[This Transcript is Unedited]
Department of Health and Human Services
National Committee on Vital Health and Statistics (NCVHS)
Full Committee Meeting
May 28, 2015
Hubert Humphrey Building
200 Independence Ave., SW
Washington, D.C. 20024
TABLE OF CONTENTS
- Welcome – Walter Suarez
- Subcommittee on Population Health – William Stead, Bruce Cohen
- ACTION—Community Engagement Letter – Walter Suarez
- E-Vitals Demonstration Project
Speakers: Delton Atkinson, Michelle Williamson, John Stamm, Mike Gahagen, Jacob Tripp, Leisa Finch, Jeffery Duncan, Daniel Friedman - Working Group on Data Access and Use – Vickie Mays
P R O C E E D I N G S (8:00 a.m.)
DR. SUAREZ: Good morning everyone, I think we should get started. This is the second day of our National Committee on Vital and Health Statistics meeting, and this morning we are going to have a meeting of the Population Health Subcommittee. After that, we will have a full session of the Committee with a couple action items and some discussion about next steps. We will have actually a demo of the E-Vitals initiative, too, so we have a full morning agenda.
I will start introductions. My name is Walter Suarez; I am a member of the Population Health Subcommittee and Chair of the National Committee, and I do not have any conflicts.
MS. JACKSON: Debbie Jackson, Acting Executive Secretary NCHS, CDC.
DR. MAYS: Vickie Mays, University of California, Los Angeles, member of the full Committee, Privacy, POP Health, and I have no conflicts.
MR. SOONTHORNSIMA: Ob Soonthornsima, CVS Health, member of the full Committee, Subcommittee on Standards, no conflicts.
MS. GOSS: Alix Goss, member of the full Committee, member of the Standards Subcommittee.
MS. KANAAN: Susan Kanaan, writer for the Committee.
DR. CORNELIUS: Llewellyn Cornelius, University of Maryland, School of Social Work, member of the full Committee and the Population Health Subcommittee, no conflicts.
DR. COHEN: Bruce Cohen, member of the full Committee, Co-Chair of the Population Health Subcommittee, member of the Data Workgroup, no conflicts.
DR. STEAD: Bill Stead, Vanderbilt University, member of the full Committee, Co-Chair of Pop Health, no conflicts.
MS. SQUIRE: Marietta Squire, CDC, NCHS.
MS. WILLIAMSON: Michelle Williamson, CDC, NCHS and staff to the Standards Subcommittee.
MS. KHAN: Hetty Khan, CDC, NCHS, staff to the Subcommittee on Privacy, Confidentiality and Security.
MS. JONES: Katherine Jones, CDC, National Center for Health Statistics and staff to the Committee.
DR. FRANCIS: Leslie Francis, University of Utah, former member of the Committee and member of the Working Group on Data Access and Use.
MR. TRIPP: Jacob Tripp, Intermountain Healthcare.
MS. LOVE: Denise Love, National Association of Health Data Organization, member of the full Committee and Standards Subcommittee. No conflict.
DR. SUAREZ: Do we have any members on the phone at this point?
DR. WALKER: Jim Walker, member of the full Committee.
(Introductions around the room.)
DR. SUAREZ: Okay, we are going to get started. I will turn it over to our esteemed co-chairs of the Population Health Subcommittee.
MR. SCANLON: Good morning, Jim Scanlon, Deputy Assistant Secretary for Planning and Evaluation and Executive Director for the full Committee.
Agenda Item: Subcommittee on Population Health
DR. STEAD: Good morning. We wanted to begin this block by discussing a proposed convening activity that would be tentatively targeted for the fall, but we will have to confirm that once we better understand the activity.
After the process we used in the February meeting of getting people to prioritize the potential next steps that came out of the roundtable, we narrowed first to three choices and then, in the last Pop Health call, to one. What we are proposing is that we focus on this idea of a standard set of core metrics that would be used across all population levels and across all components of the health system, broadly defined, to identify opportunity and to track problems.
In the agenda book you have the Executive Summary of the Institute of Medicine Report on Vital Signs that came out in April. That report we think would be the basis, the fact base if you will, that we would use to ground this activity.
We will highlight at a very high level key aspects that we think will be helpful from that report in our thinking this morning. Then the questions that we want to try to answer are who is the target audience, and which of the recommendations should be our priority. Those two sort of relate if we follow the structure of the IOM report because, in essence, there were separate recommendations for the different target audiences. And then which domains or core measures should we focus on, and finally, the meat — how should we do this and who should be at the table for the initial look into this.
Also, just to remind you, both Larry Greene and Paul Tang were members of the IOM committee, so we have people who are familiar with our work who could help us in some way.
This is put in front of you simply to remind you of the breadth of the thinking about what a relatively small set of core measures could handle. In the report you will remember what they have really done is frame the key domains, a little bit like our conversation yesterday, and then went from that all the way down to the current best measure in each key element of that domain and the current national performance off of that measure.
These were the potential target audiences identified in the report. Obviously, the federal government, and I think it is particularly relevant to NCVHS, particularly given yesterday’s conversation, because the report really does view the Secretary of HHS as the logical person to try to coordinate the very complex activity that would be required to move in this direction. But then work your way down to governors, mayors, health leaders in the communities, the clinicians and the health delivery system, employers, community leaders and then, obviously, everybody because if we are going to move in this direction it really would be a national effort.
We could pick one of these target audiences or we could pick several. If you think about communities, as we have been working them, they tend to at least relate to three of those bullets, so we would have some flexibility in that.
Just to remind you of the recommendations, we pulled them out and grouped them in that order. Three of them related to the federal government; all are referencing the Secretary of HHS and, in essence, say that the core measures should be incorporated into the current federally administrated programs in that process, standardizing and eliminating duplication. They should then work with others to have these measures used in their programs, and finally, a mechanism for developing and maintaining and evolving the measures.
The recommendations for the governors, mayors and health leaders were that they should use them to tailor a measurement dashboard to their setting and to use it to drive a focus on outcomes and programs to improve outcomes.
For healthcare delivery organizations, they should both use these measures to assess their progress and work with others across the communities they serve to collaboratively improve performance. Community leaders, much the same idea — use these measures to drive their purchasing decisions, efforts around transparency, et cetera.
And for all individuals, this would be a public education and bottom-up use effort to take an active role in shaping their own health.
At least for a nanosecond, I believe the IOM report provides a pretty complete fact base. We didn’t send it to you, but Appendices B, C and D are exhaustive, and B lists the existing reporting requirements; C catalogs existing data sources, and then D does a comparison of existing data infrastructure in addition to itemizing them. If we move quickly, we do not need to rebuild the fact base, I believe.
With that introduction, back to the questions. But first, Bruce, would you like to expand on any part of that introduction?
DR. COHEN: No; that was great. Thanks.
DR. STEAD: I think the first thing we need to say is, as the Pop Health Subcommittee, we have had three calls between the two meetings where we converged or focused in on this as the convening activity. I guess the first question is just to confirm amongst the group that this is, in fact, the highest priority amongst the things Pop Health could be worried about for us to focus on. Are people comfortable with that?
DR. SUAREZ: Could you define this as the focus, this being the IOM report?
DR. STEAD: No. I think the IOM report is a fact base and a set of recommendations that we can elect to pick from to construct our activity. I think it is a foundation for the activity. I think our activity, I believe, is focused in on a relatively speedy — because I think it is important — a relatively speedy letter to the Secretary on, given this huge report, where should she focus. The purpose of this block this morning, Walter, is actually to answer your question, so I’m straw person’ing.
The second straw person, and maybe they could be together or maybe they would be two sequential efforts, would be how do we — if you read the report, everything through it is that this is going to be a journey, not an event. This first set of metrics has a pretty broad basis but there are many that need to be further defined. The Committee did not try to do that. They are really envisioning a top-down, guided bottom-up measurement refinement process I believe if I understand this correctly.
So it would be logical, given what we have done, for us to convene communities and understand that journey in essence with them in a way that would then come back and inform what we should say to the Secretary about what the Secretary or HHS broadly could do to help with that piece.
I think those would be two possible frames, but I think we need to turn — our goal here was to have an open discussion of these alternatives.
DR. SUAREZ: So we have a few people and we will start with Bruce and Jim.
DR. COHEN: Historically, the Committee has struggled with the fact of should we get involved in this effort. The good news is IOM has done an exhaustive project that reviewed all the work in this area, and they have added their expertise and have come up with recommendations for a primary set of core indicators and a secondary set of indicators. So that is not our job.
Our job really I think is to figure out how to make recommendations to the federal government to implement these indicators, whether it is through additional surveys or technical assistance or creating these standards. So life expectancy can be done at the community level; it can be done at the county level; it can be done at the state level and it can be done at the federal level as guidance for a variety of different programs.
I think building on these activities will help us provide direction to the Secretary, and this is I think an opportunity that we should really seize and build on.
DR. SUAREZ: So you are referring to the focus would be helping to find the how, not the what. The what has been identified by IOM. The how is sort of what are the better methods.
DR. COHEN: Yes. I think the how rather than the what, although there are options for the what. There is a core set of 20 indicators and another supplemental set of another 40 indicators that support these general domains. We can decide if we feel that there are indicators that should be elevated. They tried to be very judicious around selecting indicators that were measurable and actionable to populate the different domains, although clearly the data do not exist for some of the indicators.
So we can get into that area, but I think most of the indicators — well, they use some classic examples. Paul Tang and I always talked about the definition of smoking. There are 61 different measures of smoking cessation used in a variety of surveys. If we are recommending a standard, here are the options that folks should focus on, and if they need to use something else for another reason, that’s great, but here is the consensus on how to measure smoking cessation at the clinical level and at the community level. They have done a lot of that work. We just need to push the ball down the field about implementation. Thank you.
MR. SCANLON: I like this direction conceptually; it is a good thing for the Committee to be doing. I do think it is not just the how question.
These recommendations are made in sort of an ecosystem where a lot of other people have tried to do this and continue to try to do this. I would not consider them sort of in the absolute, so I think any convening would involve at least some discussion of that. You have Healthy People objectives; you have other efforts at core indicators; you have warehouses and so on.
I think number one should be — I have the greatest respect for the IOM, and they are probably going to have their own follow-up. But I think we are trying to see how it fits into the overall health statistics and community and national ecosystem for health data. It is a good place to start to get reaction to what they are.
There are other ways to approach core and supplemental; there are menu type approaches. People may or may not agree that this alone or maybe with modifications. I would start there probably and then see where it takes us.
I would not have the data producers there, to be honest. This data is not going to materialize unless we sort of know what is realistic and what is not. And where really is the community level data? Other than Vital Statistics and administrative data there is not much.
The Committee I think can bring the view of a convener, of sort of a critical look at how does this fit into the ecosystem. Then what would make sense — and again, I am not suggesting that we discard what the IOM did, but we have done this before. I would like to get a sophisticated and sort of a nuanced look at what would make sense to move forward.
I mean this in a very positive way. I don’t know that these are the 20 that anybody would want to use. In many cases, for our own programs, for example, they have much more specific objectives, and having general objectives would not help them. Mental health, substance abuse, delivery, access issues — some of these just would not be as helpful for a program as something more specific. And they have many of them.
The whole quality area, as you know better than I, Bill, we have plenty of measures. I think we need something to align those measures. Picking one I don’t think is the answer.
Anyway, I like the direction, but I would say how does this fit into the overall ecosystem, and then what makes sense to proceed, and for whom. That is my thought. I really like this direction, though. This is a classic population health issue.
DR. MAYS: I am actually very close to what Jim was saying in terms of what I was going to say.
I think what the IOM did was good. There has been discussion about the results of this, and the question is how does it really map onto what has to be done relative to HHS. There is some discussion about whether or not these were the perfect set, how well they fit, whether or not they are mapped onto all the populations that we have to worry about.
HHS has kind of stepped up to the plate to really begin to talk about some of the populations that don’t end up there, and there have been questions about whether or not this is as inclusive as it needs to be.
I think starting with it is great, but I think there still needs to be a preliminary set of hearings or analysis to say how did this line up with what we need to do within HHS in terms of how it is we collect data and how it is that we move forward relative to the demands of answering these questions for what HHS has prioritized as its populations. I think it was a little bit different than what the IOM did.
I think it is great work but I think it needs some further refinement.
DR. STEAD: Personally, I resonate with all of that. I see this as a very well thought-out fact base. Often, when we want to start a convening effort, the first challenge is what has been done, how do we get the information, and we don’t have the staff to do that work and it has been done for us. So, take it as a foundation, and I think the report said this pretty clearly, that they view whatever happens to this as a political process and not as take it and use it.
I think we are ideally positioned, and I think the way Jim and Vickie worded that resonates well with me. We will need to get it down on paper.
Again, as we think through this, the fall is a period of time. Do we envision, for example, that at the September meeting we could, in fact, have a briefing on the report if that would be useful to the full committee. And then if we can really figure out how we want to do this, maybe then we could be trying to get it done in late October or something, which would give us some time after that. We would obviously have to be pretty well loaded to know how to pull off the thing later in the fall.
Or we could simply have a convening activity worded much as you all are suggesting. It could be done fairly early because we would say this is preliminary. The first thing we have got to do is get key people together to understand what this looks like given the context of HHS and we could include the briefing in that. So I can see a number of ways we could pull this off.
What I am hearing is a direction to focus in on how this fits in with HHS, and getting data producers and so forth — really, an HHS-centric convening activity in many ways, not a community. That is what I’m hearing.
DR. COHEN: To build on Bill’s thoughts, I just wanted to ask Vickie and Jim who do you see should be at the table for these discussions around focusing on HHS programs and coming up with the indicator sets and implementation?
DR. MAYS: First, I think we need to specify the system that we are really talking about. Are we talking about the healthcare system? Are we talking about public health? This is just my thinking. There was an article in JAMA that was really criticizing this and I pulled it up to make sure that it was actually what I was thinking about. I will read just a couple sentences.
Implementation of the committee’s vision for core measures will face challenges. Consistent development is difficult. Many stakeholders are invested in current measures and will not readily agree to modify or abandon them. Some will be disappointed that this report has not provided a fully specific menu of measures that meet all needs and all situations, even though such a menu would likely have encountered fierce criticism from parties not consulted in the development.
So, part of what I am saying is I think the beginning of this needs to be not just a briefing, but it needs to be the landscape, for those who did not drink the Kool Aid and say it was perfect, for what are the issues and how are we going to deal with those issues. To me, that needs to be the beginning planning for this.
You’re starting with the report and saying it is great and wonderful, but I am just saying we need to start with what happened beyond the report in terms of what the field has said about what is missing, kind of what the issues are. And it will be hard, I think, to bring everybody in and have a briefing and then start talking about what the field has said are some of the issues about the report.
I think we need to start, if you ask me — I don’t know that who is the issue yet to start with, but it is the what. Here is what the report said; here is what we think we should cover. I think you are far down the path and we need to just step back and take in the input that occurred after the report was released.
DR. STEAD: I think it would be appropriate to take in the input. I read the article you’re talking about; I read the report. I couldn’t tell any difference between the two. I wasn’t drinking the Kool Aid. I began to read this report in late April, and to my reading, the report says almost all the same things as are in that article. They deliberately did not do those things.
There is nothing about the report that is a prescription. There is nothing about it that could be directly implemented. It is a framing and a fact base that could lead to a journey.
DR. MAYS: That is why I’m saying that is where we should start.
DR. STEAD: Amen. So we need to capture that. So, how to capture that is a statement of what we want to do, and then who do we need to bring in for a first activity.
My sense is unless we get into this and decide the country does not need core metrics, this is, in fact, a good way to standardize. We could arrive at that conclusion. Unless we arrive at that conclusion my suspicion is we are going to have multiple activities before this is done. This would simply be the first of what probably would turn into a series.
So I think what we have really got to do is how do we frame this first one. What would you like to say its purpose is? Maybe it is best to email those to us.
MR. SCANLON: Yes, I am with that. We are not talking about implementing this particular core set, and I don’t think that is what the Academy meant. I think this is meant to be thinking about it. This is how it could work. There are always arguments about what should be the measures and who is going to use them.
I think our focus should be — for the convening — given this concept and given this work and this thought — what would be the best way to assess it and get it discussed. Where does it fit in? What would be the next steps in terms of assessing and evaluating? And I think the IOM expects this. They are a committee; they meet six times, like us. They know that a lot of other work has to be taken to assess what it is and assess its utility.
But I think it is not how do we implement; it’s where do we take this next. What does the community feel? I would include folks who are data producers; I would include the Healthy People, I would include community warehouse, I would include those kinds of folks who have tried to do measures before. I would include CDC on the community assessment, and then I would include the public health community.
I don’t know how far we want to go on quality; we would have to decide that. That is a whole rabbit hole, as you know, of its own, except for access issues and things like that. But I think it would be a very nice convening to have discussions.
Some people will say we are done talking; let’s just do it, and others will say, wait a minute now. We did central indicators before and nobody hears about them anymore. What about Healthy People? Are you sort of pushing us off to the side? What is the gestalt? Healthy People has a gestalt for what it is supposed to be.
I think those are the kinds of questions. Where does it fit in; what is the ecosystem it is into? Are there other ways to approach measures? Are there menus, are there cores, things like that? It is a relatively general level. What about specific program issues? What if you have a specific issue in your community or state? Then these measures probably won’t help you that much.
It has to fit in. I know the appeal of having common measures; it always come up, but you have to know where it fits in and what the limitations are. But I think it could work.
Maybe it would start, Bill, with a briefing to the full committee on what the report is, or maybe people could read it and get enough out of it. Then I would think we’d want to get other folks who didn’t necessarily participate in the discussion from the public health community, from the data producers, from healthcare, and we could see how that works.
DR. STEAD: What I am really hearing is it would be better to have the briefing to the full committee at the September meeting and have the convening activity subsequently so that the convening activity is not too tilted toward the report
DR. SUAREZ: Hearing what Jim points to, and Vickie and others, it seems to me that between now and September the Subcommittee could also flesh out or further define this concept that Jim brings up about sort of stepping back and looking at this more holistically and see where does it fit in the ecosystem — measurement and vital statistics and vital signs.
I think that might be somewhere that the Subcommittee can take it so that when this gets presented in September to the full committee, the presentation from the IOM report can be followed by additional deliberations that the Subcommittee has had as a way to facilitate the full committee in September discussing sort of the next steps.
DR. STEAD: We have already scheduled three Pop Health Sub calls, one for June, July and August so we will have three of those before the September meeting. We could set this briefing at that meeting and we could use the Pop Health time to refine what by then should be a fairly baked plan for the convening activity.
DR. SUAREZ: Perfect. That is exactly what I’m thinking.
DR. STEAD: The thing that we would maybe need your guidance on or we need to know when we would decide along that journey is if this convening activity was going to happen, say, late October, around when we did the roundtable last year, then we would need to actually have it calendar-spaced and people invited by then. We could be refining some of the details of the activity but we couldn’t be starting from scratch in September. If we are starting from scratch in September, then you’re talking about a first quarter 2016 activity, not a fourth quarter 2015 activity.
I don’t know what your plan is. I am hoping we will get to regularly scheduled Executive Committee calls so we could have this discussion, and we can use a little more time here to try to refine this. We can have the June Pop Health call, and then maybe that would get us to the point that at a July Executive Committee call we could pull the trigger on yes, we are going to do it, and calendar it and so forth, or that we are going to hold until next year.
I am just trying to be realistic about how we manage this.
DR. SUAREZ: We do have actually the November 16 National Committee meeting, so we have the September-November full committee, and if there is a need for convening a session after September we could actually tag it to the full National Committee meeting in November.
DR. MAYS: That is actually what I was going to suggest because then, in terms of doing the September meeting, everything gets fleshed out. If what you are worried about is the staffing, it means you haven’t had the opportunity for someone to actually start talking to people within those groups to make sure you have enough coverage. So then it means that in the September meeting, because of who is convened, you actually then get more insight.
I’m trying to make sure that there is going to be a balance between dealing with this at the public health level and dealing with it at the healthcare level. There is I think a little more in the report about the healthcare system, and I think we also have to move a little bit more into the public health part.
For example, it is very interesting, if you were to go to L.A. County, and say Jonathan Fielding was still there, you might get a debate as to whether to do this or not because you have to deal with local needs versus common good. You start going across these different disciplines. There are just a lot of things that I want to make sure are on the table as we do this.
I think if we attach it to November there is enough time. I think October is just too quick.
MS. JACKSON: Just need to recognize that Vickie Booth is here. You were talking about staff and CDC and people at the table. Vickie has been involved. In terms of getting input, she is one of the few staff you have for population.
Regarding the timing, I’m feeling a lot more comfortable considering you are looking at the August Executive Subcommittee retreat. When we get back we will be looking at the first couple in August, and that will kind of help set — you have a July call for the Executive Subcommittee and if you get something for a retreat or a session where the executive folks can get together, then a lot of this will start to gel at that point. So I think that timing works well.
DR. COHEN: Let me ask the difficult question. Vickie, does that mean you would help staff this effort?
MS. JACKSON: I didn’t mean to put her on the spot. I just wanted you to know that you have that expertise.
MS. BOOTH: I spent a lot of my life, as many of you know, in the last seven years looking at indicators of population health so I can certainly say that I would be involved. I can also say that there are a number of other people who have been working in the whole community health improvement effort at CDC that I think could really assist with this exercise as well.
DR. COHEN: I think it would be a great opportunity to strengthen the HHS here in Washington and some of the emerging strengths at CDC.
MR. SCANLON: We would get all of them. We will round up the federal agencies. And here, honestly, NCHS just has to step up. I can’t believe the statistics agency for the nation —
DR. COHEN: So the first minimum set of core indicators that I recall being useful was the NCHS minimum dataset about 30 years ago. That was widely disseminated and actually used effectively at the state level. We are returning to the mother ship.
MR. SCANLON: Yes. Again, I think we just need a very critical assessment of what we really have, and it starts with Vital Statistics and population data, and we have the administrative data, maybe surveillance data. You can’t build a skyscraper without having a strong foundation. You just have to be realistic about what is possible. Administrative data, vital statistics, and civil registration kinds of population data — those are sort of the foundation. If people can’t get the vital statistics, Bruce, locally, or the administrative data how do you think the federal government is going to get it. I would just like to have the support that is needed, the infrastructure to support all this.
But I just think there is not agreement on what exactly — there is a lot of support for the concept of core measures, but there’s disagreement on how many and what should they be. When you look at quality measures alone, it’s just overwhelming. Some people argue there should be fewer. The clinical specialty side says no. No; that is not helpful to us. I’m an entomologist; that is not going to help me.
DR. STEAD: I also see this tying back across to the issue of public health standards, because I think the degree to which we actually can standardize on, for lack of a better word, a cascade of metrics where core indicators tend to be at the top of a cascade, and you have got to, in some cases, roll up — this plays back a little bit to some of the conversations around the framework. But in other cases, they are summative that sit up above it.
Getting to where people, when they need to develop a measure that is not part of some core cascade, when they explicitly say how it relates to that, then we in fact converge on things that are useful to each other. When people create something for their purpose and we are left after the fact to decide how they fit together, we have troubles. And the latter is closer to where we are now than the former.
I really do think we really ought to incorporate this in the Standards Committee thinking of the next step of the public health standards. How does this relate to that?
One of my thoughts — I will try, and maybe Vickie can help me with this. Take the thing the IOM had, the best measure of each of the core metrics. They actually tried to set each of them in the appropriate cell in the data structure of the framework. That might give us some idea, Vickie, of how tilted it is one way or another, and give us a way to achieve whatever the appropriate balance is. The data structure of the framework is intended to handle that continuum. This could be a way of showing it does or does not.
DR. COHEN: I saw Chris Fulcher come into the room. There is lots of work being done outside of the federal government in data organizations that is certainly relevant to this enterprise that I think could help inform public programs.
Thinking about the IRS requirements for non-profit hospitals, why isn’t there a set of core indicators required so that we could compare across all these hospital reports at the community level? To me that is a relative no-brainer for us to be involved in. Some of these suggestions could push.
Jim, I agree with you. Do we have the data? Do we have the technology? We can’t do surveys — the federal government does not have the capacity to generate surveys at the neighborhood level, but certainly it can provide resources and technical assistance to hospitals that can do these surveys or to community groups who want to do these surveys. So there are ways to collect some of the data that don’t exist now that the federal government can help promote.
I feel it is almost a field of dreams. If we build it, they will come. But we need to build something in order to stimulate the development of many of these data sources.
MR. SCANLON: Those are the kinds of questions I think we would discuss. I think that would be more useful. I wouldn’t limit this to federal; I would include organizations like Chris’s and Remington and others that are trying to do this as well.
Census is probably the only place, other than Vital Statistics and administrative data, that actually does look at community level data, so I would like to get them involved as well. And there is a whole American Public Data Users Association where people who do this for a living associate.
But I think the idea — I would use the opportunity of the IOM report to look at this area in terms of options, what would be optional and what would make sense. I think that is a good idea. It is a classic public population health question.
DR. WALKER: Every one of these measures or indicators, data elements, would have a predictable benefit contribution to potential improvement in quality adjusted life in the population and a cost. There are already a bunch of measures; you would have to get a bunch of people together at whatever the cost would be of agreeing on the core drafter.
Has anyone made any effort to estimate the cost-benefit ratios of various measures and identify 50 with the highest benefit-cost ratio?
DR. STEAD: The IOM report does a pretty nice job of estimating the cost of the current fragmented measurement system.
DR. WALKER: What was the benefit of having a standard measure of smoking? What would the cost of achieving be?
DR. STEAD: That the public health community would have to answer, and it would involve some modeling around what you think the impact of having the measure will be on smoking rates and what that will then do to health outcomes. I am not sure I know where you are headed.
DR. WALKER: Well, it seems to me that that question is at the heart of this. Should there be 12, should there be 7,000? One way to at least be arguing about the same things would be to try to estimate that benefit to cost ratio and then identify 50 or so with the highest ratio and say why don’t we see if we could do those 50 together, and maybe that would be the way to get this thing off the dime in a way that is manageable.
DR. SUAREZ: I guess, presumably, based on the IOM report estimation of the fragmented system, if one models a standardized way of doing it there would be cost savings ultimately to be achieved by virtue of standardizing and eliminating the fragmentation.
I don’t think that answers the question, Jim, that you posed which is, for each measure there is a benefit associated to it and there is a cost associated with collecting that measure, and you are trying to get to that cost-benefit ratio per measure and see which one might be a better cost-benefit factor. Right?
DR. WALKER: Yes, exactly.
DR. COHEN: We have two excellent resources to find out more information — Paul and Larry, who sat on this committee. I would like to see if we could get them involved in some of these conversations as we move forward. I think that would be really helpful.
MR. SCANLON: I think Jim is looking at that. We have a fragmented healthcare system so it’s not like data is fragmented. The data is trying to mirror what is happening and that is why we do surveys and other research, because we don’t have a systematic — so that is what you’re dealing with. You have to take samples of data or use administrative data.
There is data for almost every one of these measures here at the national level at least. We are building on it. We have a fair amount of resources already. But, number one, is that what matters to all the different audiences? Does it matter to the healthcare community? Probably not for this particular set. Does it matter to a community? Maybe. Does it matter to a state? Well, Healthy People — that’s why I think we have to hear from others.
This concept of a national framework with objectives and cascading the states and localities people have tried before, and they have indicators and they have criteria for what they use as indicators. Again, I think we have to look at it, and we will help set up the people who have done this and what they have learned and what they think would be useful.
That is, I think, the value of the Committee rather than how do we implement this.
MS. LOVE: Just anecdotally, if cost-benefit were the only thing, then we would use the administrative data measures because we can populate quite a few measures. But in the last decades that I have been involved, it has been this conundrum of, well, if it’s easy to build we don’t want it. We want a quality measure. And then the hospitals say but that would cost us $1,000.00 per measure to abstract the clinical measures. Then we are back to administrative data. But we don’t want the easy measure because it is from administrative data; we want the hard one.
To be honest, some people really just don’t understand the data flows, but others use it as a means for no public reporting. So you are stuck in a state between these factions. It is a real interesting political dynamic. I just thought I would throw that in.
MR. SCANLON: The economics side has a lot of indicators — all of the standard economic indicators, most of them at the national level, some of them at the state and regional level. A lot of it is based on administrative data if not just the surveys. Those were kind of standardized and agreed to after the Depression, I think, and we keep adding to that.
But I think healthcare is a whole other — I wish it were that simple, convert everything to income and employment.
DR. MAYS: There is one group that I think is important to bring to the table and that’s the foundations. They are actually going to carry this, some at the national level and some at the local level. So I definitely think we want RWJ at the table with their whole culture of health and some other things they are trying to push out. And then I think we can decide if there are other foundations. I can come up with some but they are like the California Endowment and things like that. They may be useful to us, also, in terms of being willing to kind of spin off doing a white paper or something like that. They are very good sometimes at funding those kinds of things.
DR. STEAD: My sense is there is a great deal of consensus around the table about what we ought to do, so we will be working by email in advance of our first Pop Health calls. They should have gone out as calendar invites. One is Friday, June 19, one is Thursday, July 16, and one is Tuesday, August 18, so those will be the three calls on our run-up to the September meeting. Thank you.
DR. SUAREZ: Thank you so much for this. You even gave us an additional 15 minutes.
MS. JACKSON: Did the populations group want to discuss the letter?
DR. COHEN: Yes; if we have a little time. Can we put up the latest revised version?
DR. SUAREZ: We do have an action item on this. We have been meeting this morning as the Population Health Subcommittee although basically the full committee is participating, so the question is whether we want to just bring the two together and just say let’s discuss the recommendations now and make a decision.
Do you want to first have a smaller discussion within the Population Health Subcommittee or do you want —
DR. COHEN: No. We are ready to roll.
DR. SUAREZ: Okay, thank you.
Agenda Item: ACTION — Community Engagement Letter
DR. COHEN: We got great feedback yesterday and we tried to incorporate it into this version. I’m sorry we don’t have hard copies, but let’s just go through the redline version.
The first addition was based on Jim’s I think really great comment about focusing or reminding the Secretary that we are well aware of the activities that have taken place to open up data, and that this letter is to really build on those efforts and expand to focus more on the data use side at the community level, building on the open data initiatives. So there’s the language. You can read that to yourselves; I don’t need to read it aloud. It’s the redline — We applaud the progress the Department has made.
PARTICIPANT: I think it would be helpful to hear it.
DR. COHEN: Okay. We applaud the progress the Department has made in expanding access to HHS data — should it be sets or data assets? — through healthdata.gov and related open data initiatives to increase the impact of these and other data-release efforts. The recommendations in this letter focus on increasing communities’ ability to access and use the data being made available.
Do we want to say datasets or data assets?
PARTICIPANT: Assets
DR. COHEN: Okay, fine. That’s broader. The second reline is just taking out a sentence that appeared previously.
Overnight we had some suggestions to strengthen the linkage between building the federal virtual data — not hub, but — enterprise, just integrating all the recommendations, that they need to be included and linked. They are not nine separate recommendations; they are an integrated set of recommendations, and we want the Secretary to consider them as such.
This is a description of the recommendations, the introduction, and it reads: The focus of these areas of action that should be approached — sorry. They focus on three areas of action – and we just added the clause, “that should be approached as dimensions of an integrated federal strategy.”
Are folks comfortable with that? Great, let’s keep moving, unless I hear otherwise.
This just reiterates the point that the strategy should encompass all of the areas of activity addressed in the letter including technical assistance and stewardship education. Further, HHS work with and for communities should be coordinated with the work of other federal departments and non-governmental organizations.
MS. KANAAN: Then we go back to the things that were suggested yesterday.
DR. COHEN: So this is changing — create a mechanism to expand opportunities. And then the addition of some other organizations with whom HHS should partner in these efforts — DOD, Census, Bureau of Indian Affairs and Department of Transportation.
We had a back and forth around whether the focus should be on developing tools or using existing tools or stimulating the development of tools.
DR. SUAREZ: We are on Recommendation 8.
DR. COHEN: Yes. The new recommendation reads, “Develop resources to help communities find, choose and use appropriate data tools and work collaboratively with governmental and non-governmental organizations and businesses to develop additional tools as needed.”
MS. LOVE: It says, to develop tools or to fill data gaps.
DR. COHEN: We are going to address data gaps in the fall. The idea here is we want the federal government to help communities figure out what tools are out there. We want to stimulate other tool development, but we recognize that the federal government needs to develop tools where they don’t exist.
MS. KANNAN: Here’s the final language without the redlining, to confuse you.
DR. SUAREZ: One minor suggestion — changing the word “choose” with the word “select.” Find, select and use.
MR. SOONTHORNSIMA: When you say data tools, does that mean tools like to access, or does that include data assets, when you say tools in this context?
DR. COHEN: Ohers might have slightly different thoughts but I’m thinking a data tool to me would be a web-based query system, or it could be something like Epi Info which provides a structure to do community surveys. Or it could be NCHS has developed some very easy stat application packages to analyze health data.
MR. SOONTHORNSIMA: So in this context, you assume that the community has already selected a data source that they want to use the tools.
DR. COHEN: Not necessarily. If a community in Massachusetts wants to know what its health issues are, they go to the state web-based data query system called MassChip, and then they can pull up 38 different data sources for their community. It is larger conceptually in my mind. It could encompass access to data as well as manipulating the data themselves.
MR. SOONTHORNSIMA: That is probably fine, then, because you really mean tools, because the tool sits on top of the data in this context.
DR. FULCHER: We work with different stakeholders, and they look at not just tools, because it means different things to different people, but talking more around features and functions and services. That resonates depending on who you are talking to — a service or a feature that may not be data-centric but relies on the data as a visualization, for example.
MR. SOONTHORNSIMA: Shouldn’t you then expand that? I like the way Chris said it.
MS. KANNAN: I do, too.
DR. FULCHER: I think tools, and then parentheses, features, functions and services — something along those lines.
MS. KANAAN: And would you say i.e. or e.g.?
DR. FULCHER: Probably e.g. because there’s more than that. Depending on the audience, that helps some people get their heads around it.
DR. WALKER: This is Walker’s rule of multiple adjectives, subjects and objects. Every increased subject, object or adjective decreases the likelihood the recipient will read the document by .5 percent.
DR. COHEN: Jim, I appreciate that. There were several other suggestions. If you notice, Susan did a remarkable job of keeping each recommendation at one sentence. Some of the sentences are getting complicated, but our goal here is to keep each recommendation at one sentence.
DR. WALKER: The document is great. It’s just every time we add something in the interest of supposed precision or whatever, we decrease the likelihood anyone reads it.
DR. COHEN: Point well taken.
MS. KANAAN: But there’s enough confusion about this word tools that maybe it’s justified.
All right, back to number 11.
DR. SUAREZ: So we are going to number 11 now.
DR. COHEN: This is just edits. This is more, I would say, wordsmithing to be more precise in our language.
MS. KANAAN: And we have a footnote for the stewardship tool kit.
DR. COHEN: The final change here is just a stronger closing, I think. We will continue to engage stakeholders, maintain related work products and promote community-focused best practices. Additionally, we will explore other ideas generated by the roundtable with an eye to developing further recommendations.
As we said before, we feel that this is the first letter to the Secretary on this subject and we plan to expand. And actually what comes out of whatever convening we have in the fall will build on this.
DR. MAYS: This is a very simple one and I am kind of trying to represent privacy. We want to make sure that the stewardship — can it go up? The tool kit, I’m sorry. Can it go up a little sooner in this? Can it be referenced earlier in the document? We’re talking about data stewardship earlier and I thought it might be good to also reference the document.
Do you have a specific place that you were thinking?
DR. FRANCIS: One way to think about it is earlier on we talk about things that NCVHS has done. We might mention there the tool kit in addition to other things. There are several places where stewardship comes up. There’s the stewardship letter, and then the tool kit.
Footnote 7 has — it doesn’t have the letter and it doesn’t have the other references.
DR. COHEN: Do you want us to add through footnote 7 those items, or is there somewhere in the body?
MS. KANAAN: I think frankly it is most effective where it is because it’s a very practical, focused —
DR. FRANCIS: I was actually suggesting that it go earlier as well; I wasn’t suggesting you take it out.
MS. KANAAN: Okay. The counter-argument would be there are many NCVHS products that we have not referenced. This is not an exhaustive reference of documents with all of our relevant reports. Even our original community health as a learning system is not — but I can certainly look for another place to put it.
I think all of the references to stewardship education, Leslie, are explicitly related to the recommendations, if I’m not mistaken.
DR. MAYS: I think the question was that they sent a letter to the Secretary, so would you not want to also reference the letter as well.
MS. KANAAN: I see. Could we do it at the same place? You want it earlier.
DR. COHEN: Okay. So let’s just expand footnote 7 to include the reference to the letter to the Secretary and the tool kit.
MS. KANAAN: I think that is a good way to do that.
DR. COHEN: We will add those.
DR. STEAD: Could you go back to the ending, because it still worries me. Now that we have added the detail — “We will continue to engage stakeholders, maintain related work products and promote community-focused best practices” — could we then kill “Additionally we will explore other ideas generated by the 2014 roundtable”? Could we go just from community-focused best practices with a nod to developing further recommendations in the coming year or two?
PARTICIPANT: That is a nice ending.
DR. COHEN: Parsimony reigns. So I guess I should ask — do we have a motion to adopt this letter?
PARTICIPANT: So moved.
DR. SUAREZ: We have a motion on the table to approve the letter as modified and a second. Any further discussion or comments about the letter?
I want to make a comment just to follow on Linda’s comment yesterday. Certainly this will be the first letter I will have the honor to sign, and I am very impressed with the simplicity and clarity of the letter itself, and I really commend the Subcommittee and Susan for an incredible, straightforward and simple, to the point letter. This is setting a new bar, as Linda was saying, for all our letters. So thank you.
All right, hearing no other comments, anyone oppose approval of the letter? Any abstentions? All in favor say aye.
(Chorus of Ayes)
Thank you very much, the motion is approved. Thank you so much for this output.
I think we are going to be able to probably go into a break before we get to eVitals, but I am going to ask Terry — I think she had an announcement she wants to make about some news on the Health Plan ID plans.
MS. DEUTSCH: Yesterday, Shana mentioned that as a result of the NCVHS recommendations about the HPID and comments from the healthcare industry there is a review going on to see what should be done about it. I wanted to announce today that, today, CMS placed on display at the Federal Register a request for information, which we call an RFI, regarding the current health plan identifier.
This RFI seeks information from the healthcare industry about the HPID enumeration structure outlined in the previously published HPID final rule, the effects of any changes in the national healthcare system since the issuance of the Federal Register on the HPID requirements, and the use of the HPID in HIPAA transactions as specified in the Transaction Standard ASC X12 Version 5010.
It will be published tomorrow; it is on display today.
PARTICIPANT: Do you know what the comment period is?
MS. DEUTSCH: Sixty days.
DR. SUAREZ: That was very quick. Thank you very much for that announcement. Anything else before we go to break? Let’s take a break until 9:45, a 20-minute break.
(Break)
Agenda Item: E-Vitals Demonstration Project
DR. SUAREZ: I think we are going to reconvene. This is really a very special part of our session today, and we want to in advance thank everyone that is going to be engaged in discussing with us on this project. For some time, we have been hearing about the work that has been done about eVitals and the use of electronic standards to communicate vital statistics, and we are very pleased to be able to have you all here today to tell us about the project, and then more importantly at the end consider what the national committee could do to help prioritize this important work.
So I want to ask Bruce if you want to say anything, because I know you had a very special interest in inviting the group here. We all had, but certainly I recognize your and acknowledge your significant role in that.
DR. COHEN: For those of you who don’t know, I was in the Bureau of Health Statistics, and my boss, my old boss, Dan Friedman, who is right across the table from me, and the Registry of Vital Records and Statistics in Massachusetts goes back to 1841. I still have datasets that go back that far on my desktop.
(Laughter.)
Actually this was before the germ theory. So counting deaths, you know. I’m happy to share it with the vitals folks.
So the way I got involved in the national committee actually was through NAPHSIS, and so my background is in vital statistics, and I’m really, as I said to Michelle earlier, I really feel it’s time to put the V back into NCVHS. So I am really happy that you are all here, and I’m excited to educate me more and my colleagues more about what’s going on in vitals. It’s great to see Delton here as well and really making this a very wonderful connection. So thank you all.
PARTICIPANT: We appreciate the invitation.
DR. SUAREZ: Yes, and I just wanted to make the point, I think you all have introduced yourselves in the first part. If you haven’t, and I know Dan hasn’t, so I’ll ask Dan if you can introduce yourself to the group here and then anybody else that has not yet introduced themselves. So do you want to introduce yourself?
DR. FRIEDMAN: Thank you, Walter. I’m Dan Friedman. I’m a consultant to the Public Health Informatics Institute, and I was a member of the committee for several years, and I learned many things that I really enjoyed learning, a few things I did not enjoy learning, and I would say the main thing I learned was keep presentations short here and to the point. Thank you.
(Laughter.)
PARTICIPANT: Walter, can we just go around and just introduce everybody?
DR. SUAREZ: Sure, absolutely. Go ahead, please.
MR. DUNCAN: Jeff Duncan, from the Utah Department of Health.
MS. FINCH: Leisa Finch, Utah Department of Health Office of Vital Records.
MR. TRIPP: Jacob Tripp, Intermountain Health Care.
MS. WILLIAMSON: Michelle Williamson, CDC’s National Center for Health Statistics.
MR. ATKINSON: I’m Delton Atkinson, Director of the Division of Vital Statistics at NCHS.
MR. STAMM: I’m John Stamm, Public Health Interfaces at Epic.
MR. GAHAGEN: I’m Michael Gahagen. I’m the Director of Compliance at Genesis Systems.
MR. SIMONETTA: Anthony Simonetta, senior technical writer at Genesis Systems.
MS. FOURQUET: Lori Fourquet, working with Michelle on the interoperability standards.
MS. WILLIAMSON: Okay, so we will get started with Delton Atkinson.
MR. ATKINSON: Good morning. I am very, very delighted to have this opportunity to talk with you about vital statistics and about the eVitals project. This is a special project for us. Ever since I took over as director, one of my goals has been to really improve the national vital statistics system, and with that, the goal has been focused on the timeliness, the quality, the usability of vital statistics data in this country, and that effort requires a great deal of work here.
We also have been focused on expanding the use of the vital statistics data for surveillance and for health services research, and I’m not going to talk as much about that this morning, but I wanted to make sure that that is part of expanding how we enable these data to be used throughout the country.
Now, we have undertaken over the last two years, last two to three years, a number of major initiatives in trying to strengthen our national vital statistics system. One of these has been the enhancement of a comprehensive network of state based electronic death registration systems, a major initiative to try to have each and every one of our states using an electronic death registration system to capture the information.
We have worked with states to put in place electronic birth registration systems. We have actually funded what we think are the last four or five states to move towards an electronic birth registration system. So really trying to make vitals collected in an automated way.
We are right now working on a concept of not just thinking about today but really thinking about the future and thinking about something called the next generation electronic death registration. So we have been actually got a project out of HHS, the Entrepreneurs Program. We have an Entrepreneurs-in-Residence who is working with us to really rethink how information needs to be collected, stored, processed, and so forth, and eVitals really fits, provides in a central element there in helping to facilitate that future for us.
We have been working on improving the quality of the cause of the death. We just recently got some fundings from Dr. Friedman to allow us to work with states to help them improve the quality of the cause of death that you see on death certificates, everything from e-learning to death audits that we are going to be doing to establishment of a system that we call VIEWS, which is Validation and Interactive Edit Web Service. What this does is when a person is completing the cause of death that the system will automatically pick up that information, send that cause of death literals, send it to a web service, and the web service will give that person, that data provider, information about how good is that information that’s there, whether it’s specific enough, whether it’s misspellings, whether it’s a whole host of kinds of problems and things that we see when we analyze the data and try to work with states to try to improve the quality of it. So we are looking to implement that system this summer across the states.
All of these and others that we are doing are all part; a central part of what has to happen is having standards in place for the sharing of information electronically. One of the things that has been a goal of mine is to minimize the number of manual interventions that it takes in order to process a record, and if you think about from the time point that it is filled out in a hospital or at a funeral home or so forth, all of the manual interventions that take place.
Now I’ll give you an example of this right in my own shop. Two years ago, it took lots of interventions by the people in order to process a record. We have now gone through and squeezed out all of those manual interventions to the point that I now can code electronically 80 percent of the records within 20 minutes of the time that the records hits my office, and we’re taking that approach throughout the whole vitals process. So we are really working with states and helping them to look at their business processes to see whether or not they can minimize the steps that they have in there. They can squeeze out the number of manual steps and moving it towards an automated process.
For the projects that we have been working with in states now, we are able to get 46 percent of the death records to us within 10 days of the data being available. We still need a lot more work to do, a lot more effort to do here and in terms of working with that, but that’s the goal of moving to this. I sometimes have joked with Michelle that I want to be able to get to the point that records are registered electronically in every state, that once it is registered, that information comes to us and we code it and we get it back, those codes back to the state. They submit it before going to lunch. By the time they get back from lunch, those codes are on their computer. That’s the goal, and we are making sort of progress towards that.
If you think about our current environment here, we have separate electronic data systems, each with their own user IDs and passwords, very little to any interoperability. That’s kind of what we have been thinking about here, and I’ve talked about the manual coding and keying of data elements from medical records to entry into the vital records systems.
The whole hypothesis behind all of this is that if we can do this, we not only can improve the timeliness, but we can hopefully improve the quality and accuracy of the data. Those are the things that we are going to have to improve as we get these things. Right now we are releasing about six special projects to states. One of those is looking at the quality of the information in electronic death registration system versus that information in electronic medical records. So we are beginning to look at how good is it once we collect it in that particular way.
Now to help in this process, and we really thank Michelle and Hetty and her staff and Lori, because they have been working with us over the last several years to try to put into place national standards for the interoperability of systems and sharing of information across systems. Those standards now have been approved as drafts by the HL7 organization, and we are now beginning to start the phase of doing pilot projects in a number of states throughout the country, both on the birth side and the death side.
So I’m going to let Michelle kind of go from here to talk about the eVitals project.
MS. WILLIAMSON: When Delton came to our group, I worked for classification and public health data standards staff within NCHS, and he asked us to collaborate with DVS on this project. I wish I had known what we were in for when we started, but I say that jokingly. I love, absolutely love, working with Delton and his team and the activities that we have been engaged in to develop these standards. It has been an interesting journey, and we still have a long way to go, but we have made a lot of progress.
So what we came up with when we started was what we call the eVitals standards initiative, and we collaborated with Delton’s group to really kind of have a statement that we could take to our state partners and to others to say what we were doing, and the focus has been on developing standards that would support interoperability between electronic health records systems and vital records systems, and the ultimate goal — and we put potentially — is that we would be able to use those standards for also exchanging information with other systems, and we will talk about that a little bit as I move forward.
One of the biggest focuses of the initiative has been to capture information at the point of care or the point of contact with the patient so that you are not retrospectively going in as a provider and going to a vital records system and writing in the information, trying to recall what was documented, but you can actually have the system pull the information that’s already been recorded and prepopulate a form or something that will represent the content for vital records. Our focus has been on doing that using the HL7 standards, as Delton mentioned, and then also the standards provided from Integrating the Healthcare Enterprise, which are also based on the HL7 standards. So we have done a lot of work with both groups, and the focus, as Delton also mentioned, is on improving the timeliness, the accuracy, and the quality of the data that we are getting for birth, death, and fetal death reporting.
What I’m showing on the next slide is primarily an overview of the various types of standards we have developed within HL7 and IHE. We have developed messaging standards, document standards. We went the route of developing both, based on the recommendations of our stakeholders. We had kind of a major discussion with several state representatives trying to decide do we just do messaging or do we just do documents, and several told us don’t tell us one way or the other; give us the option so that we can do what’s best for our state and what we are prepared to do and then we can move forward. So all of our work has been kind of on a parallel path to make sure that we can accommodate both and that there are similarities in how the information is represented.
We have also developed at IHE what we call content profiles. That piece of the puzzle provides for that prepopulation. So that’s the trick that pulls the information from the electronic health record system and puts it on the form for us, and we will show that as part of our demo.
The earlier work was on developing data models. When we first went to HL7 and said we want to create standards for vital records, I thought it was great guidance at that time. They said before you just jump in, sit back, explore, understand your workflows, understand your data content. So we sat down and we worked with a contractor who helped us to model everything, and I think it was a good foundation for moving forward.
And then we also developed functional profiles so that we are looking not only at the movement but the functionality that a system would need to support for vital records to get the data in the format that they need.
A huge piece — and I emphasize the huge; I had no concept of how much vocabulary would be needed to support these standards. We have, I think, over 130 value sets. So within the CDC’s Public Health Information Network Vocabulary Access and Distribution Systems, PHIN VADS — it’s a mouthful — we are storing all of the value sets. When I say value sets, I will give you an example. We collect information on Apgar, Apgar score. So we have SNOMED codes to represent Apgar. We have a value set of Apgar codes, and the value set is in PHIN VADS and it will have one value set Apgar 1, Apgar 5, Apgar 10. So each of the codes that we need for that, and the reason we need those codes is so that they can look in the system electronically and say, ah, an Apgar 1 was documented for this patient. That’s information that we need to put on that form to send to vital records. So a lot of work in the vocabulary, and I have to give kudos to my team members, Tamara Jean-Paul who is on maternity leave and Hetty Khan who is here who has really put a lot of focus on that activity.
So it is great that we have standards. It’s wonderful, but a standard is only as good as the testing shows that it is, in my opinion. Once you develop the standard, you really have to test it, and we have had the benefit of testing for several years now. We have tested in trial implementation at IHE. Integrating the Healthcare Enterprise every year has an event that they call the Connectathon. I like to call it the geekathon, because when you go there, there’s all these engineers and vendors, and they are all sitting together, and these are competitors, but in this environment they are testing to see how this information flows, are they able to exchange and use these standards to do what we are hoping that they can do. The first time around it may be slightly off on the numbers — I’ll look to John — but I think we were around the 20 to 30 percent range for prepopulation with the first year of testing. Now we are more in the 80, 90?
MR. STAMM: Counting change proposals and development that’s ongoing right now, we should be around 98 percent.
MS. WILLIAMSON: So you see the progress and the benefit that testing does. So we value the testing that we have done at IHE.
One other point there is that NCHS funded and supported participation by two of our state partners from Minnesota and Utah to actually come to Connectathon and test, which was a different concept. They were only used to seeing vendors, and who are these public health folks and why exactly are they here? And it was great, because they engaged and it was a great partnership, and we made a lot of progress. So it has been a good experience.
Money is always a challenge. We want to see more testing. We want to start moving into pilot implementation activities. Delton mentioned some of the activities that NCHS is supporting. So we are starting to find some funding, and we are looking for other ways to help sponsor and support.
Two of the pilots — and I know you mentioned briefly, just to emphasize the Patient-Centered Outcomes Research Institute project, PCORI project, will be focusing on death implementation, and the beauty of this one is it’s not just the testing for the state providers and vendors; it also involves NCHS showing that we can receive that information using that standard and use the standard to return a coded cause of death, and then also to appropriately disseminate that information for research purposes.
Another focus, obviously, is quality, and I know, Delton, you mentioned that. So one of the projects that we will be funding as well that we have a task order out for now is to assess the quality of the data, so to kind of do a comparison of two jurisdictions, looking at what would be collected electronically, what would be collected otherwise, and what’s the quality of the data that they are receiving.
Now to date the focus of our standards development work has been on the national standard, so what we collect that every state is required to report, but states have their differences in terms of the additional requirements they may want to put on the standard, and we have and we will demonstrate that today a way to accommodate those differences, but what we are planning to do, after much discussion with many of our state providers, is expand the standards to almost have a superset standard is what we’re calling it, that it will not only represent the national standard but will keep adding state requirements and work on harmonization with the states as well to identify the vocabulary they would need and so forth, and those fields would be optional in the basic implementation guides at HL7, but then a state, any state at the point that we are hoping at the end of this go, will be able to take that superset implementation guide and use it for their purposes. So we are actively engaged in that work.
We are also involved in standards harmonization. One of the states, Michigan, came to the table at HL7 to develop an implementation guide for birth defects, and we reviewed their standard and identified that some of the information that they were collecting is also collected for vital records, and so we convened a series of calls with them to see where we could harmonize where they might take our approach so that if the information is being collected we can try to be consistent in how that information is collected across the various standards.
And we have been doing that ongoing, I know, with Lori Fourquet and work on early hearing detection and intervention and the healthy weight profile, that if we see something when we are doing vitals, let’s do it consistently in these standards.
Outreach is a big part, because as I said, if you can’t implement, we don’t really know what we have. So trying to get states involved, we are on monthly calls with our state vital records offices talking about the implementations. We are having them call us and say, you know, can you give us a little bit more information. So we have gone from a paradigm where the very first NAPHSIS meeting, the National Association for Public Health Statistics and Information Systems, the vital registrars meeting, where at the very first one I said, oh boy, they looked at us like what are you talking about electronic health record systems for vital records, until this point where they are receptive and they are asking us to come and do demonstrations. And so we can see that paradigm shift.
But trying to do more outreach and engage others to try to get involved in the testing, to get involved in just the activities to move this forward. One of the states mentioned to us that we need more resources to provide to them so that they can disseminate it to the policyholders and others to maybe make changes within their states. So that’s where I don’t think we have done as much. We are trying to put a focus on developing better content on the website and other material that will make that available for them.
We also had a state ask us, well, do you have a project management plan that we can use if we want to get started, and we said we don’t but we will, and so we actively became engaged, collaborated with Epic and sat down and came up with a proposal for a plan, worked it out with them, and now that’s in their hands and we plan to make it available as a template for others who may want to move forward.
So at this point we are really focusing on the engagement. It’s building bridges with hospital providers, our vendors, our state agencies, to really come together as a team to make this happen. There are so many pieces of the puzzle, so many partners that have to work together, for this to effectively happen.
And I like this quote from Peter Drucker; I think it really reflects NCHS’s commitment: unless commitment is made, there are only promises and hopes, but no plans, and I think we are showing. We have a plan. We are moving forward, we are engaged, and we are trying to make this a reality.
So at this point we are ready to start our demo. I will paint the picture for you in terms of the scenario. We are starting with our birth scenario. We will present a birth scenario and then a death scenario. For the birth scenario we have Mrs. Parker. Mrs. Parker was recently seen by her OB/GYN, Dr. Frank, and she is there for her last scheduled prenatal care visit, and Dr. Frank has assured her, as she is documented in Mrs. Parker’s record, that that information will be shared with the medical facility where Mrs. Parker will be going for her, to deliver the baby.
So once she is all done with that information and that information is shared with the facility, about a month later she goes into labor. She starts having contractions, and her doctor is notified and she quickly rushes over to the hospital, and we are very happy to say that she had a beautiful baby boy, and very soon after the delivery, the birth information specialist knew that she had to complete the facility’s worksheet to provide the information that would be needed to record this birth, and so she signs in. You can see her sitting at the desk. She is on her Epic system looking at Mrs. Parker’s record, and that information will need to flow to Genesis, the vital records system, and Genesis is representing the state system.
So she completes that information, that information is able to flow; it’s exchanged using the HL7 and IHE standards, and once all of the facility’s worksheet information is done, she finishes up the information, the demographic component. That needs to be done in the vital records system. She is all done. The state has what they need. They verify with edits in their system and the Genesis system, and now that birth certificate is generated and NCHS is anxiously awaiting the final data, and that data now flows to us.
So that is the scenario.
(Applause.)
MS. WILLIAMSON: So we are going to move to John Stamm.
MR. STAMM: Thank you, Michelle, and thank you, committee members, for giving us the opportunity to be here today. We are actually going to use a patient with the last name of Strong for the demo. So we will run through on that patient.
As Michelle said, the background here is that our mother Isabel has been seen at prenatal clinics that are associated with the Epic Medical Center, and documentation on her conditions of her pregnancy have been put into the electronic medical record at that point. She came in yesterday and delivered Jamie, her baby boy who is now 17 hours old, and both mother and child went home happy and healthy yesterday.
What we have here is we have a view that a birth clerk would look at today in order to summarize for them the information that has occurred across the continuity of care for this patient and for her child so that that information can be submitted on to vital records.
To call out a few pieces for you here, on the right-hand side some of the information that is related to her prenatal care is shown, including, as Michelle mentioned, that a little over a month ago was when her last prenatal care visit was. She came in about seven times for her prenatal care. As I scroll down here, there is information that’s related to her hospital stay and information, for example, on the left-hand side related to her problem list where she has had both gonorrhea and diabetes associated with this particular pregnancy.
Now in today’s world, this is sort of where the automation stops, as it sounds like folks are pretty familiar with. At this point, the birth clerk would use this report as a summary of the information in the EHR to either transcribe the information out by hand onto a piece of paper for submission, or in some states they might be lucky enough to have a second monitor and pull up a webpage and transcribe the information directly that way, and we are going to see what we can do to show you that it can be easier than that.
I’m going to press a link down here at the bottom that just says submit to Georgia Department of Health. Now what that does is it packages up all the information that you just saw and a lot more related to the mother’s record, related to the child’s record, the facility information, and it sends it over to Georgia, who is, again, as Michelle mentioned, being played by our friends at Genesis.
What they return to me is a webpage that looks very much like it would look had they pulled it up on a second screen. It’s just a version of the facility’s worksheet. But whereas in today’s world they would have to type all that information into the facility’s worksheet, since I actually sent them all of that information for them to use, they have been able to pull a whole lot of the discrete data out and into the form.
You can see already here that the demographics have pulled through, so the mother’s name, the child’s name, the facility information is in here, and as I scroll down here we can see as well that here’s that prenatal information. You can see that they have pulled through the last prenatal visit from a month ago as well as the first prenatal visit, other information about her menses, and number of live births.
DR. SUAREZ: One quick question. So this is from, let’s say, my hospital going out to Georgia State Department of Health and pulling up the birth report form, bringing it to my system, and then filling in the data from my system into that form. Is that sort of the way it works?
MR. STAMM: Correct, and it is all using standards. So different states can use the same standard to retrieve the data and different EHRs if they support the standard could use it as well to submit.
To give you just a little more information here, you can see as well that some information that was discretely documented there on her problem list is pulled through as well. So at the top we have that gestational diabetes was pulled right off of the problem list that we had on the EHR site, as well as here we can see the gonorrhea infection that she had during her pregnancy.
Going down a little more, there are also sections here for the labor and delivery information in the hospital setting that came through yesterday, the onset of labor information, date of birth information. There are something in the range of 130 different data elements, and so I don’t know if I need to run through them all for you, but for the next component of the demo, I’m going to just jump down to the bottom.
Now as I mentioned, what we are doing here is we are actually living within the state system at this point, and this is really just a webpage that’s being hosted on their side, and that means that the state has control over how they want this to look, what they want to be prompted for. In this particular case, the state of Georgia has extra information that they want filled out before I can do the submission, and because it’s extra information that’s above and beyond the set of standards supported by the facility’s worksheet, you can see the information down here at the bottom is empty for the metabolic screening for the state of Georgia, and in that way they are still able to prompt my birth clerk user here to enter additional data in that form before we go through and do the submission.
When all the information has been verified that it’s accurate, it’s up to date, all of the state-specific requirements have been met, I can go ahead and submit the information, and that will send it along to Genesis. So I will turn the view here over to our friends at Genesis and they can show you what happens now as the next step once it has left the EHR.
DR. COHEN: So, question. So the birth certificate has two components. It has the facility’s worksheet information and then the parent’s worksheet information. Are we going to get to see how the parent’s worksheet information is put in?
MS. WILLIAMSON: Well, when you show on the Genesis system, he can show the system in the tabs that are there that are available for them to enter that information. The reason that we did it this way was really on the guidance from NAPHSIS members and DVS when we were initially developing the standards. We had actually completed all the specifications to represent facility’s worksheet and mother’s worksheet, but they said for the beginning starting this off they felt more comfortable only pulling information that came from the medical record that was the medical information. So facility’s worksheet only. So right now the way it works is you would pull that information, that content, using the standard and then be able to use the vital records system.
There has been mixed feelings about that approach. Minnesota in the testing that they did at Connectathon with us recommended that we reconsider and put mother’s worksheet information in and expand the standard to collect it all. So that may happen.
DR. MAYS: Can I just follow up on that, because that is the part that I actually wanted to see is that worksheet and understand the complexities of how you get some of that information, because that’s where people come back and it’s debatable and they don’t know what the child’s name — there are all the complications that are there.
What does the mother know about what’s transmitted? Like, does the mom — is there a point at which you tell the mom about all these other variables? Like, does she know that her gonorrhea history and all that kind of stuff is being recorded somewhere?
MR. STAMM: So there are release of information procedures that most hospitals will go through in order to provide information like that to patients that are involved, in this case to the mother. It varies a little how much gets released from state to state, but to your concerns, that sort of information is the sort of things that I typically see released to them as far as what all has been sent on to the state.
DR. MAYS: So do they get a copy of this, of everything? Or what is it that they are getting, the mom?
MR. STAMM: So not typically. Typically it’s sort of a summary view of what all needed to be released. I would also point out, which I think is what Michelle was gesturing at me for, that the information is identified at the time that it goes to the state, because they need to know who the mother and the child are, but as you will see in the Genesis flow, as they finish filling out the information in the state system, their information is de-identified before it’s passed on.
DR. SUAREZ: I have one more question to you. So the facility worksheet, is there any data in the facility worksheet that is not captured by the EHR?
MR. STAMM: Good question, and I think again that will vary from organization to organization. What I can tell you is that the standard itself is aiming to be able to include all of the data elements with the exception, I believe, of the certifier. That is something that they typically want entered manually, the certifier, and again, after change proposals going through this year and current development, we believe that we will be able to support all but one or two of the data elements that are supported by the standard.
MS. WILLIAMSON: We have spent a lot of time analyzing where the content needs to come from in an EHR, comparing with what the edit specifications from NCHS says as far as what should be the source for that information and making sure we are pulling it from the appropriate place.
DR. SUAREZ: I don’t know if this makes sense or not, but so did you, for example, in doing that analysis, did you take the 57, including all the jurisdiction forms from the various states and map all the data elements and data master — these are all the data elements that one or more of the states ask about, and so this is the large common denominator of all, and this is the data, or is that something that — so the standard by which EHRs are going to be measured or will have to have the capability to do includes all possible data elements, because I mean Epic has claims in all the states, for example.
MS. WILLIAMSON: So the basis of the standards has been the national standard. So that’s the core that every state — so we started with the core, and the current work now is we identify implementers. We are saying, okay, what does this state collect beyond what’s in the standard, and we are expanding the standard, the vocabulary, everything, and the goal is that eventually it’s going to be the superset that represents everything that the states collect.
DR. SUAREZ: Excellent. I am sorry, but one other question is about that you mentioned vocabulary. So the terminology elements are standardized as well. I noticed that in the spreadsheet there were like the mother’s education, all that, well, it says 12. So that code for example for coding education level, all that is standardized and it’s all using national coding standards?
MS. WILLIAMSON: The particular one that you selected, the education, is a demographic piece that’s from the mother’s worksheet. So we haven’t gone into that space, since we were told to leave that alone for now. What we have standardized is all of the 130 elements that we have from the facility’s worksheet. So all the clinical content. For instance, this patient has gonorrhea, a history of gonorrhea. So we have looked at SNOMED, and we work with a team of physicians, pediatricians, vocabulary specialists, DVS representatives, state representatives, and convened calls to identify and kind of data mine what are the codes within SNOMED that would be appropriate to use to select this code, so that if the mother has that code in her system, it would prepopulate the form.
MR. STAMM: What you saw within the EHR as well there was a view that made the most sense for the birth clerk, but the detailed information that’s stored is intended to be in such a way that it can be translated to those datasets so that whichever user is looking at it in whichever system or by whichever interface, they can get it in the value set that they need so that they both understand it.
DR. CHANDERRAJ: Some of the cultures don’t have naming of the child until three or four weeks afterwards; one of the sacraments is naming the child ceremony, and the naming of the child comes later. How are you going to incorporate that for the child, last name, as well? First name as well as the middle name, middle name is not given until the baptism occurs.
DR. STAMM: That is a good question. On the EHR side, that’s something that occurs to some degree today. Typically the mother will fill out a sheet, for example, while she is still in the hospital, and that information won’t necessarily get entered in until just after she has left, and so there tends to be some delay there as well.
On the EHR side it’s managed by having basically a queue of the children that are ready to be submitted or the children that they are still waiting on more information so that they know they still need to go through and do those submissions, but I can tell you as well that the way that the standards world works is that these standards are designed at an international level, and then extensions to them can be applied to help them fit to different national localities.
MR. GAHAGEN: Thank you, Michelle and John. Ladies and gentlemen, thank you for coming for this demonstration. My name is Michael Gahagen. I’m the director of compliance at Genesis. Genesis is based out of Harrisburg, Pennsylvania, and we have been participants in the vital records industry for almost 30 years. At Genesis we build state level vital event registration systems. Specifically we build interfaces that are used to record birth, death, and domestic events. Our systems are used for the registration of approximately 33 percent of all births in the United States and approximately 25 percent of all deaths in the United States. Additionally, Genesis is the first vital records vendor to participate in these types of interoperability demonstrations.
For the purposes of this demonstration, we are receiving a CDA 2.0 document. CDA stands for clinical document architecture, and essentially what that is is it’s any type of exchangeable clinical document. For this demonstration, it’s the labor and delivery summary which was shown on screen by John earlier. What we are going to do is we are going to receive that information and we are going to populate it into our form, which is shown on screen, which is the Georgia Electronic Birth Registration System.
On the left-hand side of the screen, you’ll see some tabs going up and down. Those tabs collect some of the information that we touched on earlier. It’s a mix of demographic and medical information that pertains to the mother, the father, the child, as well as characteristics of delivery. On this first tab here, the newborn tab, you will see the first name, the last name, date of birth, sex of the child, as well as the facility information. All this information has been imported into our system from the Epic EMR.
PARTICIPANT: I didn’t know Epic had opened a medical center.
(Laughter.)
MR. GAHAGEN: They have their hands in everything, apparently.
Moving along, if we go down to the mother medical 3 tab, you will see that this tab lists some exposures and infections that may have been present during the pregnancy, and you will note that the gonorrhea box is checked, and if we jump over to the risk factor tab, as John indicated earlier, you’ll see that the mother was affected by diabetes of the gestational type. We can see she had a previous preterm birth, and we can see that she had a C-section. As you may have noticed, this is information that was not keyed in by Genesis. This was imported from Epic, which is going to be a common theme through this demonstration.
If we move down to the newborn medical 1 tab, we will see the infant’s medical record number, we’ll see the Apgar score, duration of gestation, as well as the weight. Once again, all imported from Epic.
And finally, if we move down to the newborn medical 3 tab, what we see here is the metabolic screening that John mentioned at the end of his presentation. That’s been prepopulated, but also what you’ll notice about this page is that the medical and health information as well as the hearing screening information, it’s blank. The reason for this is because these fields are specific to the state of Georgia. This is not part of a national standard. So what would have to happen here is the birth clerk would have to enter this information via manually keying it in.
An alternative to that would be if there was a state stakeholder that was collecting this information, that state stakeholder could act as an HIE, health information exchange, and if they collected information and acted as an HIE, that could be prepopulated into this form here the same way the Epic information was.
DR. COHEN: So, Mike, at this stage the birth clerk would enter all, everything from the mother’s worksheet into this form.
MR. GAHAGEN: Yes, sir. You are correct.
DR. COHEN: So that’s where all the information that you were talking about, Vickie, gets put in the system for the issuance of the certificate by the jurisdiction.
DR. MAYS: Let me ask one more thing, and that is when it gets put in, for example, if this person only had a month or two of prenatal care, would I know in terms of looking at the information that you didn’t really know throughout the entire pregnancy whether they had an infection or not?
MR. STAMM: I believe that information would be limited to what was embodied in the prenatal record. Whatever was sent to the hospital was what the birth clerk would be keying in.
DR. MAYS: Oh, see, that is a very interesting issue, because it means that for those women that don’t get prenatal care or get very limited prenatal care, that we think — see, I never knew this. So we think we have a lot of information about them, and we actually don’t, and we don’t know.
MR. STAMM: For the sake of the demo here today, Epic is being used across the continuity of care.
DR. COHEN: The real issue is creating the linkages to get all the information from the prenatal care sites into the system, and in this demo, it exists; in reality it doesn’t.
MR. STAMM: I would say in reality it depends on the organization, and there are many document sharing standards that can enable you to send information between settings, like between hospitals and clinics. Some of that was demonstrated at HIMSS. We were able to receive referral documents from other family planning clinics and incorporate the data into the chart and then it gets submitted on. So I completely understand where you are looking as far as where the gaps could occur, and I would say that in today’s world that probably does leave a gap still sometimes, but there are standards out there that can be used to accommodate it and especially in areas where the different health care organizations have really worked together to be able to share the data, that information can flow over.
DR. MAYS: Yeah, see, I think it would be good to know that you don’t have the full record so that you know that it was possible that other things happened.
PARTICIPANT: So when you say the full record, you mean prenatal plus the hospital?
DR. MAYS: Like the full — so if what you are doing is — in the prenatal care. So for example, if you tell me that this woman had — I see no gonorrhea, and it may have been that you didn’t see her throughout the entire pregnancy. So I make an assumption about, you know, kind of what the pregnancy has been like, because I don’t know from your record that you only had two months of prenatal care as opposed to the full —
MS. GOSS: Actually there was a data element that showed how many — but to your point, Vickie, this is why we need health information exchanges, because people do not always go to one place in one health system. So some of the dynamics that I’ve heard they’ve mentioned in the demo are really about the larger infrastructure and the ability to have everybody be connected using national standards so that ultimately when you get down to it and you want to be able to rely on data, you can actually get at the best amount of data that’s out there in the market, but I don’t think that that’s an integral of your demo today. I also don’t know that it has really been integrated as a part of the standards world, although the nice thing is is when you look at what we are using within the health information exchange world, there is a tremendous amount of synergies because of meaningful use, the tie-in of the IHE profiles, and the leveraging of the underpinning HL7 standards. So the right building blocks are there, but boy do we have a long way to go.
MS. WILLIAMSON: Yeah, and I think I would actually like to ask Lori Fourquet maybe to speak to some of the demonstrations that we have done at HIMSS to show using an HIE and using some of the other standards for public health to get that information.
MS. FOURQUET: Yes, we have a standard that we have in IHE for the antepartum summary, and what we typically do in the demonstrations, it’s not — that prenatal care visit would come from another EMR system, not necessarily — and flow into the hospital so that the hospital is able to determine that prenatal, number of prenatal care visits and some other information that they would not have known in isolation, or perhaps by asking the mother how many or other paper mechanisms. So they have an attribute at the hospital level to enter number of prenatal care visits and some of this other information, but with the interoperability specifications for antepartum, they can flow that information in.
MS. GOSS: So really the key here is ensuring that certified electronic health records and their related certified vendors have the ability to integrate these standards and they become a standard part of the functionality that builds on the health information exchange functions, but also ensures that this is just an inherent feature that providers don’t have to pay extra for to do the reporting, because they are obligated under a lot of state and federal rules to supply the data, but they are not necessarily given the effective tools to do it in a cost-efficient manner.
MR. GAHAGEN: Just to complete the circle here, any information that wasn’t populated via the Epic EMR or via an HIE, the birth clerk would in fact have to key in manually. Once all the information has been keyed and the record is complete, what happens next is that record is sent out to various state stakeholders, as well as national stakeholders such as NCHS, SSA, CDC, and that enables the child to get a Social Security card and also, as Michelle mentioned earlier, that would allow the state to issue the birth certificate. So what we wanted to demonstrate here is, one, the efficiency that can be achieved by an interoperability protocol such as this where you eliminate a lot of the need for double data entry and also by nature of that you are going to improve data accuracy, because there’s going to be less opportunity for user error to take place.
With that being said, I’ll turn it over to Michelle.
MS. WILLIAMSON: Any other questions on the birth scenario before we move to our next one?
MS. GOSS: While you are switching over there, Michelle, just generally how long has this project been going on?
MS. WILLIAMSON: Too long.
MS. GOSS: But you are doing it thoroughly. You are actually designing it with people. You are building the standards. You are testing them. You are using the right kind of systems development life cycle and standards development life cycle that we have been trying to get in HIPAA for years. So kudos to you. It also speaks to your experience, and you’ve been in this market for a while.
When do you think this is going to be like in a primetime status, as it sounds like you are past DSTUs, but when are you going to get to the point where you think it’s actually going to be an inherent part of your CDC state directives?
PARTICIPANT: We’re looking at each other here.
MR. ATKINSON: I think the biggest thing for us in terms of moving forward is finding the resources to make it happen. We kind of get the resources from wherever we can, and at this particular point we have had about five or six different entities who have provided resources to us over the years to try to make some of these things happen. The next step for us is really beginning to get in states to start the implementation of this. We are going to go through a pilot phase and we have done one or two pilots and we are going to do another one on the death side starting in about September. We need to try to find moneys to do one on the birth side, but as we finish with that and get the draft standards finally approved by HL7, then that’s going to be the stage where we are going to be trying to work with states to go through the process of implementing.
DR. COHEN: Just to add, I think the issues, there are 57, as Walter said, jurisdictions, vital statistics jurisdictions. They all have their own set of rules and they all have their own set of requirements, and it’s a cooperative health system. It’s as much about state politics as the technical issues that we are hearing about today in terms of implementing these interoperable systems. I think clearly resources don’t exist but the — you’re a state IT person; you know what it’s like to try to do these kinds of projects through some of the issues that emerge in state government. So it’s juggling lots of moving parts.
I think states, the culture is changing in vital statistics, which is really important. You know, it all used to be a very tedious paper-based system, and now I think I would say we are on the second, almost moving into the third generation of electronic vital statistics, and the cultures are slowly changing to embrace these kinds of efforts, but it’s very slow and to do it more completely.
MS. GOSS: Well, and I think that there are a lot of resources already funneling between the feds and the states related to this data, and I think it’s aligning the incentives with some of the other policy levers that sister federal agencies have, whether it’s Medicaid or ONC meaningful use, national government standards groups, and spending a lot of time in figuring out how to bring provider efficiency to data reporting and state structures using national standards and leveraging a health information exchange framework, so I’m happy to talk with you offline, Michelle.
MS. WILLIAMSON: Yeah, and just before we go to the death demo, you mentioned about the standards, whether they were out of a draft format. What we have done is we had them as draft standards for trial use. They are still draft standards for trial use, because we took what we learned from that earlier testing and we went back and published that release so that we improved the standard. Each time that we have learned something, we have gone back to the standards organization and said we need to republish that, and so we are warning everyone to have the latest and greatest and the best version of the standards when they pull that down.
So we have about, I think, another year of the standard as a draft standard before we want to move it forward.
Another quick point is, Alix, you mentioned about providers and I talked a little bit about provider engagement. One of the challenges is that we are expecting what we need on the NCHS side and vital records side has to be an initial investment on the provider side. They have to be the ones that put the money or the efforts in their EHR systems so that the downstream flow gets to public health. So it’s kind of a challenge convincing providers to make this initial investment so that they can do that.
MS. GOSS: Yes, and they want to be able to, at least in the Commonwealth of Pennsylvania, we know that there’s an interest in leveraging the EHR connection to an HIE to get at public health efficiencies, but really your best lever is going to be meaningful use and ensuring that this is — this is a cost of doing business for humanity, and if the feds want the data, then we are going to have to align the incentives to get that, and the vendors need to come along and enable that to happen, and we shouldn’t be — if we are developing this once within a product, we shouldn’t have to charge everybody for the interface cost for a module that we are rolling out in a standard product.
MS. WILLIAMSON: And on that note of vendors, the challenge has been, because we don’t have this tied to legislation or something to move it forward, when we solicit vendors at the various Connectathon and different activities to participate, Epic has been probably the only EHR vendor that has been actively engaged in this testing with us. When we have gone to vendors in general, I always get two responses. If I say, you know, would you test the standards, they say, well, is it a meaningful use? We are talking, we’re — and then they say, well, who is my customer. So it has to be one of the two for us to be able to move this forward.
MS. GOSS: And the other thing is that we are in the big bang theory moment of trying to pull everything together, and EHR vendors are just strapped, and then the certification meaningful use 3, all those changes are really a strain on them, and we need to stabilize some environment. So I think you have some good time to get from a DSTU to the next stage in the production and then also align with meaningful use, and I think today’s demo will help elevate our awareness that maybe there are some things that we can do in further promoting the importance of this.
MS. WILLIAMSON: So I will paint the picture for our death scenario. Mr. Extest has been at Intermountain Health Care for quite some time. He has been very, very sick and unfortunately we are sorry to say that he did pass away. His physician was able to go into the Intermountain Health Care system to complete all the information in preparation for sending information to the state. So he was able to access the clinical notes and all the details that he needs to get this to Utah Department of Health. That information is being sent, once he’s completed it, to Utah Department of Health, using the HL7 version 2.5.1 standard.
The funeral director also has to complete his portion of the death certificate information, and so he goes into his system which also ties into Utah Department of Health, and he is able to look to see is the patient’s record in a waiting area, and he finds it there because the physician had access to the system and was able to complete all the information that he needed. So now he is able to go ahead and complete his documentation in that system and doesn’t need to try to contact the physician. So it saves some steps there.
So now we have the information that was documented by the physician in the Intermountain system. We also have the information that the funeral director has documented, and we need to pull that information together at Utah Department of Health. They have to merge and match that information in their system so that once that is done they are able to then generate the death certificate.
DR. COHEN: So before you begin, I guess for folks who are less familiar with the vitals world, I think the primacy of the funeral director in the organization and collection and compilation of information is really a key point historically, and I just want to make sure you guys cover how that works.
MS. WILLIAMSON: I believe that they will. We will start with taking a trip from Intermountain Health Care.
DR. MAYS: Can I ask a question first? Whose responsibility is it to record race and ethnicity? Is it going to be now taken from the electronic health record or — in the past, the funeral director does it, and they typically wouldn’t ask. So if you’re going to cover that, then I’m a happy camper.
MR. ATKINSON: Well, this won’t change any of those requirements. It still will be the responsibility of the funeral director to do it.
DR. MAYS: So my question is going to be in this E-Health whether or not you build in things where you give them guidance about how to ask for this or give them some help. I mean, I have been on both sides of it, at the side of having to fill it out with them and watch to see what they did and on the side where we have tried to talk with them about, you know, kind of this and what would they need to be able to do a better job. So if it’s going to be on a computer, can’t there be some prompts or something to ask them to ask?
MR. ATKINSON: There can be. In terms of particularly — well, if the information is left blank, that can be an edit built in such that the funeral home can’t file that record without that field being filled, completed. Now we will have to look at — let’s say if he puts something there. Whether or not that is accurate is a big question, and how do you get around that? We can make sure that they put something there. How we guarantee the accuracy is a question.
DR. MAYS: See, I think the difference is to ask them to ask, because they don’t ask and they fill it in kind of based on looking at the person. I mean, you all are experts, so you know who it is then that ends up with the most misclassifications. So if we are now in an e world, it would be great if at this you said please ask.
MS. LOVE: We do that with hospital discharge data, and they are taught to ask, ask the patient, at the time of admission. We still a lot of forcing of observed, because we still have clerks apparently or in the hospital systems where the culture is just to assume this person is a certain — it really is pervasive in a lot of the death certificates.
DR. MAYS: I am going to hope you can work on that.
MR. TRIPP: Just by way of brief background, my name is Jacob Tripp. I’m in the medical informatics department at Intermountain Health Care. Intermountain Health Care provides about half of the health care for the state of Utah, and we have a long history of collaborating with the state health department on various projects in the public health reporting of, I guess, reportable diseases, along with vital records, and in 2009 we began this project of looking at how we could give the physicians a way to complete their part of the death certificate from within the EMR without having to log into the state’s EDEN electronic death registry system, and in 2011 we started a pilot of this system that we are going to demonstrate for you. In 2013 it was rolled out to all of our providers as an optional way for how to complete this. They can still log into the state system if they prefer.
Then about ten months ago we implemented the draft standard that we worked together with Michelle’s group to create, and we continue to use that pretty much every day. We see the system being used. So I’ll just give a brief demonstration here.
This is within our HELP2 EMR, which is a homegrown system at Intermountain Health Care. If there’s time later, we can talk about some of our plans as we transition to the Cerner Millennium system.
So I’m not a physician, so hopefully Delton will cut me some slack on my cause of death, but as you — you probably can’t see this, but there is a link here for death certificates. Once I have selected the patient, I can go to that module. It’s going to automatically populate the things that we already know from the patient’s record as far as their name, if we had a Social Security Number on record, gender, and date of birth, and then as the physician I need to go out and fill out the rest of the required fields.
DR. COHEN: So, excuse me, if there was race ethnicity information in the medical record, would it auto-populate from there?
MR. TRIPP: We do not currently have that filled on the physician’s portion here. As was mentioned, that generally comes from the funeral home. So we haven’t added that here. We could add that.
DR. COHEN: That should be recorded in most medical records, if it’s coming from the hospital.
MR. TRIPP: I believe that it is recorded. We did not consider that as part of the physician’s portion of the death certificate.
PARTICIPANT: They could.
MR. TRIPP: Then the question would be who do we give, I guess, preference to if we get answers from both sides that don’t match.
PARTICIPANT: The family or the —
MR. TRIPP: So I will just quickly work through this record here. As I get to the cause of death portion, one convenience for the physician here in working within the EMR, they have access here to vital signs, the problems list, clinical notes. Speaking with our physicians, you know, they said that the history and physical is something that they often will look at, and we have kind of marked one up here for our demo, that this patient has a 10-year history of chronic kidney disease and a 30-year history of tobacco use, also that they were diagnosed with type 2 diabetes about 5 years ago. So that can help me as I click back here to the death certificate. He did die from kidney failure, if I could spell it correctly. They have been suffering that for 4 weeks, but chronic kidney disease here is the underlying cause, as well as type 2 diabetes here as one of the conditions contributing to death.
One thing that we have seen with our physicians who use this system, we recently did a study on a year’s worth of records here and found that on average we see an increase in the number of coded causes of death that we get back from the NCHS. We get about four causes per record on ones using the state’s system, and when they are done within the EMR, we are seeing an average of 5.5 causes per record. So we are seeing a little more data being entered.
As we get down here to the certifier’s portion, if I were a physician, my license number would pull in here. Right now just my name comes in. Just what we can do to save the physician some time as they fill this out.
And then once it’s completed here, we have an option to save and hold if they’re not sure yet, if they are waiting for some things and they just want to get started.
PARTICIPANT: One thing I notice is that there’s a lot of free text typing.
MR. TRIPP: Yes. The national standard for causes of death is that the physician should enter that as free text. We are not allowed to prepopulate that or to give them a pick list to choose from, and then the coding all happens at the national level. The ICD-10 codes come back to the state and then we are hoping to work at getting those codes back to us as well.
As Bruce was mentioning, kind of the traditional order of death certificate has been that the funeral home will begin the certificate and then once they have done their part they will chase down the doctor and harass them until they finish their part. One thing that we have seen more and more frequently with this module available in the EMR is that as the physician is completing the rest of the documentation they do with the death that they’ll just go and they’ll fill out this portion of the death certificate and send it off to the state, because it can be done asynchronously or before the funeral home. So that’s the way that we are modeling it today, just to show that that is happening more and more frequently.
So when I’m done, I hit submit. It’s going to pop up and remind me that this does constitute my electronic signature on the record, and if I’m okay with that then I hit okay. This data gets packaged up into HL7 message using the draft standard format and gets sent over to the state system, where it goes into, I guess, a waiting area until there’s a funeral home record to match it together with.
So we will turn control over to Leisa Finch, who is going to represent a funeral home today for our demonstration.
DR. MAYS: Can I just ask a question before we go to the funeral home? I want to stay with the doctor. One of the things is we are talking about it as the doctor, and there are certain things that are in the electronic health record, and I guess I’m trying to push to see if we can’t broaden that so that some stuff can get populated from that. I mean, are we dealing with a rule or a regulation or something, or are we at a point where given that we have moved into the world of EHRs that we can’t take some things from there, like if they have the race and ethnicity, like if there are other things which get put into the EHR? Can you include other things? I don’t know what’s going to happen, but there’s a push, for example, to have sexual orientation in the medical records. I don’t know if there’s any thoughts about whether that would get put onto this or not. There are some issues around wanting more information in the death certificate around suicides. I mean, so is there also this — are there any expansions that are possible that come from the EHR?
MR. ATKINSON: I’ll kind of answer that from the national level. When we first got started with this, and this is really back in 2007 when we spent this whole effort, it really was more of an effort to try to see if we can get people on board with the concept, and so one of the things that states were very protective of is what is called the legal items, which are your more demographic — the name and demographic items, and they really did not feel that they wanted at that point in time to trust the medical records in providing the legal items to the death certificates.
Now, as we have — that was 2007, 2008. This is one of the things that, as Michelle talked about earlier in terms of going back and expanding kind of what is collected, some of these are items that are going to need to be revisited, and so you are going to get into that question of race, if it’s there, do we go ahead and prepopulate it, then giving the funeral home the ability to look at it and change it if you have a different response from the family, or something else? So those are the kinds of things that we are going to have to look at and debate and really talk with our state partners in terms of what they are most comfortable with, but we are dealing with the legal items, and those are the ones that they tended to want to make sure that they control more of.
PARTICIPANT: Thank you. That’s very helpful.
MR. TRIPP: For the provider side, I would say technically if we can record it in the system, we can send it to the state. It’s just more about policy and how we want to handle that.
Take it away, Leisa.
MS. FINCH: Again, I am Leisa Finch from the Utah Department of Health Office of Vital Records and Statistics, and I’m actually the coordinator of Utah’s electronic death registration system, and I’m going to be acting as the funeral director, and as the funeral director, and we have discussed this a little bit here, it’s my responsibility to collect the demographic information, collect it on the decedent by interviewing a family member or the informant, and that information is actually entered into Utah’s electronic death registration system, and again, it’s called EDEN and it stands for Electronic Death Entry Network.
You were mentioning, one of those demographic items is the race field that you’re talking about. Now, if you’ll notice, it’s a required field in our system. It won’t let them past it without answering very specific items on the death record, just to kind of give you just a little bit of view of that, in the Utah system, what won’t let them past it. As you are talking, again, the physician may be able to collect it. It could populate in here. But this is a funeral home field, and as Delton was saying, it may populate there, but they may — the family may be reviewing this and saying you know what, that’s not right, and the funeral director would need to be able to correct that at this point.
One of my other responsibilities is collecting the health care provider that is supposed to be completing their portion of the death record. EDEN also allows me to identify if the physician has already completed his death record through his EMR, and I look at that by clicking on a link here in the certifier screen, their section right here, and it takes me to the death matching table, and as you can see, the decedent that the physician actually put in is in the table, and when I see that this decedent is on my table, this tells me that the provider is done — as a funeral director — provider is done. I don’t need to call his office. I don’t need to bug him or give him his four phone calls. He is done, and that is score. That is frosting to us.
Okay, so the merge process takes a — it runs through about a 20-minute cycle looking for like-like things to merge with. Once that validation happens, Jeffery, can you kind of touch base on that for me while I pull this record back up?
MR. DUNCAN: Once it merges? Okay, so it matches on demographic characteristics like first name, last name, date of birth, and we have to have a perfect match for them to merge, because we don’t want to have a false merge. If they don’t match, then staff can go in and review the records and they can force a merge, and Leisa and her staff will go in and do that several times a day. They will look for records that are waiting out there and haven’t merged, and then they’ll make them merge.
DR. SUAREZ: This is merging the record from —
MR. DUNCAN: The one that’s coming from Intermountain with the one that’s coming from the funeral director. In the past workflow, everyone did their work within the same system. So you didn’t have to worry about these asynchronous flows of information. Now you have a flow coming from Intermountain and one from the funeral director, you have to merge it, and they do have some fields in common. So we have to have some logic about which field takes precedence, and to the point about race and ethnicity, we have some other fields like the date of birth, for example; there may be a discrepancy between date of birth in the hospital record and in the one that comes from the funeral director that would cause those records not to merge, and there has to be some kind of manual resolution as to which date of birth is the correct date of birth.
DR. CORNELIUS: Could you clarify offhand about what percent of those records you have to do some type of hand matching as opposed to these perfect electronic matches you were talking about?
MR. DUNCAN: I believe about 50 percent have to be, because there are discrepancies, particularly in names.
DR. COHEN: I think name is a really important issue. Thanks for bringing that up, Jeff. When we talk about EHRs populating vital statistics, there are some fields that make a lot of sense, but other fields, when you go to the hospital, you could be Jack Johnson, Joe Johnson, Joey Johnson. You know, on your birth certificate and your death certificate, you are Joseph Johnson III. So, you know, there are different sources of information that have higher quality data. So doing these kinds of matches, you really need to think about where the gold standard might be.
DR. DUNCAN: Right, and the gold standard on death certificates is typically what the funeral director gets from the family, and it’s verified with Social Security. So there’s a button in every state’s EDR system, I think almost every state, where they will put in the name, the date of birth, the sex, and they will click a button and it actually verifies with SSA computers to say that the number is correct, and if it is, then when the death is registered, another message goes to Social Security that says to terminate that account.
MR. TRIPP: As I mentioned, at Intermountain Health Care we are in the process of transitioning to the Cerner Millennium EMR. As part of that process, we have actually gotten them to build a screen where we can do this death certificate stuff, and an interface to go with that, and we have registered a few deaths that way at the one pilot hospital that we have up and running. So now that they have employed that, potentially they could use that with other customers if they were in a state that had implemented the receiving side of that interface.
PARTICIPANT: While we have a little bit of time, I might mention as we see the physicians completing the record a little more quickly, as part of our study we saw that on average it’s taking about 43 hours from time of death until the physicians completed their portion of the death certificate when they use the EMR module. Using the state system, the average is about 80 hours, and if they are using it on paper it’s about 112 hours. So we are looking at half the time that it takes using the state’s electronic system or a third of the time of using a paper record.
MS. FINCH: So I just wanted to validate the importance of that timeliness in filing of that death certificate. That death certificate is extremely important to the family members, and any kind of delay in that death process or that death registration, families can’t inter their family members. They can’t close bank accounts. They can’t get their life insurance benefits. So it is really difficult for the families if that isn’t done timely.
Jacob mentioned before that as — because I see this every day. We are seeing an upswing in the records that are being processed through the EMR, that are oftentimes same day that the person passed away, even before the funeral director has initiated the death record. So we are seeing a huge upswing in that, which is very exciting for us, because that benefits, again, the families, and again the physician has 72 hours to have that death record completed by state statute, and oftentimes we don’t see that, and again, when we see that it’s done the same day, it is just the chills just happen.
Now as the funeral director, I have five days to get that death record to the local registrar for recording and registration, and that’s where I’m going to turn it over to Jeffery, if we can turn this screen over to Jeffery, as the local registrar.
MR. DUNCAN: So I am demonstrating the role of the local registrar, and Leisa mentioned the death certificate is important to the family. It’s also very important to public health, and the registrar is the last step in completing the death certificate. The physician has done their part. The funeral director has done their part. The local registrar finds that record and reviews it and makes sure everything is complete.
So the funeral director, or the local registrar rather, I’ll just choose — we have had a lot of dead Extests in the last couple of years.
(Laughter.)
So the local registrar will log in. They will review the record. They will see that everything is complete, and a lot of that is done through computer edits. Specifically what they will do is look at the cause of death, and they look to see is it ill defined, something like cardiac arrest. Sometimes doctors like to put cardiac arrest as a cause of death, and that’s not very meaningful for public health statistics. So local registrars can call a physician and ask for a little more detail.
They will look for causes of death that are suspicious that should be reported to a medical examiner or a coroner, and if they find something like that, they can push a button and refer that to a medical examiner or investigator for review.
Third, they will look for causes of death that are reportable: influenza, for example, or HIV, or a couple of years ago we were reviewing cause of death for bird flu. So the things that we are looking for can change as conditions change.
So once the registrar is happy with the record and the cause of death that’s on the record, they can click to register it, and I don’t think I’ll actually go register it. Once it’s registered, the family can get that — once it’s registered, the cause of death is available to the epidemiologists in the local and the state health department, and it’s available to NCHS. Once a week we batch records and we send them to the National Center for Health Statistics, and those cause of death literals that are entered go through a machine process, and they are assigned ICD-10 codes and we get those codes back. Like Delton mentioned, we are working on streamlining that process. So we get the codes back more quickly. So they are more useful for public health surveillance and things like that.
And I just want to point out that this is an operational system, like Jacob mentioned. We have been doing deaths in this system since 2009, and the use just keeps increasing and we are seeing dramatic improvements in the timeliness and the quality of death, cause of death information that we get.
MR. ATKINSON: Just to kind of follow on, from a national perspective in terms of timeliness, we right now are getting, as was mentioned before, about 46 percent of the records within 10 days of the date of event. Without electronic systems, we could not do that, and as we see the growth of these electronic systems and not only the systems but states changing and modifying their business processes, because sometimes we have seen in some states they have implemented a system but didn’t change the business process along with it. So all it did was just slow down the flow of records.
But you are beginning to see different models, just like in Utah where the cause of death component of the death certificate is tied in with the electronic medical records, you are seeing in other states where the state office has completely given up the task of issuing death certificates or birth certificates at the state level. It’s now only through the counties, because now you have tied in all of the counties and the local registrars to a central computer. So the whole model of how vital records and vital statistics is being done is changing. As Bruce mentioned earlier, there is a change going on there, and how and where this will end is going to be interesting to see.
MR. SCANLON: Quick question: are you using, when you code cause of death, are you using ICD-9 or I thought —
MR. ATKINSON: We use ICD-10.
MR. SCANLON: And the world didn’t come to an end?
MR. ATKINSON: It didn’t come to an end. We started using ICD-10 in 2003.
MR. SCANLON: Last question, when Utah, when the Utah registrar approves the record, you shoot it off to Social Security at that point, or do it without the cause of death?
MS. FINCH: In the decedent information screen, there is an SSA validation portion that the funeral director does as he is processing the record that he just clicks on a link that’s in our electronic death registration system that does that validation through SSA.
MR. ATKINSON: Now, once that record is registered, we are now trying to get states to send us that record at least on a daily basis. That’s where we are trying to get to, and our national goal is to be able to get 80 percent of all deaths within 10 days after the date of the event. That’s the national goal.
DR. COHEN: I had a question about automated geocoding. Is it built into your system in Utah? If we are trying to generate cause of death information within 10 days, we are also trying to generate more detailed geocoding into that process?
MR. DUNCAN: It is not in our current system. It will be in — we are developing the next generation EDEN that will use a web service to geocode. It’s not reported to NCHS, because I don’t know that the standard supports it yet, the reporting standard to NCHS, but it could be expanded to include geocoding information.
MR. ATKINSON: And that is something, Bruce, we have been talking about from a national perspective is we know that some states are already doing some geocoding, and so that’s something that we really want to be engaged with the states about. Whether or not there needs to be some national effort with this or whether or not states prefer it to be on an individual state-by-state basis, so that’s something that we will be engaging in discussions on.
MS. LOVE: This may be a rhetorical question, but I am sitting here thinking why can’t we or should we think about hospitalization data that 48 states collect or now probably all of them? We have them standardized, automated. I mean, do you see the day, or is this realistic, where we could start taking in other core datasets and enhancing them perhaps with the medical record with clinical data as we so much want for risk adjustment and outcomes, and can you do it by next year, Delton?
(Laughter.)
MR. ATKINSON: Maybe by the time I retire. I think right now we can’t do it. Vitals is — there is not a federal rule that’s associated with vitals. Our authority is really only to purchase the records from states, and so the authorities for vital records and vital statistics is at the state level. State laws, state statutes —
PARTICIPANT: Similar to hospital data.
MR. ATKINSON: Yeah, and it is kind of working with each state to look at different strategies and different methods, and so it’s somewhat of a slow process, but the kinds of things that are acceptable now in vitals, even in 2007 when we got started was not acceptable at that point. So it’s changing, and maybe we will someday get there.
DR. SUAREZ: So I was just thinking, and before — I was thinking that one of the ways, for example, that the country has been pursuing some of the advances or increases in data availability is through grants. So for example, you know, CMS provides a new set of grants called State Innovation Models to states and give them money to build an APCD system to collect APCD data. Ideally it should be using standards, but in some — is there a way to — I mean, is there a way to help states build this data through some funding back to them? I mean, besides the political point that I think Bruce mentioned about state politics and all that, funding I presume is one of the bigger barriers.
MR. ATKINSON: It is one of the bigger barriers, and we have now a way in which we can get moneys to states for efforts to enhance the vital statistics system. We have a VSCP contract, which is where we purchase the data, but in our last contract for the first time we worked with our purchase and grants office, our contracts folks, and we were able to insert language in that VSCP contract that allows us to do special projects with states. So as we get special project moneys, we can funnel that to states to carry on innovative kinds of tasks or activities. So yes, it is possible, indeed, to do it that particular way.
DR. MAYS: I just want to ask: something that NCHS did that I thought was absolutely wonderful was this linked mortality dataset. So you now have the mortality data linked to other things like NHIS, and it’s great. So congratulations on that.
Now how quickly is that refreshed from, you know, like if things are coming in pretty quickly, is that particular dataset like refreshed constantly with new information? And the second is originally when that dataset came out, it had all causes of death, and now it’s changed where it has more limited causes of death, and I just wonder why that changed.
MR. ATKINSON: The refreshing of the dataset, that dataset is refreshed on an annual basis, because you want to have all of the final records. There is a — especially if you’re looking at the cause, there’s a lot of back and forth that goes on with states in terms of that, and a big group that the states have to work with is the medical examiners and coroners and trying to get final resolutions of the cause of death. As was mentioned earlier, we get a lot of causes for which there’s just a pending or ill-defined cause, and my staff spend a great deal of time working with the states who then works with the funeral homes, the physicians or medical examiners, trying to get the corrected cause.
Now, the speed at which we are doing this now is that we are able to — in 2013, for the 2013 data, we released the final death file in December of 2014. First time ever that we have done that. Maybe we can do it in November of this year. We’ll have to see. But it’s that timelines are really improving in terms of that.
DR. MAYS: What about the change in the cause of death? It used to be all causes of death were available, and it seems like now it’s down to specific causes, not all of them.
MR. ATKINSON: Yes, not all of them are available on that sort of file that’s released. You can go through a research protocol and get access to the more detailed data. That came about through discussions with the states who were concerned about some confidentiality and security and some of that, and so again, how we operate is sort of a negotiated policy between the states and us in terms of the use of the data.
MR. SOONTHORNSIMA: I think you mentioned there’s a number of levers like funding, like HHS leadership, and so forth, but one thing you mention oftentimes is really you rely heavily on the states. So when you look at the enthusiasm, Leisa, the state of Utah has displayed, do you see a similar groundswell or sense of enthusiasm or some common needs are being felt among the states, considering the complexity that the states have to deal with with the various offices?
MR. ATKINSON: Well, I think yes, I am seeing that enthusiasm, especially over the last 3, 4, 5 years, as we have put into place these special project provisions and so right now, for example, I have six special projects that we just released to states, and states can submit a proposal that allows them to carry out that, and one of the projects on timeliness, I’m going to be working with 13 states. We are going to fund 13 states in this fiscal year and another 12 states in next fiscal year. So there is more of these opportunities to help states to improve their infrastructure, and that’s what we are talking about, their ability to be able to do some things, and that’s kind of what these special project funds are being directed at.
MS. WILLIAMSON: We are definitely on the standards side started to get more and more calls from state representatives asking, you know, can we have a call to talk with you to better understand what’s available with the standards, how would we begin if we want to start this process. So definitely seeing an increase in that.
One other that we found was very interesting was we got a solicitation from a provider from one of the states who did a search on the internet and found my colleague Hetty Khan’s presentation form what we had done at NAPHSIS some years ago, and she was interested because she’s in an OB/GYN physician who says I’m just tired of duplicating this information. We are implementing a new system in our hospital, and I don’t see why this information can’t flow. Are there standards? So she found this, reached out to us, and we have been in discussions with her. We connected her to the state office so we could talk about possibilities, and actually she will be presenting to NAPHSIS at the meeting next week.
MS. GOSS: So in thinking about the policy levers, incentives, alignment, getting everybody going in the right direction together and knowing that the dichotomy sometimes will be across the state rules, how do you see this work fitting in with the interoperability roadmap? How do you see it fitting in with some of the communities of practices that are out there? Any thoughts on that?
MS. WILLIAMSON: We have actually responded. The NCHS response to CDC to submit on the interoperability roadmap to specifically recognize these standards. I think we made the point that we have said many times that this is the longitudinal record that we are talking about, and this is the beginning and the ends of that record. So we think it makes sense to include these standards as part of that roadmap. So that has been our response, and we have also presented to the HIT Policy Committee, Delton and I, a couple of years ago, encouraging even in meaningful use that this be recognized.
MS. GOSS: I think you have got a great story to tell, and it would be neat to see even through some of the other state leaders, not just within your vital records community, but to get the word out to some of those other counterparts that are doing SIM, that are doing HIT coordination across states and commonwealths.
MS. WILLIAMSON: We have a question on the phone. This is from Gib Parrish. In the birth demo, are hospital birth data entered directly into the Epic EHR or is the Georgia state EVRS screen made available to data entry within the Epic EHR environment?
MR. STAMM: Good question. This is John, again, with Epic. So Epic has fields for I think very close to all if not all of the fields that are both for the facility’s worksheet and for the mother’s worksheet. In the demo that you showed, the information would be documented for all of the above on the Epic side, and then we showed the ability for it to flow automatically over to the Genesis system for the facility’s worksheet. The additional steps of entering the maternal worksheet information would, in that particular demo, still be transcribed into the Genesis system.
MS. WILLIAMSON: Are there any questions for the demo team?
DR. SUAREZ: Probably the bigger question, how can the national committee help advance this initiative forward?
MS. WILLIAMSON: Can I ask to hold that question until Dan presents, because I think some of what Dan has to say will play into that response.
So we will turn this over to Dan Friedman.
PARTICIPANT: I know there is some public comment, too, and questions. Okay, go ahead.
DR. VAUGHAN: Hi, just excellent presentations. I wonder — one of the questions that came to mind is I wonder how in particularly around the death data, there is a lot of discussion about asking the family, the funeral director. I wonder, given the evolving environment, whether there might be consideration of the patient being the key informant for their own demographic data. Some hospital systems already have, require you to opt out of medical directives, medical power of attorney, and it’s not necessarily saying perhaps what you want on your death certificate, but having the patient be the authority of their life.
MR. DUNCAN: We talked about that in Utah. We did a pilot under the Beacon Grant of an electronic, an e-POLST, we called it, a physician order for life sustaining treatment. So we talked about gathering some of that information and using it to populate the death certificate, but we never got that far.
MS. WILLIAMSON: I will add to that when we first started the project, looking at vital records standards, we were proposed the vision of handing the mother or someone, a family member, an iPad and saying, okay, here’s the form, fill out that information. So you would get it directly and feed it in, use the standards to pull that information in, but as Delton mentioned we were told at this point to focus only on the clinical content that could come from the EHR, and then maybe later we will expand to those others if they decide to move in that direction.
DR. VAUGHAN: One of the things might be that again the patient is the ultimate authority, when I enroll in a plan that those are already there, and so when that time comes, many years in the future, that’s there and that’s the ultimate reconciliation.
MS. JACKSON: That question was from Dr. Leah Vaughan of the data working group, data access and use. Thanks, Leah.
DR. FRANCIS: Leslie Francis, also from the working group. Just an observation: the meaningful use stage 3 NPRM is out. Comment period is, I believe, just closed. Closing tomorrow, that’s right. So there still is time if you look at the public health portions, they are very limited and something that I think is really critical to have on the agenda is that stage 3 is the last stage for meaningful use, and after that, the kinds of standards for things like public health are going to be up for the annual standards for EHR certification. So I fear that a window is closing and people are going to have to be very careful to keep that window open if good robust public health capabilities are going to be in certified EHRs.
MS. WILLIAMSON: And we did provide that as one of our comments from NCHS to CDC, and we are aware of a few other states who also indicated that they plan to provide that as their response.
DR. SUAREZ: The point being basically this is not included as a neither a metric in CMS NPRM, meaning meaningful use metric for stage 3, nor is included in the ONC certification and standards as a name, identifying name standard. Is that correct?
PARTICIPANT: That’s correct.
DR. MAYS: Leah’s question reminded me of something. Right now one of the problems is these kind of contested deaths. So a police officer shoots somebody, and then it takes like two or three years sometimes before what — the truth of what happened. How long is the record open for you to make corrections, and is there some process by if somebody wanted to amend a death certificate that it could be amended?
MR. ATKINSON: I think if states get an amended death certificate, then they would go through their — they have an amendment process that they usually go through. From the statistical standpoint, once we close the files, they are pretty much closed and we don’t come back from them. We do like to get an amended record from the state, but that’s for our national death index, because we have the national death index. We keep records of every death, and it goes back to about 1979 to the present, and so anytime there is an amendment, we would like to get an update so that we can put it into the national death index. But we don’t go back and do anything with the prior statistical data.
DR. COHEN: So I just want to respond, and Jeff might want to from Utah’s perspective. You know, particularly with the epidemic of opioid deaths for example, states are very focused on working more closely with the medical examiners to deal with unknown causes or pending causes. Pending causes are a real issue, but they are a state-level issue, not a national issue, and I think all jurisdictions have tried to speed up that process and handle pendings and amendments, and it usually comes down to having the resources to do this in a timely fashion, and in most states historically this has not been a priority, but now maybe with more emphasis on perhaps some of the violent death situations and substance abuse issues, there might be more resources to clean up the backlog of pendings that, you know, the state health department hasn’t been able to get from the ME’s office in order to finalize the death records. I don’t know your experience.
MR. DUNCAN: There is a resource issue in Utah, and there was something in the news just two weeks ago about the backlog of pending. So when someone dies and the medical examiner reviews and they can’t establish a cause of death, they’ll put pending, and it’s usually pending a toxicology lab analysis, and that’s where the recourse constraint comes into play in Utah, because those can sometimes take weeks to months to get back results from toxicology, and when the ME gets those results back then they will amend the cause of death in our system and we’ll amend it to NCHS.
DR. SUAREZ: Let me see if there’s, before we go to Dan, is there any other question from anyone from the public?
[No response.]
Okay, anybody on the phone? Thank you, Vickie. Anybody on the phone with additional questions?
DR. FRIEDMAN: Thank you, Walter. I am going to actually start with a question, which is I’m mindful of your agenda and the time, and how long do you want me to speak for?
DR. SUAREZ: Well, so it’s about 11:35. We have probably, what, about 10 minutes? Is that good?
DR. FRIEDMAN: Sure, I’m infinitely flexible. This is a — I’m speaking about a project which was funded through a cooperative agreement from NCHS from CDC to the Public Health Informatics Institute as a consultant to PHII. We generally start our presentations with big pictures of Dave Ross in various heroic poses. Dave on horseback, Dave in a Formula One racecar, et cetera.
(Laughter)
But since he’s a member, I’m going to take the liberty of — and also since we don’t have enough time to go through my entire slide set of those heroic pictures, I’m going to skip that.
PARTICIPANT: Well, hey, Dan, this is Dave. I am on the phone.
(Laughter.)
DR. FRIEDMAN: Well, Dave, I hope you’re not offended and that I still have a job even without showing all of those pictures.
The purpose of the project was twofold. First of all, specifically to develop a roadmap with possible routes for certification of electronic health record systems capability for generating and exchanging data with electronic vital record systems. More generally, the purpose is to develop a roadmap for enhancing EHRS to generate and exchange data with EVRS, and just for the sake of saving time, I’m just going to refer to this as EHRS- EVRS integration.
What did we do in the project? We basically did two things. First of all, we did an environmental scan. We looked at websites. We looked at published reports, unpublished reports, et cetera, relating to EVRS and EHRS standards, certification, et cetera. We also spent a lot of time looking at the excellent reports from the Minnesota and Utah pilot projects.
In addition to that, we interviewed 42 experts, including some of the people in this room, in vitals, public health, and medical informatics, clinicians, certification experts, and EHRS and EVRS vendors, and we talked to them essentially about three things. First of all, we talked to them about their perception of barriers to EHRS-EVRS integration. Second, we talked to them about their perception of facilitators to integration, and third, we talked to them about their recommendations for next steps.
Okay, what were the three main things that we learned from these interviews with 42 experts? Well, the first thing we learned — and I’m going to pose all these as questions. The first thing we learned was basically what was said to us often, particularly from clinicians, and particularly from people in state health departments, and the first thing was what’s the goal? And here I’m harking back to some of the things that particularly Delton and Michelle said earlier. You know, is the goal EHRS-EVRS integration, per se, or is the goal improving the timeliness, the accuracy, and the completeness of vital records and doing that as efficiently and as economically as possible? What we heard from clinicians, what we heard from the state health departments, is we need to constantly keep in mind that the goal is the latter — in other words, improving timeliness, et cetera — and not specifically EHRS-EVRS integration.
Second thing we learned is essentially where’s the evidence, and we heard that particularly again from state health departments as well as from providers, hospitals, as well as from clinicians, and when they said where’s the evidence, essentially what they meant was, as Delton said to start with, EHRS-EVRS as — it is a hypothesis that that improves timeliness, accuracy, and completeness. We have some very promising data to start with from Utah on the death side, but it is still a hypothesis, and what was stressed to us was the need for planned pilot projects of the sort that Delton has been describing.
The third thing that we heard — and this is particularly from EHRS vendors — was where’s the market, and we heard great concern. One of the firms we spoke with said, look, I haven’t been asked to develop this product by a single state health department or by a single hospital system. We need incentives for development. We need incentives for implementation.
What else did we learn about barriers and facilitators, and what I just said about the three main things may sound discouraging, but it really isn’t necessarily that discouraging, because we asked people about barriers. We asked them about facilitators, and essentially for almost every barrier we really heard a facilitator.
The types of barriers and facilitators we heard about included planning and stakeholder engagement, national and state policies, including mandates and incentives, and especially state variability, and we heard this particularly from the EHRS vendors. Data quality, provider workflows, HIT, and of course certification. Health care providers, state health departments and vendors, again stress the need for incentives for development and implementation.
Clinicians talked to us about the need for vitals and EHRS-EVRS to meet their clinical needs. They are not — they are much less concerned about the national standard certificate. They are much less concerned about state requirements. They are concerned about quality improvement and about patient safety.
Informaticians talked to us about the needs for long-term planning for state health department adoption of HL7, as well as the need for long-term planning about moving to EHRS-EVRS. All of — not all of the experts, but many of the experts from all the different sectors we talked to discussed the needs for planned pilot projects and planned pilot projects with common metrics.
Okay, what are the roadmap and routes to enhancing EHRS to generate and exchange data with EVRS? This looks like gobbledygook from where you sit. What’s important is — this is the overall roadmap that was developed — what’s important is the banners, the headlines along the top, which many of you will recognize as a minor adaptation of the ever-popular HHS enterprise performance lifecycle that’s been adapted by CDC; it’s been adapted by NIH, been adapted by PHII. Here we happen to adapt it by having nine phases instead of ten.
Phases one through three — and this is really the guts of the project — in the roadmap that we envision, phases one to three are it’s a common roadmap, and what’s really important here and important throughout the roadmap is, as Delton and Michelle were talking about, engaging stakeholders, determining stakeholder information needs, describing vital records workflow and variation from state to state, and this was a particular — we kept hearing this from the EHRS vendors. Assessing current performance and then, of course, requirements analysis.
So this is for all the different possible routes within the roadmap. One of the issues that obviously needs to be addressed, that could be addressed at this point, is revising vital certificates to better meet stakeholder needs.
Phases four and five are the design and development phase. Here we have a set of exemplars, a set of ideal types of possible routes. This could be twelve routes. It could be three routes. It could be fifteen routes. The specific routes that we have laid out are really much less important than the concept of really following that EPLC and developing with stakeholder engagement different possible routes, implementing a selected number of those routes in pilot projects. The routes that we laid out here, two of the routes, routes one and two, involve no health information system changes at all. Basically it’s working within the current processes, the first one, and the second one is just improving current workflow processes. Third and fourth routes are essentially working within — using the current EVRS and still transmitting data from hospitals and from hospitals and funeral director homes via current EVRS, even if some of that is essentially imbedded into the EHRS.
And then the fifth and sixth that we have laid out here as exemplars are two different examples of collecting the data and transmitting the data through EHRS. So those different routes are tested. Some of them are chosen for testing with common metrics, with strict common metrics, and then implemented and then evaluated before national deployment.
The individual routes, whether it’s these six routes or six other routes, whatever, will vary along a set of common conditions. Place of data entry, is it within the EVRS or is it within EHRS? Is there any manual data entry, and if so, how much? Clearly manual data entry for everybody involved is a huge bugaboo. Is there duplicate data entry? Again, this varies from exemplar to exemplar, and certainly at the data provider level, for hospitals, for funeral directors, again, if we can get beyond double data entry, that will be a huge gain.
Place of data storage, is it within the EVRS or the EHRS? And then means of data transfer, is it within — is it via the current EVRS or is it via HL7?
How should specific routes be chosen for integration? Well, the first decision that needs to be made is are we going to choose routes for births, births slash fetal deaths, or for deaths, or for both, and as you have heard via the demos today, you know, that’s a complicated choice, and on the one hand, there are some advantages to going with births, because all of the data reside at the hospital, even though some of them come from the prenatal care provider, but births is a really complicated data set. There are so many data items. Deaths is much more straightforward, even though the data come from both the certifying physician as well as from the funeral director.
Other criteria for choosing routes include maximizing timeliness, maximizing accuracy, maximizing completeness, maximizing efficiency, maximizing economy, and those are all things that will need to be common metrics through a set of carefully planned pilots, minimizing design and implementation costs, maximizing speed of implementation and deployment, maximizing likelihood of state health department acceptance. You know, this is not something, as Delton and Michelle were talking about, that NCHS can mandate. NCHS could, funds permitting, provide incentives, but the states still would need to accept those incentives similarly with the data providers.
And then finally, inherent in all of this is maximizing the value proposition for all of the different stakeholders.
That’s been a couple minutes beyond the 10. I thank you for your patience, and I also want to acknowledge, of course, not only Dave Ross on horseback but also my colleague Gib Parrish in Maine. Thank you.
DR. SUAREZ: Thank you very much, Dan. Thank you for that.
Well, first of all, let me see if there are any comments or questions or reactions to Dan’s presentation.
DR. COHEN: So, Dan, did you find any systems that — pilot systems — that represent pathways that are advanced enough that they should be explored more before common metrics are developed, or do you think we need to develop common metrics first before we engage in piloting?
DR. FRIEDMAN: I think that I am going to turn that over to Delton, because their movement in terms of pilots — you know, they are moving rapidly in terms of pilots.
MR. ATKINSON: I am not sure that we are at the point of saying there’s one sort of preferred method of the pilots and the ways that we have done it so far. We have tried to give a lot of flexibility from one state to another in doing this, and so these pilots for us are a way of really learning kind of what is possible and so forth, and so maybe we can get to that point of making some decision. So I think some attention needs to be focused on the metrics part of it.
DR. SUAREZ: I think one of the main questions is really how can the national committee — I don’t think there is any doubt in anybody’s mind here about the value or the benefit of pursuing these and helping advance this appropriately, and so what we need to think about is what is the best way that the national committee could help advance this, and I think that will be something that we would have to probably take back to our executive committee to better understand and better see.
We have a number of vehicles and ways we can perhaps help advance this, through letters or through other mechanisms, but we will have to probably think about it. But in the meantime, if you have any thoughts about specific things that you would like the national committee to consider to help advance this, we would really appreciate that.
MS. WILLIAMSON: I will say I am looking for ways for our group to overcome some of the barriers, some of the challenges that we have been dealing with, in order to try to move this forward. The biggest one, which everybody knows, is always funding, funding to support more pilots. One of the first things that we hear from states when they express an interest is is there funding to support it, and so that has been a big one. So if there was something behind that, and then Dan spoke to the interest in incentives. If it was incentivized if someone wanted to do it, even being able to do something in a pilot implementation. We are not saying, okay, go ahead and implement the whole thing, do it now, please. You know, just a pilot would help us to advance it.
I mentioned how much we have been able to improve standards just with the trial implementation efforts and Connectathon activities. So if we can really pull this out in a few state pilots, I think it would inform the process and help us to feel even better about moving those draft standards to normative standards at HL7. So that would be one. And then just to be able to move this from a hypothesis. You say the evidence. We will have the evidence then. So that’s the biggest one.
The other is the challenge with vendor engagement. I don’t know if there’s anything that the committee can do, but this is a challenge for us, because vendors are tied to meaningful use. That’s the biggest driver. So if we were there, it wouldn’t be a problem. We would have vendors coming in and investing and engaging, but without something to drive that process, we continue to struggle with trying to get vendors to invest in this process, and I don’t know again if these are things that the committee can help with, but I’m just telling you some of the challenges that we are dealing with.
MR. SCANLON: Maybe they need more recognition in the interoperability roadmap as well, sort of an area where it hasn’t been part of the interoperability framework so far, but showing that it’s, as you say, it’s the bookends for EHR and it’s also well along.
MS. GOSS: To what degree have you done any coordination with the National Governors Association?
MS. WILLIAMSON: We have not reached out to the National Governors Association. The one organization that we have talked to quite a bit about reaching out to is the American Hospital Association and the various state hospital associations, kind of starting from there, asking that they possibly consider this, knowing about this, as they look at the plans for IT implementation efforts and then they can put this in their line item for future implementations. So trying to get that message out, but no, we have not engaged NGA yet.
DR. SUAREZ: We are fortunate to have David now as part of our national committee, so I’m sure he will help us figure out some of those possible ways in which the national committee could move forward.
Vickie, you have a question.
DR. MAYS: Yeah, I was just going to say, because there is another group that I think would be very interested in this, which is the National State Legislators group, and they really are very interested in a lot of like equity issues, data issues. I think if you could get a presentation before them, you really might make a lot of headway. People will probably self-select who then would want to work with you. I’m not sure you can get them all, but that’s a really good group to do a presentation.
DR. SUAREZ: Okay, without any more comments, anybody on the phone wants to make any comments?
DR. ROSS: This is Dave Ross. This sounds gratuitous but nonetheless sincere. I really think what Delton and Michelle and the whole NCHS team has done on this has been very methodical and smart, and we are seeing progress. I’m really impressed with the rate at which they have improved electronic death registration. So I think kudos to your whole team, Delton.
DR. SUAREZ: Thank you, Dave. I think that’s really a great way to express our appreciation on behalf of the national committee for the work that you all have done nationally to move this forward, because this is truly an incredible benefit for the country. Thank you so much for coming, all of you, to do this demonstration. It’s been very, very informative to the national committee.
I think to take words from Bruce, bringing back the V in NCVHS is going to be one thing that we are going to want to do more, and so certainly this offers us an opportunity to consider ways in which the national committee can help move this forward. So thank you so much again.
MS. WILLIAMSON: We appreciate the opportunity and thank you, Bruce, for pushing us to make sure this happened.
(Applause.)
DR. SUAREZ: All right. Of course, yesterday I saved about 30 minutes of our meeting. Now I have to ask your indulgence for a few more minutes before we go to lunch to hear from Vickie about the workgroup on data access and use. So we will be taking a lunch break. Then we will adjourn and move to the workgroup access, but for now, Vickie, I will turn it to you.
Agenda Item: Working Group on Data Access and Use
DR. MAYS: It actually is fairly quick, except I just got an email from NCHS that I have to talk about.
Actually I’m going to get started as she pulls it up, just in case, so we can save time. Part of what I wanted to talk about is what the workgroup is about to do at its meeting today, which is probably the start for us is to think about alignment.
One of the things we have to think about is both in terms of the resources we have and kind of the skillsets that we need sitting around the table is what our actual customer base is and what it is that we need to do. We want to rework our work plan so that we can kind of be clear about what we can accomplish and what we can’t. We had hoped to be further along in terms of our guidance for data access and use, but we have been slowed down.
So the first thing to think about is we have two constituencies that we need to work with, and one is HHS and the other one is the subcommittee, and I actually put them in the priority of our charge, which is in terms of working with HHS, the workgroup was put together to really give the department feedback, ideas, information about data use and access. So that means that one of the groups that we work with is Damon’s group, which is the IDEA Lab, and I want to kind of get us clearer about how we are going to best do that work. Right now he comes in, he does a presentation to us, and we kind of very nimbly respond on our feet to whatever he requests.
So I think we are going to explore a little bit more whether or not there are some specific things that are short-term long-term that he would like to see us engage in, or if he’s real comfortable, we can just stay with come talk to us and we can tell you kind of off the top of our heads, you know, anything that is of concern to you, you can get advice from us. So we are going to kind of explore that a little bit more and see if there is any other way to work with the Idea Lab.
The other thing we talked about, and this seems to be a more substantial piece of work for us, is to talk about guidance for data access and use in terms of the specific datasets that exist within NCH. We talked about that. We kind of did a trial basis of having one come in. We haven’t finished that process. We had them come in, because we were at Hyattsville, so we kind of picked on one and had them come in, but we haven’t figured out how to give them feedback. We haven’t figured out what kind of information we want to collect. So we have talked around these things, but we need to be about finishing it and getting that pulled together.
PARTICIPANT: That was the health interview.
DR. MAYS: Yes, I’m sorry, it was NHIS.
And then we also talked about, well, we want to talk to SAMHSA, and SAMHSA wanted to talk to us as well. So that would kind of be our next one.
But we have to kind of pull this together. I’m not real big on you bring people and kind of talk to them and then everything is nice and there’s no follow-up or anything. That happened months ago. So this is a piece of work that’s on our agenda that should be primary for us, because those are our customers that we have been asked in our charge to respond to.
Then the next group is the subcommittees. So we want to try to be accommodating to the subcommittees as well. We want to try and either do things with the subcommittees kind of on our feet. If there’s something you want to come in and just ask, like, for example, one of the things is the launch plan for privacy. We don’t need to do a long drawn-out thing. I think that we could probably talk about Thunderclap and a couple of other things in terms of giving some suggestions. But those are the kinds of things that as your agendas develop, we are at the table with you and we want to be able to be responsive.
So at the workgroup today, we are going to talk to each of the subcommittees and say can you tell us ahead of time what kinds of things that you see as dovetailing with what our mission is, and then let us try to work them in in terms of our work plans. If it can be like in the meeting you just want to make, get some suggestions from us, we are happy to do that. If you have something that’s more substantial, we really have to make sure that we can do it, because I don’t like start and stop, because it just doesn’t get things done efficiently. So we are going to ask you to talk with us today about ways in which you think we can be helpful.
The other thing is today we are also asked — privacy asked us, you know, to kind of give some comments about the transaction issue in terms of what happens when this is done using social media. Again, those are very easy, very simple, but I think pop is going to talk to us today. We want to see if there’s anything from standards, and we want to see if there’s just anything else that is long-term so that we can plan well. So let’s do the next one.
Products and deliverables. We do number one, which is advising, and that’s what we were asked to do. So we have to remember that. We are going to try to not get into substantial things that we can’t finish. So our group should be nimble. It should be able to just advise and move on.
We want to develop something in terms of the guidance that’s actually a document that we will try to put together, and when I say a document, it’s something that probably we want to see thrown up on the web. We want it publicly available. I don’t think we need letters to the secretary or anything like that.
And then the other is a potential article on those principles and guides, and we talked about that because frequently in this space, where you have developers, entrepreneurs, et cetera, there is an article sometimes in their trade magazine or something. We may think about doing it there or doing it in a journal article or something like that, but it’s — this, in the space that we operate in, people want to kind of be able to understand the pros and cons pretty quickly, and so we want to do something other than to just have a document up.
Datapalooza, we are going to have a discussion about that. I think we have it under control in the sense that Walter is going to be there. He is going to be able to attend on the day in which there will be a co-moderation with Damon for a session in which — what’s the title of the session? Data Lab. We are going to ask him to solve the issue which is privacy, wanted to kind of unveil its work at Datapalooza, and we may have just found the ideal way is to be able to do it. Well, we actually have a few slides that Lily made. So we have to see whether or not we can get the slide deck, because I think the slide deck is due today. So we will discuss that.
Now, the last thing, which I just got an email about — talk about timing — is I had talked with Jennifer Madans, I can’t remember exactly when, about presenting, because I thought we would be further along on this guidance document at the NCHS data users conference, because it’s the perfect place to do it, and I was thinking about it much more as kind of a listening session, where you start talking about what some of the issues are, some of what our thoughts are, and then getting some feedback from people.
We are not as far along as I would feel comfortable about, but they have offered us potentially a slot to talk about NCVHS work. So I would put it on the table. It may not be necessarily that it’s all about us. We might use part of it. It’s not a given. We have to apply. So they asked me to fill out a form and tell them what I want to talk about first. But it — I’m going to kind of suggest a thing, which is that meeting is August 24 and 25. We just talked about doing a retreat in August. If you want it to be, you know, to coordinate it, it would be, that’s the time at which to do our retreat, and then we could go to the data user meeting, or vice versa.
PARTICIPANT: And I think there was also some discussion that the last half of August wouldn’t work for some people.
DR. MAYS: Oh, I didn’t know that. Oh, well, then that’s out, okay.
MS. GOSS: Not necessarily, just it is going to be a balancing act on the schedule. I’m good in August.
DR. MAYS: But anyway, so I would throw that as a possibility. So the question that I need answered is whether or not the committee thinks it wants to do something at the data user meeting.
(Comment off mic.)
DR. MAYS: Oh, I’m sorry, the National Conference on Health Statistics, which is put on by NCHS, and it’s almost always here, and this is where usually all of the data people come. This is where, you know, particularly when you want training on the NCHS datasets. So that’s who the audience will be, and public health comes to this.
DR. SUAREZ: We have done, actually, sessions in the past at the national conference, and I would take the opportunity of doing something. I would certainly appreciate if we have the opportunity to do something, participate in a session or organize a session or whatever. So, I don’t know what would need to be done.
DR. MAYS: Well, we would have to — they sent me a form. So I can send the form. We can talk about it. It’s more whether or not we want to do it, what you would want to present. You know, I’m assuming that, you know, we would to talk about some of our work over the year, and it wouldn’t be at this point just about the guidance, because we are just not there.
DR. SUAREZ: Yeah, I mean, we have enough products to showcase.
DR. MAYS: Okay, then I am going to send you this and then you can kind of determine how to organize it.
So I’m done. I don’t need a whole lot. I’m just giving you a preview. That’s all.
DR. SUAREZ: Any questions for Vickie?
MR. SCANLON: Let me add two little things, friendly amendments to — where HHS is going to be asking the workgroup to give us kind of that reactor advice information to some other websites and some other datasets as well, and the other thing I remember is that ARC is also reinstating its national research conference, which I think is later in the fall, which we might want to think about doing that as well. Let me get the details to everybody.
DR. SUAREZ: Any other questions for Vickie?
DR. COHEN: No questions for Vickie, but Jim, you just brought up another point that I had on my agenda. I think all the work we have been doing today is really very closely connected to health data council activities. So I would like to see us have more of an ongoing relationship with them as a source for feedback from us and information to help them do their job.
DR. SUAREZ: I know that in the past we have presented to the data council.
MR. SCANLON: We still do that.
DR. SUAREZ: Actually Bill presented on the social determinants, but yes, I think for a given set of recommendations or issues, we could have the presentation. At one point we had a liaison or actually the chair of the NCVHS, used to be, come to the Data Council meetings, but the lawyers said it wasn’t — makes us a FACA if anybody else comes on a regular basis. So we can have individual resource and discussions and presenters, but we can’t do it — we couldn’t do it on a regular basis, because then it begins to look like a FACA in their view, so they asked us not to do that anymore.
But I think certainly for individual reports and recommendations and issues, we can do that. We can certainly do that.
DR. MAYS: Can we like get the agenda ahead of time or something and know if we want to say something? Or is the agenda being —
MR. SCANLON: I have to just be sure that we are not crossing the FACA line. That’s my only concern.
DR. SUAREZ: I would just rely on you to identify times and opportunities.
MR. SCANLON: I think it works best if it’s specific issues directed at HHS, but otherwise it’s a FACA issue.
DR. STEAD: Our letter in particular probably ought to go to the council since it recommends — first it goes to the secretary, but it recommends working with the council.
PARTICIPANT: All the recommendation letters go to the data council.
MR. SCANLON: They automatically come to the —
DR. SUAREZ: I think we copy the council on all the letters. The bottom of the letter should say cc: HHS — good, okay. Well, certainly, we will — this group, the group meets every month, I think, right, the data council. So yeah, we will rely on you, Jim, to see which times and opportunities.
MR. SCANLON: June is kind of a sausage-making meeting. So we don’t usually invite to that, because it’s more pre-decisional kind of stuff. But the other meetings are based on issues kinds of events.
DR. SUAREZ: Okay, thank you. I am going to — some of you know I’m a stickler with notes, and I take notes almost verbatim of people, but I just wanted to highlight some of the very critical next steps, very summary, quick, the next two minutes that we have before a full hour for lunch. So I think I am going to just say it very briefly here.
So the executive committee, I think we are going to be discussing — well, first of all, we are going to document the discussion points that we have on this meeting regarding our larger strategic plan and mission and goal development, particularly the guiding comments from Jim. We are going to put them out. Then Debbie, you were going to pull out the statutory requirements and required outcomes from NCHS, and I’ll help with that as much as I can, too.
Then, the smaller group, Alix, Bill, Terri, and me, we are going to get together and refine the strategic framework that was presented, turn it more into the concept of the mission and goals and outcomes and audiences, and then schedule a possible one-day executive committee retreat sometime in mid-August hopefully, so we can work on the scheduling. Then with the intent of preparing our revised strategic plan in September and presenting it to the September group that would include hopefully most of the new members of the national committee.
From there we will define the next steps. I think with the executive committee, we are going to still schedule at least one call to prepare for this and probably more than that, and if we have a retreat, maybe one more or something like that.
The standards, I think, the standards committee, I heard we have the June big hearing. There’s going to be a lot of post-hearing activities to prepare the report and letter or letter and report with the findings and observations and recommendations, and then for the fall, I think there is a number of topics, operating rules, attachments, all payer claims database, the public health standards activity, and this is fall, kind of winter and early next year. So I think those were the things that I kind of heard in summary.
Privacy, certainly prepare the letter from the HIPAA 1179 hearing, finalize and release and disseminate our toolkit, and then consider the next round of activities related to privacy, including the topics that I think Linda presented yesterday, wearables and mobile technology, predictive analytics, working with OCR and ONC on possible additional updates needed for HIPAA, as we heard yesterday, and monitor legislation and legislative action in Congress that modifies HIPAA and affects us, and certainly a big, big area right now of a lot of attention is cybersecurity. So potentially some work around that.
And the population health, I think I heard completing next steps on the framework document that Bill presented yesterday, work on the strategy regarding the IOM vital science report and helping put that all into a context of the larger ecosystem of data, and then have a presentation potentially at the September meeting of the IOM vital science report, and then possibly planning a roundtable or hearing, some sort of a meeting, at the November meeting of the national committee with letters of recommendations to follow up from that.
DR. STEAD: And given logistics, Bruce and I would like to go on and get 11-17 penciled in the calendar on people’s schedule, because if we are going — that would let us do it the day before, and then we would have a hearing for a day and do the — and then we could use a pop health block to do the debrief on what we learned. So that could be an efficient way to do it.
PARTICIPANT: So we are reserving November 17.
DR. SUAREZ: And then an important element now is what we heard about eVitals, and I think an opportunity really for the subcommittee to begin considering what are the kind of things that can be done to help advance this concept. So that’s just one additional item to add to the plate.
And then lastly, jointly working with the standards subcommittee on this public health standards area, which touches also on eVitals, and then we just heard the workgroup on data access, so I’m not going to really repeat what we heard from Vickie, but those are the main action items I have.
Are there any other action items that anybody heard or I missed?
MS. GOSS: More of a clarification of what I heard, in regards to the strategic discussion by the executive committee for a one-day retreat, did you note that there would also be some prep regarding that and how to — beyond the logistics?
DR. SUAREZ: Yes, exactly. We would do prep on that, yeah, conference calls, yes.
Okay, hearing none, I guess I’ll hear a motion to adjourn?
PARTICIPANT: So moved.
PARTICIPANT: Second.
DR. SUAREZ: All right. We are officially adjourned. Thank you very much.
(Whereupon, at 12:20 p.m., the meeting was adjourned.)