help some of our state and local data sets immensely. I just do not know how to make it happen.
Technical. It goes hand in hand with the legal. We just need to update our de-identification and encryption and some of the cloud technologies that are available today that were not available in 2007.
And then we have this political and cultural. That is a little bit of my fear of bringing in legislatures before there is some thinking and solutions because the sledge hammer approach could create more issues. I just feel that strategies need to be intertwined legal, technical, and then the political let alone the cultural shift that has to happen in sharing data. And public health needs some modernization as well. That is my two cents for solutions.
PARTICIPANT: Frank, I hope you will feel free to jump in if you feel like it.
PASQUALES: I guess the only thing intervention I want to make at this point was that as I worked in this area, one of the things that I found very useful in the privacy and security literature is a typology of threat scenarios. Getting a report like this – sometimes these matters of data and data management and policy can be extremely abstract and people — it is just the Mago affect. My eyes glaze over. To sort of be able to give some people who are thinking about, the policymakers, examples of some very troubling developments or developments that need some attention I think would really help.
I think, for example, Sharona Hoffman’s work on employment scores, the potential for business advantage by having a score that would predict who would be a very high cost employee. That is a natural concomitant of employer-sponsored insurance. I think we really have to think about that. Life insurance connected to Fitbits. A new proposal in New York with respect to your life insurance connected to your Facebook profile where again there are inferences of health status or predictions of health status made by that. Very concrete examples.
Or the story of a book called Girl in Glass about a particular employee of AOL who – she was essentially outed because there was supposedly data that was general, but that it said that the CEO of the company said nobody is getting bonuses because there are some people with distressed baby cost us like $2 million or $3 million. She wrote a whole book about this.
I think giving very concrete examples of invasions of privacy that led to very troubling outcomes I think would help ground some of the public concern and some policy interest in the report.
KLOSS: We do have a number of good rich examples in our environmental scan, but that is a great idea.
BERNSTEIN: Listening to the discussion and I appreciate having Frank’s perspective that we are lucky enough to have today and really I am going to miss him in March.
I am fortunate to have a copy of the slides here that was circulated. I wonder if we could back up one and try to pump up a little because I think one of the things we wanted to talk about with the Committee was the general themes and so forth. I think we may have skipped that. I did not know if you wanted to say something about that, Linda, so that we could try to get the Committee’s view on the major themes that we want to discuss in March and help us with the statement of the problem rather than to dig into the particular recommendations that might come out of that. I think that would be helpful for the group to discuss if you agree.
KLOSS: Thank you. These are the themes that are in the 2007 work.
SEEGER: Specifically, it would be helpful to get folks’ input in the 2007 work. The issues identified at the time were treatment payment and health care operations, quality measurement, reporting, and improvement, research and public health. It would be useful to get all of your perspectives whether those are still some of the standout issues, if there are other issues today.
The landscape is very different today in 2019 than it was in 2007. But some of these issues are still perhaps front and center.
BERNSTEIN: You might frame them differently or you might call out some of the things that Frank was talking about. I wonder if you think about at that level what kind of themes you would want to see the Subcommittee address.
GOSS: Rachel, this is Alix. In regards to our emails sent yesterday, I really feel that we need to more clearly call out social determinants of health data.
BERNSTEIN: Could you say more?
GOSS: I think if we start to look at the whole person and understand food insecurities, housing insecurities, transportation aspects that enables us to get full – more fully achieve the health outcomes that we seek so the quality and well-being of individuals. There are lots of different kinds of data that Linda noted earlier kind of go in and out of being regulated or unregulated, but become part of the full view of the citizen’s health status. We need to think about what does it mean to the data standards, the data flow, the data handling aspects when it is not just your traditional what we might think of as the medical or pharmacy type claims data.
COHEN: So sort of building on that, clearly my interest is public health, but personal. I think Alix is talking about the intersection of personal health and public health. There is a slight distinction, but certainly the explosion of data that might not be protected about personal health as important.
The previous conversation has gotten me thinking about how this discussion can help inform what I am thinking about as the gray area where there might be policies developed, but that could be enormously strengthened and made more consistent across organizations that have attempted to provide some protection, but have not thought it through strategically or would benefit enormously by a federal guidance for more consistent approaches. That is the piece that I was trying to identify in public health. Non-covered portions of public health would certainly benefit through a focused lens on the extension of these themes.
Just popped into my mind. Barrack Obama’s birth certificate. It is a use case of a very private individual, health-related data source that was protected by or attempted to be protected by a public health department regulation. I am thinking of other cases where individuals have been identified in a public health context through the lack of secure protections that might be addressed through the work in March.
KLOSS: You are saying that of this list of specific users that was addressed before public health remains. Research certainly remains.
COHEN: Yes, clearly.
KLOSS: Quality management reporting is not quite the burning platform today. It is? It still is? Everything that is here is still a burning platform.
PARTICIPANT: That absolutely is just because of the way the payment models are changing.
KLOSS: We may frame how we describe it and then we have no use cases to add to this, the personal health apps.
STEAD: I think one of the things that has changed is the way these things merge and influence one another. Alix raised social and behavioral determinants and the intersection of that public health. The increasing understanding that the separation of mental health and medical health is not biologically sound and therefore an approach to privacy management that involves segregation is not probably the right approach.
I think in a way the points with the others are similar. The merger of Internet of Things, which – activity and all other sorts of things with personal health and public health.
I think the big difference between 2018 and 2007 is in 2007, you could sort of think of these as discrete. I think now we understand that you need to have one aggregate data capability that provides different lenses, but in fact, those lenses are informed by the collective set.
You are almost going to have to shift the privacy and security protection paradigm from the data to the purpose of the use. It is going to somehow have to be done in a way that lets the computer operate across those things under appropriate protection for inappropriate use.
In 2007, we still thought a perimeter protection model might work in some way, shape, or form. Nobody that I know of think that will work today. I think the shifts are much more fundamental, which is why I am not sure fixing this report may not turn out to be a bridge too far for March. It may, in fact, be the bridge that comes after March. I do not know if that is helpful.
KLOSS: But we will see. I think we can proceed on that assumption. I keep going back to the environmental scan that we have done most of the descriptive work. We need to focus on what the policy alternatives are or the practice improvements are. I think we can be kind of short in describing it. That is not easy to describe the scope of the problem. But I do think we need to focus on what are the ways forward and not go back and try to redo the environmental scan. We need to work at the level you are talking about.
STEAD: If anything I suggested said to redo the environmental scan that was not my intent.
KLOSS: I think spending too much time defining the use cases does not seem to me to be —
STEAD: I was not talking about defining the use cases. What I am saying is the protection. What you now need to do is consider the purpose of use in your threat and risk model because it is less a use case than it is a purpose. Bruce taught me this early in my population health tenure. The purpose is the key question that a lot of other stuff close from. Purpose of the use.
And what we have done is shift from the purpose of the capture and the purpose of the use and therefore the purpose of the management were somehow demarcated to where that is no longer true.
LANDEN: In thinking about the differences between 2007 and 2019, one of the aspects that occurs to me that a lot has changed since then and the origins or the sources of data have increased I would guess exponentially as have the destinations to which the data goes and specifically I am thinking in terms of what 12 years ago was probably a single, low double digit penetration rate of EHRs and EMRs. Now we have nearly 100 percent of hospitals and physicians have the electronic medical records. We have had a half a dozen years of very highly visible federal policy promoting interoperability.
We also have patient access to that data, the Internet of Things. Both are for consumption and the production of data. We have the consumers in the mix now, none of which we had back in 2007. I think that is another factor we need to look at. It is just the tremendous change in the scale of what we are doing and our ability to fluidly move the data to many different points.
GOSS: I have a question about whether or not the team that is preparing for this March event has been able to weave in the work that OCR has been doing in their recent request for information.
KLOSS: Have not to date, but certainly are looking at those subjects that are raised.
PHILLIPS: One last plea. Our charter says in three places that we are advisory to Congress on Title IX of the Social Security Act specifically Part C around HIPAA. We have congressionally appointed members of the Committee. I just think it would be helpful to know what they need. It would be helpful to know how they need it framed to Frank’s point. And it would be helpful to know if there is a vehicle moving that might carry some of this work, not to write legislation. I hate to see another 21st Century Cures bill go through as a vehicle to could carry some of these changes and not know what was happening and not having had our very important informative work considered in that context.
MAYS: It is my understanding that HHS has an office – I think it is like governmental affairs or something like that. If you wanted to take the work of this, I think you would have to go that way. If Congress asks for things that is a different story. I do not think that what we are supposed to do is to go beyond the Secretary with reporting things. Congress can always ask us, but I think otherwise you have a rep that does this.
PHILLIPS: Our charter actually directs us to do it annually. I am just saying that there is a stipulation in our charter for direct.
STEAD: That is the stipulation for the Report to Congress.
BERNSTEIN: I am thinking about what Bob is saying and trying to figure out how to incorporate it. The Committee advises the Secretary and we do have the opportunity to make the Report to Congress about HIPAA every year. By the way, I think of it is we compile our work. We collect it once a year, but we report it to the Congress every two years. But we do report annually.
As a longtime bureaucrat, my bias is that before we go to the Congress, we try to find the least fortune requiring way to do something. If we do not have to go to get legislation if we can do it by regulation and we do not have to go ask for regulation if we could do it by some kind of sub-regulatory guidance and under that – a memo from the Secretary or just tweaking the administrative bureaucracy in a way that improves things.
I hesitate a little bit about thinking about going to the Congress before fully exploring all the possibilities of things that we might do. Certainly, it is the case that the Committee can recommend to the Secretary that legislation can be generated from the administration or it can be generated from the Hill. You can recommend to the Secretary that he promote legislation to do whatever it is you want to recommend. And that is a way to get to that through to the Congress. But in general, we try to do things at the level that it cost the least bureaucratically. Others made comments about what happens when things go to the Hill. They do not always get it right, but it is also hard for agencies to get things right. That is why we go out for comment and we try to change our regulations. We think about what has happened to them over time.
I think OCR has been particularly with its RFI trying to figure out – we have some experience now with this important law, what is working, what is not, what kind of tweaks can we make or major changes that we need to make. I think the Committee can help in that way. I do not want to discourage us from thinking about what might change legislatively, but encourage you to think about what is smaller, less – smaller hammers if you want or smaller tools you might use to affect the change that you want to see.
LOVE: You articulated nicely. I would just thinking about 2007 and what might have changed in a use case that could carry us into this next report. I just am thinking, and this may be just silly me, the use case – I think back if we had our act together in data exchange and privacy together, could we have seen the opioid crisis before it hit us over the head? Thinking of the future. How do we integrate our data and data flows and data systems smartly and not have all these silos that I fight every day, but also not just free for all either, but meaningful exchange that we can see these – this is one emergency. We are going to have others too and can we do a better job in surveying those and not letting privacy be the excuse.
I have too many privacy officers that just say no because that is the risk-free, legal, consistent solution, but it does not serve the public good.
KLOSS: We have five more minutes. I just want us to look one more time at the goals for this meeting and see if you can suggest any ways that we might sharpen them. We are talking about outlining principles for stewardship of health data in the environment described in the scan and the essential public and private levers and we mean that public is regulations or laws or sub-regulatory and private levers to ensure more appropriate governance and use.
Two, develop recommendations for a contemporary framework of data stewardship including a pathway for improving private and public sector governance.
Themes for communication with policymakers and stakeholders. This conversation that Bob spent talking about as well as a conversation with current covered entities and other data holders that are outside of HIPAA.
And targeted recommendations for the Secretary.
MAYS: I thought about this when Denise was just talking about what the next crisis is. One of the things that has come up is the issue of sharing data during disasters. I do not know if at all we want to go there in terms of that. But it really becomes very integrated in terms of being able to get your pharmacy data and being able to have people who speak on your behalf to be able to get access to that and whether or not we have sufficient both laws or policies and protections in place during disasters.
PASQUALES: I just also wanted to add in terms of aspects of laws that block data that could be used for very valuable purposes. I do think the trade secrecy is very important there as well. We often hear about privacy as being sort of the big blockage, but I think a lot of literature in pharmaceuticals, other areas show that trade secrecy is a huge issue as well.
KLOSS: Anything else for the good of the cause?
HINES: There is, Linda, and that is your thoughts on the March meeting. Are you thinking that an invitation go to the Full Committee? What is your thought on that?
KLOSS: I think we always have had these sessions open to those who can make the time available. It is always beneficial.
HINES: Just to clarify because a member asked recently in the last hour. You all should have an Outlook Invite from Marietta, but we will just send another reminder to accept that if you plan to come just for planning purposes, members. Right now, we are planning to have the meeting at NCHS. We are going to look one more time to see if anything has opened up in this building.
KLOSS: I will say that our invited guests will be working with us on that Thursday and then Committee members will continue the work for two-thirds of the – until 3 o’clock on Friday. We want to pick their brain and then we want to dive into developing recommendations and work product. We would appreciate if you can come to come for both days.
It is fitting that we are ending this meeting with Beyond HIPAA because it is probably certain that you will begin the June meeting with Beyond HIPAA.
STEAD: Thank you, Linda. I think we have reached the time for public comment.
Agenda Item: Public Comment
HINES: There are no comments that have come in on the WebEx as of now. The WebEx viewers are seeing the information to submit public comment either by the WebEx Dashboard or to the NCVHS mailbox. The website is there. Is there anyone in the room present? No one in the room present to make public comment. We will wait 60 more seconds, if you have not sent it in and you had meant to.
While we are doing that, I would just like to say what an incredible privilege and delight it has been to work with Bruce and Linda. I know you are still with us until June 1, but this is our last Full Committee meeting. And although I can no longer claim I am new to the Committee, just in my growing into this role, you two have played really just fundamental roles in your respective areas of expertise. It just has been a wonderful experience.
COHEN: Thanks. It has been a wonderful ride. It has been fun and I have learned so much from all of you. It has just been a really enriching experience in my personal and professional life. Thank you all.
KLOSS: I would echo that. I have learned so much and it has just been a great capstone to my career. We have grown old together, Bruce.
COHEN: What happens at NCVHS stays.
STEAD: I prefer to think that NCVHS has kept you young.
HINES: One more pass at the WebEx. Any public comments? I think this is one of the few times when we actually had no public comment.
For the good of the order, anything else at the table?
STEAD: We are adjourned. Thank you.
(Whereupon, the meeting was adjourned at 2:50 p.m.)