[This Transcript is Unedited]
Department of Health and Human Services
National Committee on Vital and Health Statistics
February 23, 2017
Hubert H. Humphrey Building
200 Independence Ave., SW
TABLE OF CONTENTS
- Welcome – William Stead. MD, Chair
- Healthdata.gov Report/Update on Progress and Recommendations – Vickie Mays
- Action Items Follow Up – Chair and Full Committee
- Predictability Roadmap – Alix Goss, Nick Coussoule
- Planning for April 2017 Full Committee Meeting – Chair and Full Committee
- Next Generation Vital Statistics – Bruce Cohen, Dave Ross, Bob Phillips
- Strategic Discussion, Workplan and Next Step – Chair and Full Committee
- Public Comment
P R O C E E D I N G S (8:30 a.m.)
DR. STEAD: Welcome.
(Introductions around table)
DR. STEAD: Let me just briefly say how we are thinking about handling the agenda and then we will start with the healthdata.gov discussion. We are going to do that discussion and manage it so that we bring the input to closure at ten o’clock. Our goal with this discussion is to make sure that we are all onboard with, in particular, recommendations so that Vicki and the workgroup team and Erica can get a final version and form that we can then try to approve on our April call. That is the objective.
At ten o’clock, we will take a few minutes, I hope, to approve the DID letter. Since we were able to get the Pop Health report approved yesterday, we want to free up as much of that 10:00 to 10:45 block as we can to work the report to congress. Then take a break, work the Predictability Roadmap. Come back after lunch, deal with next generation vital statistics, planning for the fall committee, and then the discussion of Beyond HIPAA and terminology. If we cannot get RTC done in the morning, we will have to probably constrict and finish the report to congress in the two to three block. We have got to get all the way through the report to congress today because we have to be able to edit it in a way that we can approve it in April. That is our game plan if that works for people.
DR. MAYS: Thank you very much. In terms of the goal today, it really is for us to focus on the recommendations. Comments that you have about any part of the report, we really appreciate, but what I want to try to do is to spend most of our time on the recommendations so that we can get a sense of whether or not you are on board for that. I think the other parts of the report, mainly factual, we can actually receive edits and comments from you on that. The plan is we will bring this to the April 18th conference call for final approval. What we need help on is making sure that we are landing where we need to land relative to the recommendations today.
First, I have to acknowledge the workgroup. We have had a short amount of time in which we have done a lot of work. Many of them worked during December and January to kind of get drafts of this pulled together. In particular, Erica has been – who is a workgroup member – has been the writer for this and worked long and hard. Some of my workgroup members have travel woes so they are not online, but I just want to acknowledge what they have done. It has really been a team effort. Thank you to everybody.
First thing I want to say is in terms of healthdata.gov, I want everybody to understand we are talking about a platform. I realize that that was really important to start so that people aren’t thinking we are talking about actual data.
What we are going to do today is talk about what is the problem we are trying to solve, what was the charge to us from HHS, and what are the anticipated products. Then I am going to talk about the framing of the report, what it is that is driving the structure so that as you comment, you get a sense of what it is that is the background that is leading us to write the report in the way that we are. Then we will focus on the recommendations. My question is are we at the – somebody said 50,000. I said 30,000-foot perch. I like using perch because that is Bill’s word. I am perching pretty high today, Bill.
Rashida, if you could start us off by talking about what the charge is so that they have a sense of what we have been asked to do?
DR. DORSEY: First, I want to start off with, again, a thank you to the workgroup on HHS Data Access and Use for its valuable efforts in making recommendations about how healthdata.gov can be strengthened and moved forward. As you know, since its inception in 2010, healthdata.gov and its community of users has grown considerably. We are at a place where we are considering how best to preserve its progress and also to continue its momentum. To that end, we were looking for the working group to identify ways in which the Department could enhance healthdata.gov. In particular, as we go through transition, this is a great opportunity to have something in place to share with leadership who will be overseeing healthdata.gov moving forward.
Your report and letter is an important contribution and it really supports the core of the workgroup’s charge to assist and advise the Department of Health and Human Services on recommendations to promote and expand access to, and innovative uses and applications of HHS data to improve health and healthcare. That is important. I thank you for that.
Again, we appreciate your recommendations. We look forward to seeing how they might be shaped through the input of the full committee. Again, Vicki, I thank you and the workgroup members for all of your hard work on this.
DR. MAYS: Thank you. We appreciate being given the chart.
DR. MAYS: The anticipated product from this and Rashida alluded to them, a report. We are trying to keep the report ten pages-ish. We may put some graphics in it to extend it a bit. On top of the report, as would be per our usual way of doing business, we would have a separate letter to the Secretary.
I want to make sure that we keep separate that today is the report and then from the report, we will write the letter to the Secretary. There will be more things in here that we would never put in the letter to the Secretary. At the same time, I want to make sure that people understand we will send something separate to this Secretary.
I am going to go through a series of things that really drive how we structured the report. First one is in 2014 – that was when we had our first chair, Justine Carr, and the group had just begun – what we did was we sent a letter to the Secretary at that time. It really was about use, usability, and timeliness. Usability of the data, in terms of having platforms that were user-friendly. That is part of what we will be really talking about. We made that recommendation. It is as if right now, what we are doing is expanding on the fact that the platform has done well, which is healthdata.gov, but there are ways in which we can maximize and enhance its effectiveness to really meet I think the mission of HHS, in terms of health for all.
The second was the user of the data. That was thinking about data documentation and metadata. That is still an issue. The workgroup has kind of a draft of its earlier work on best practices. That is still a relevant issue for HHS. We are not going to deal with that in this report, only to a minimum amount.
Third thing was usefulness of the data, which talked about the timeliness of the data. That, actually, in terms of HHS, has been implemented. We have heard reports from NCHS about how they are getting data out faster. They are telling people it is provisional, in terms of estimates. The timeliness, I would say, of these three things, is the one that HHS has improved on. Timeliness, for us, is not going to be a big factor in this report as well.
The first issue when we began to talk about this report is thinking about the value of data. I think in this day and time when you are investing data, you want to get a sense of your return on that investment. Helga was really good about pointing out some of the value issues around data. Helga?
DR. RIPPEN: The committee and a lot of the work the committee has done has highlighted the value of data, one example actually being the report that we are going to be finalizing during this session, which is the summary workshop on population health and the importance of providing data to communities and all the way up. I think all of us understand the value from a population health. I think many of us also understand the value from allowing innovation. From some of the Datapalooza meetings that some of you may have attended, they do dog and pony shows of innovative approaches of leveraging the data that is available on this site. Again, that becomes really important. That means providing tools for consumers. There was one where medical students actually provided an application that helped them do differential, which we can talk about later, and then allowed them to identify clinicians that would be in their area. Again, data being pulled together. We also have to remember this site has not only federal data, it also has state and local also, a link to it.
The other one, as far as innovation, then also goes to patient empowerment, being able to look at kind of smooth statistics that are provided on quality. Obviously, for researchers, the way to actually look at different datasets and there is like one place to go. Then also, evidence-based action with regards to having access to some of the data CDC has and others on emerging diseases and things like that. Again, we are talking about a resource that is available that consolidates listing of datasets. It doesn’t have the data, but it actually does a significant amount of work of providing the sites that have the data.
DR. MAYS: The concept of the value of it, we felt was important to emphasize, in terms of why would you want to keep healthdata.gov and not just roll it in. Data is valuable. Data is an e-economy. Data drives jobs. Data drives innovation. For us, that is part of the framing of the report.
I am going to skip the value tags and let Erica talk about that because Chris was going to talk about it, but he is sick.
DR. STEAD: Can I just make one comment, just from reading that section? I think it is going to be important to be very clear which of the examples actually resulted from people using healthdata.gov and which of the examples are examples that occurred and that could have been done with healthdata.gov. To the degree – since you are going to have a relatively short report or letter, narrowing – really trying to get them to be ones that actually came from it would be cleaner. If not, distinguish.
DR. MAYS: That is good.
DR. STEAD: One other thought, if Tom Price has pointed toward opioids, childhood obesity, and mental health, if I remember from your conversation, you’ve got two – you have pieces that could easily be linked to two of those. You might think about the third. You might as well make it relevant because I think it is doable.
DR. MAYS: The other important issue for us is data stewardship. We feel that, one, it is something that we represent in this committee. We bring that expertise. That perspective is important, in terms of thinking about the platform of healthdata.gov. What you are looking at is lots of different – I think there are over 3,000 datasets in there. What you are looking at is being able to have the owners of those datasets take on a certain level of data stewardship and then have the department think about data stewardship in the sense of getting data out for the good. The data that is there, the access to the data that is there is data at the federal level and then we have some state data that is federated to the site.
It gets us back to some of the goals that Pop Health wants to do. We want to make sure that that principle is there and that people think about that. One of our recommendations actually speaks to data stewardship.
Those are the things that we thought were important to serve as drivers for how we frame the report.
In the recommendations – I will read that stuff. I just wanted to make sure I had it up there. I think there is a very important question.
In the recommendations, the question is are these recommendations where you want them to be, in terms of the level? There were tons of things that we went through. We have even other reports, in terms of very specific things about Google analytics and what have you. In terms of what we want to recommend at this point, I am going to ask you to think about and look carefully as to whether or not we are up where we need to be, in terms of a first discussion with the Secretary around healthdata.gov. I think it is really important for us to land where you want to land.
I am going to read something to you – it is not going to take that long – from one of our members because it was a question that he posed that I think is kind of the integration that we often try and achieve in this group. Mark Savage is on a plane right now. He is commenting on having read the whole report. This is kind of the setup for the recommendations.
Having just read the wonderful catalog above, which is the first ten pages of the report, of different things folks have done with individual datasets, maybe what I was trying to suggest was a set of tools, a toolkit, to make it easier for healthdata.gov users to connect or exchange data among different datasets. I am providing strategic and technical assistance to ten communities across the country, trying to pull together data from different sectors – clinical health, public health, criminal justice, housing, schools, environment, et cetera – in ways to help them identify and address ten locally defined health challenges. For example, combined weather and housing and school and clinical data to help target and reduce childhood asthma attacks. Then he gives a URL to the site that talks about this.
There seems to be similarities in these site visits. These ten communities are also trying to do the same thing. Find the datasets they need, create multisectorial exchange – now, where have I heard that before – integrate the data in ways needed to address the respective challenge. Sometimes it is the same kind of asthma data, just for different regions. Sometimes it is the same sector of data, such as housing integrating with different data sectors in different communities, public health workers, or schools, to address different community challenges. Could healthdata.gov aim to help all of these communities together by creating a national health database and tools for community learning health systems. Where have I heard that before?
He says, ONC’s Interoperability Roadmap talks about a national learning health system. I am wondering if healthdata.gov could be a core engine in that effort, a place for communities to plug and play, something to help communities integrate the relevant social determinants of health they need to address local health challenges, reduce health disparities, improve health equity. What becomes important for purposes of these recommendations maybe is not just advances in individual datasets, but advances in how to connect or synthesize them in useful ways for different communities and ask users what they need.
The question he is asking us is to think about could this site be a place where instead of just worry about there is an individual dataset there, could this site in some way build in a toolkit, build in some kind of space for communities to be able to convene – it is probably not the best word, but to convene in some way the knowledge that they have gained, in terms of you are already looking for asthma data, here is where it is, here is an example. I think what he is proposing is something I would like you to think about going into the recommendations.
Should I take questions?
DR. STEAD: I think we need to pause here. Bruce?
DR. COHEN: I think the report and the recommendations are a wonderful start and others, I think a lot way we need to go. I totally agree with Mark’s perspective. I guess when I looked through healthdata.gov, the good news is there is a lot there. The bad news is it is incomprehensible about how to use it. I think the focus of our recommendations need to be what is the recommendation to – the most important thing is figuring out how to make this more user-friendly.
To me, healthdata.gov, I feel like I am walking into a big box store where there are no labels on the shelf. I need that greeter there to help me figure out to get to where I want to go. When I am thinking about that – that also has to do with a lack of understanding about who the users are and who the users could be. If I enter this palace of data, there are some fundamental questions that should help guide me down the aisles. That should be what the user-friendly interface should look like.
It would be – well, let’s take obesity. There is a search engine. It should be divided into maybe a menu that says do you want to find out what datasets we have on obesity, do you want to download data on obesity, do you want to begin a query about obesity? This is the point Mark was making. The value would be increased exponentially if this was not only a compendium of where to go for data, but could actually begin driving people to respond to specific queries. Do you want to find out more about program and services about obesity? The initial point of entry should be user-friendly to channel users to what it is they want to know. There is a search engine, but I think people – only sophisticated users want to go right to databases. If you want a broader audience of users at all levels, it really needs to be more topically driven rather than database driven.
I think the focus should really – of the report should be how do we make this more user-friendly and accessible. I think also another recommendation that I would add would be – the focus here is on HHS, but as we know, a lot of very important health data is not in HHS. Part of the recommendations should be HHS should work with the other executive secretariats to review all of the data that is health related. This gets into the synergy with what we are trying to do with the measurement framework on the community health and well-being.
I love the fact that there is capacity to integrate data that doesn’t belong to the federal government. I think that is a real strength. There should be recommendations around outreach – a more formal process of outreach to incorporate those datasets.
I have a lot of – I think those are my basic concerns. I think the body of the report focuses too much on a couple of examples from New York and California. It would be great if there were federal examples, maybe from NHANES or HIS about how the data and healthdata.gov has helped strive policy recommendations, as well. I will stop right there. Again, I think it is a great place to start. Healthdata.gov could be an incredibly valuable product. It is all there. It is impossible to get to.
DR. MAYS: Can I get clarity from him on the recommendation? When you were talking about there is some federated data in there, was your recommendation we should be asking HHS to invite other people to put their data there?
DR. COHEN: Yes, or establish a mechanism to encourage other people.
MS. HINES: Unfortunately, then you are advocating for data.gov and getting rid of healthdata.gov.
DR. COHEN: I guess pretty much all data are health data from my point of view. As we talk about health and well-being, the Census is probably the most important source of data that we have that goes down to the community level. Certainly, we can refine what the ask is and what we think is reasonable for HHS to be doing. Getting the other data sets that are relevant to health and well-being, however we define that, is important to make this useful.
DR. STEAD: I want to come up one level because as I heard Mark’s suggestions, all of which resonate and all of which resonate with many conversations we have had in the Pop Health workgroups, that is really getting at an end vision in a way. I think if we are going to have an impact with a letter we write now, it has to be at ground level.
Having read everything we did in the last three years in the run up to the report to congress, I came to the conclusion that many of the things we wrote met our aspirations, but were judged non-actionable. I think we, therefore, need to clearly distinguish when are we presenting vision. That is one of our jobs. We should present vision. I prefer that we actually keep that in reports. I think our recommendations, we ought to have a mental model with the government as it is constructed now, who could the Secretary hand this recommendation to? If we can think – we are not supposed to say who that person is in the recommendation, but if we can have that mental model, then that is probably – that is possibly an actionable recommendation. I just would encourage us to have the discipline to keep our – to apply that kind of lens to trying to narrow the recommendations down to what are the most important things framed in a next step from where we are that somebody could take. I think if you look at what we did yesterday with DID, we really worked hard at that. I think we have got to try to work hard at that with this if we are going to be successful.
I just want to try to – you asked us what level do we want to be at. I am really trying to provide the guidance at that level. I think – in this case, I think we are actually probably talking about a letter, not a report – interesting question. The letter – as I read the structure of this, I liked it. I think the idea, a short section, what is healthdata.gov. This is the cleanest description of that I have seen. That is very helpful. Then what are examples? Well, then you say it is remarkably inexpensive. What are examples of its use that have been high value? Then the stewardship piece and then what are recommendations that could be acted on over the next 6 to 18 months within the current sort of structure of the government?
I think if we could pull that off, we would be doing a very valuable job. I think we need to try to get at that level. We note our vision. We can come back to them. Our life will continue. I just would provide that overarching guidance.
DR. MAYS: Thank you.
MR. LANDEN: First off, overview, great approach. It is intriguing. It is fascinating. It just demonstrates so much value. I very much liked the direction. Question, in what you were reading from the comment, there was a quick reference to an ONC – one of their roadmaps that talked about a learning system. That flew by too fast. Could you reread that?
DR. MAYS: ONC’s Interoperability Roadmap talks about a national learning health system. I am wondering if healthdata.gov could be a core engine in that effort, a place for communities to play and plug.
MR. LANDEN: It is putting the – I dealt with the ONC concepts. It is not at all exclusive of interoperability. I have never thought of it in this context before. I will do so. Thank you.
It all plays together, but I am not sure that is where ONC is thinking about going. Check with them. It might be interesting to them as well.
My comments are I very much liked the concept of healthdata.gov having a set of tools or query builders. I guess to Bruce’s thing, is it healthdata.gov or healthcare? I don’t care. I am agnostic. I want a portal that if I am coming in for healthcare, I want access to all of the data. I want the healthcare tools visible to me, a toolkit to build queries or design report. I would also like a library. The example up here about the childhood asthma and here is what we do to combine the data, if somebody builds that, put it in a library. Have a virtual card catalogue so other community groups can come in and say, oh yes, this is what I am looking for. Somebody already built it. I don’t have to reinvent that wheel.
DR. MAYS: Thank you.
DR. MARTIN: This is what I learned with healthdata.gov – so I really appreciate these comments. I am writing them down. There is one issue with sort of what we can do with healthdata.gov. What is unique about it, unlike some of the state and local portals, is that they don’t own the data. All they can do with it is aggregate the information about the data. What you essentially have is a library catalogue of what the data resources are. There is limited capacity to do things like queries.
States that are really ahead like New York and California, they are thinking really hard about how to do queries. It is harder than you would think. Socrata keeps promising to do this. They are a major vendor. They haven’t yet. New York and California are thinking about what they can do.
Something to think through of what is doable, I like this idea that what you can do in 18 months. Unless we want to recommend that they put all of the data on healthdata.gov, that might be something that might be harder to do. You are relying on the portals of say the CDC or other states where they are aggregating stuff to have those tools. I don’t know if that helps with thinking through what is possible.
MR. LANDEN: At this altitude of our perch, I will let others wrestle with the technical, is it feasible within the 6 to 18 months.
DR. STEAD: Thank you. Jim, do you still want to say something?
DR. SORACE: I was just going to quickly say, you know, when you actually sit down and try to write the letters, the thing that you learn about healthdata.gov, we have a vast user community at all different levels. Some of the most important ones are really the power users. They are the ones who are building apps and working with alternative care models to try to improve the value of healthcare. They understand things like ontologies and data models and this type of thing. That is where a lot of our value is, in some ways. Then you have other people who are more aspirational data users, perhaps. They are newer to the game. They may be in junior colleges or what not and they are doing this as course work or what not.
You have a big set of people to satisfy. In some cases, they could be satisfied through better human-computer interactions at the interface level. In some cases, you actually need to like do cross-indexing of databases to tell people what logical chains they can follow. That is where the power users come.
Just keep that in mind. It is sort of a very broad set. We don’t have the resources to support everybody maximally on this probably within HHS. We can’t become the user desk for the nation. We have to rely on probably Academy Health and Health Datapalooza and other groups that we are bringing along on the journey here to be – to have the tacit knowledge and to help people along. That is not to say we don’t have a role in user health, but it can’t flood us.
The second thing – so, there is value in power users. The other thing is there – actually, the biggest consumer of healthdata.gov should be HHS. It is probably one of the better indexes we have of what our data efforts are and how they relate to other data efforts out there. It has a lot of internal value, I think, in terms of supporting HHS, a database. I will just offer those as some observations.
DR. PHILLIPS: I am a huge fan of healthdata.gov. In my former role, I was a power user of healthdata.gov and built one of the applications that won awards at the early Datapaloozas. In looking through the site lately, what I wonder is – I know there are divisions between what you can do and what you can point to, but I wonder if there is a way to help people find an app store that – I mean, even if it takes you to Academy Health site or some off-site that says, look, here are the apps that people are using to search these data by geography. Here are the apps where they help people search the data across datasets by disease.
There are a number of apps out there that do that, but there is no way from the site to find those or to direct people to them that I can tell anyway. I just wonder if there is a way for you all to create that kind of a hand-off to help people find the tools that use your data in much more sophisticated ways, but in an easy platform.
DR. SORACE: It is a very good thought. I think the answer is eventually we could get there and eventually might actually be pretty quickly, but you are talking about exposing data models to the app builders and data dictionaries to the app builders. That, I think, could be done. I think it could be done safely and responsibly. I don’t think we have the mechanism in place to do that. It sounds like a type of recommendation that we could –
DR. RIPPEN: I think you are talking about something different. Right? I think that probably needs to be a third party because otherwise, there is implied sponsorship of an app. I think what – is it correct to say that what I hear is people that already have an app to do something that uses that data gets referenced so that people could look at the apps that use the data, not the app developers.
DR. PHILLIPS: Right. So, we had an application that people could search across 100 different health datasets by selecting the census tracks they were interested in. As they selected the census tracks, it would pull the datasets that had data relevant for those census tracks. When they selected them, they could select the datasets and it would just give them a spreadsheet with all of the data elements tied to those census tracks from those datasets and they were off and running. There are several apps like that out there. I just wonder if there is a way on healthdata.gov to either direct people to useful applications or to hand them off to someone else, a third party, who can say here are applications that other users have found helpful for looking across the data.
DR. STEAD: If we have captured the idea, let’s continue the discussion. Helga, did you have a different point?
DR. RIPPEN: It is the question also of standardization of summarizing the data. Again, if you have kind of the listing of all of these great data sources and I know we have talked about it before, but this might be an opportunity to include that as a potential recommendation, which helps facilitate searching. If there is a common way to summarize the datasets, then now you have a standard way of approaching the search and you will be confident that the search will be consistent. Especially now that we have kind of a – the framework for the population health kind of component that could tie in very nicely, too.
DR. STEAD: Thank you.
MS. KLOSS: I am going to suggest that the recommendation regarding data stewardship, certainly important, but I would suggest it be strengthened. In recommendation number three, there is reference to establish a governance process. That seems to be delinked from recommendation five, which is stewardship. I think they go hand in glove. I would suggest that recommendation on stewardship be retitled formalized governance and data stewardship practices. There are things that are threaded through here that really belong in governance and data stewardship.
One of the really good references, I think, that we should bring into this is that new report from NIST on deidentifying government datasets. It seems like this is one of the first products that can really put those recommendations into practice. It would be wise to explore that. For example, some of the things that should be in this stewardship are the life cycle management of the data, the policies, including issues like suppression that we have talked about, review of methods of deidentification, even thinking about how to staff this and whether there needs to be some kind of disclosure review board.
I think right now, the stewardship is really important, but it reads like an afterthought. I think it probably needs to be a little more central.
DR. MAYS: Thank you.
DR. COHEN: Just a follow-up on a couple suggestions. I think Bob’s idea is great. There are a lot of – if you just had a radio button that could direct people to the existing web-based query tools, SEER Stat, WONDER, the NCHS tools, so people who want to go right there can find how to do that that is great.
The second is a lot of the data we are talking – although the owners are within HHS, they are already aggregated in a way that I think this would be a wonderful challenge to app developers to see if they could combine the data from SEER and from Vital Statistics and from HIS and BRFSS in a one-stop search engine query kind of tool.
Third, the point that you made, Jim, about power users, it is clear to you who some – who those users are and that user community. Why don’t you make it clear on the site? Essentially, naïve users versus power users can be directed to channels that are much more useful to them. If you are a power user, you know where to go. If you are a naïve user, you have to go through all of this stuff to finally find out where to be. I think embedded in your logic in healthdata.gov, it is all there, but you need to make it easier for people to find themselves. There are ways to do that, in terms of recommendations that I don’t think are too costly or onerous.
DR. MAYS: I am going to let Erica get her slides up. We are actually going to go through the specific recommendations.
DR. FRANCIS: I thought the suggestion about the overarching governance with stewardship and the questions about deidentification, I think that is a great suggestion.
DR. MAYS: I do, too. Just something I am going to put out there to think about. That is whether or not some relationship – we can think about a recommendation – some relationship with the National Library of Medicine. Many of the things you are asking us to do, it would be great, for example, if simple things like the National Library of Medicine said what dataset is connected with an article so that people understood where to go look for it.
PARTICIPANT: I hear a recommendation. Seriously. That is actionable.
DR. MAYS: Yes. Some of the stuff that we have talked about, it really should be in, I think, the National Library of Medicine. I don’t – again, if you think we can make that recommendation without consulting them or whether we consult them, just let us know. I think a lot of this could be populated by working with them and using their search engine.
DR. STEAD: I don’t think we want to name in the report. I think what we want to do is recommend it in a clear enough way that it would be obvious to the staff or the Secretary that they should hand this to the NLM to say whether they could do it or not.
MS. HINES: I generally agree with you as a principle. In this case, I don’t know why we wouldn’t. It is like NASA. There is only one of them.
DR. MAYS: We actually had some conversations with them. We will work that out. Erika, do you want to now take us through the specific recommendations? We already know something about three and five being kind of clustered in some way.
DR. MARTIN: What I wanted to do is give you a sense of where these recommendations came from. Just like a little process thing, I am taking a lot of notes on what people are writing. I would also like hard copy comments. If you write them, please put your full name at the top and an email address so I can follow up with you if I have questions.
To start with, I actually had a lot of conversations with Chris Boone, who was previously running the Health Data Consortium about how do we have this sort of guiding vision and mission statement. The idea here is that your vision is really aspirational. You will never get there, but it is where you want to go. The mission is more about how you do it. That is really underlying a lot of the content in the report and what is driving the recommendations and is also what is driving the section about its early successes. It is also a way to make the report stick a bit more and make it more clear.
The vision behind this is that healthdata.gov drives economic empowerment, innovation, and transformation in health. We thought that those were really good aspirations and things that the new administration might be aligned with. Then the mission statement is that it does this by facilitating the transformation of public data into health innovation by powering visibility, competition, community learning, research, and evidence-based decision making. When you read the section that is around early successes, that is why things are grouped into these different categories.
DR. MAYS: I have a quick question of clarification. Is that currently healthdata.gov’s – so this is what we are -?
DR. MARTIN: Yes. So, when I shared all of our meeting notes with Chris and thought through how do we make this more impactful, he said what you guys talked about wasn’t a tagline. This is actually you guys created a mission statement. That is really what this is coming from. It is sort of I think our workgroup’s recommendation on where healthdata.gov should go.
Underlying this is that – this gets into the context, which is that once thing that has happened is that, you know, there is this initial developmental phase. Healthdata.gov was really innovative in being the first in the nation to do something like this. It started this developmental phase. It now covers all of HHS operating divisions. About 25 percent of its products are actually from state and local agencies, which is pretty unique.
What has already happened is that throughout this really organic process, there is this nation open data ecosystem. What that means is that there is a bunch of users in a community so there are data entrepreneurs, community groups. It is HHS. The idea is that you get all of these people, especially outside HHS, who are involved in playing with the data and not only do they create products, but they make better data. They put their apps out. They put data transformations. That makes it more usable and fit for us for other users, as well. That is sort of this idea that there is a cycle.
What is sort of interesting is that even though this has gone through this very unique developmental phase, it has developed organically, it has actually already achieved really important outcomes related to the vision and mission. Hopefully, that comes out in the report.
Where the recommendations come from is that we are thinking through the report and our charge is how do we enable it to reach its full potential? How do we get healthdata.gov to really hit this aspirational goal? How do we make sure that it is sustainable in the long term so that this isn’t just sort of an initiative of the past administration, but it is something that will be deeply embedded and something moving forward? Also, how do we make the future activities more strategic? Rather than this sort of organic evolution, which worked really well and had important outcomes, how do we tailor it so that it is really hitting these high-level goals? That is where the recommendations are coming from.
Where these recommendations come from is that we have a workgroup with very diverse expertise. We had members who were talking about best practices in the field. These are entrepreneurs, other sort of high-level data users. We had ideas from other concurrent activities of the workgroup. This is where things like your past data stewardship report come in. We asked the outgoing and current leadership to provide some wish lists in terms of what worked well in the past, where do they want to go.
We also relied on peer-reviewed literature. I have actually been doing some work in this for the last few years. Your point about there are too many examples from New York is very well taken. New York and other early innovators, such as Chicago and California, are just the ones that have really done a lot of this. There is literature on that. Also, we pulled from grey literature that have described some use cases, so news articles, white papers, really things that we can point to and cite in the report.
Also, I spoke with New York and California – since these are really leading state agencies, the only states that actually have a dedicated open health data portal, so it is very similar to the concept of healthdata.gov. Their leadership is kind of similar to sort of HHS leadership. These conversations are really focused on the recommendations. I actually felt a little better when we talked to the undersecretary of the Health and Human Services Agency in California, which is sort of analogous to incoming Secretary Tom Price. He said that he felt like the recommendations were pretty spot on, other than having a few other suggestions. That is sort of where these have come from.
High level overview of the recommendations. So, these are not in any order. The only thing that would be maybe ordered was maybe to possibly consider putting the program strategy one first because then it makes more sense where the others come from. The others have no ordering.
There is one strategy that is really – one recommendation that is about strategy. This would be saying how can we move this towards a more clear vision and strategically focus on that end goal, rather than doing this in the organic process that has worked well up until now?
There is a recommendation that is really about how do we improve capabilities for external users? This is actually great because a lot of the members here have already made some comments. This is one of the challenges of healthdata.gov and where we want to go next. We have a recommendation around external users.
We have a recommendation around internal capabilities. This would be recommendations for what could HHS do to build its own workforce, to make sure that it has the capabilities to do the technical stuff and to really ensure that HHS could be using the resources of their own product.
We have a recommendation on understanding users’ needs to tailor the products. This is actually something that has come up. The idea that the number of users is so diverse, what is unique about health data portals and the notion of an ecosystem is that HHS and other state agencies are very used to interacting with researchers, so people like me. They are not so used to interacting with data journalists, community groups, new users to the system. There is a lot of work in thinking through who are these users, what do they want, how do you make the products usable to them. That is a recommendation.
The fifth one is this commitment to data stewardship. I already took notes on the idea that maybe that should be adjusted to be more about governance and data stewardship.
So, that is kind of the high-level piece.
DR. MAYS: Erica, I am going to make a suggestion that we just go right to the recommendations. Just start with one and all of its pieces rather than going through each one separately.
DR. MARTIN: The first one, then, is about strategy. That is developing an integrated and coordinated strategy with HHS operating divisions, states, and local municipalities to advance the mission and vision. This is trying to make this more strategic. One of them was promoting this as a brand. You want to get brand recognition so that people want to come to this site. They know what healthdata.gov is. That is where one recommendation could be adding a tagline, such as public data unleashed. This would be so that people know that this is what healthdata.gov is. It is to make it memorable and to make people come back.
The other one is facilitating a learning collaborative with states and local municipalities to share lessons, such as sustainable business models. Inspiring data customers to do more with the data by highlighting success stories, encouraging data custodians to post blogs and other reports. This is something that certainly New York and California have spent a lot of time doing because not only does it get external users excited, but it also gets your internal users just as excited in getting the ball rolling moving forward.
Launching a public data accelerator to increase the number of innovators and data entrepreneurs using public data. That would be getting the data out in the community, facilitating that.
Supporting developer challenge competitions. This is actually one of the pieces of feedback was the idea of wouldn’t it be great to have a data entrepreneur kind of figure out how to do some of these query tools. These challenge competitions could be useful for tapping outside expertise. It is also really helpful for just cultivating the health data ecosystem and really building your user community.
The last one is establishing mechanisms to better integrate this into broader HHS data efforts. New York did this. They actually embedded their Health Data New York into a program area so that when their commissioner left, it would still remain moving forward.
The second one is around enhancing the platform’s capabilities to make the data more meaningful to a range of data customers and extend its reach. This was the idea of we want to make the platform more accessible, more user-friendly. One of them is keeping the website in a modern design so people perceive it as being fresh. A second one is implementing additional web features. The committee already had a report around this. There are things like providing sort of a feature of other users who use this also look at. This is what Amazon.com does when you shop. Allowing users to identify their roles. This would help them identify and say I am a researcher and I am really wanting to sort of do a project on this. That might change the kind of query that you would get. Then also improving the automated processes that are used to extract the information from the open data portals that house the data. This is a pretty technical thing.
In terms of helping people discover data better, showing relationships between datasets so you would know that, okay, NHANES actually has multiple years associated with it. You might want to look at not just this one piece, but all of them. Publishing data models. Making the data documentation consistent across HHS datasets.
Then this is where – we also folded in Mark Savage’s point about maybe this could also include developing a toolkit to help communities synthesize and integrate multiple datasets. The other recommendations are about how do you get one user to use a few different datasets, where this would be more community focused with the toolkit.
The third one is around developing a long-term internal infrastructure to allow HHS to tap into state of the art knowledge about big data, machine learning, and algorithms so that they can increase customer use. This is the internal focus recommendation.
DR. STEAD: Back up one if you don’t mind. I just question whether the last bullet – that is the kind of thing that Monti and all of those people are doing. I don’t think that means that healthdata.gov couldn’t or shouldn’t go there. I don’t know whether that begins to dilute us too much. That is a place where we might consider that if we could get the first things done, we could then come back to this question, depending to agree on how the rest of the community health data work’s ecosystem is evolving. I just ask the question because that is –
DR. MAYS: I think that these are the recommendations as they are, but I think reengineering them, what would really happen is Mark was actually suggesting that the first recommendation is to really think about healthdata.gov as facilitating the community health learning system. It was a much broader kind of recommendation. Rather than it being a piece, he is actually saying that that is what he thinks would be a major contribution, is not just focusing on the individuals, but focusing on ways in which communities, et cetera. That would be what would make it unique.
MS. HINES: We just have to keep in mind the power users that we are not aware of. We don’t want to mess that up.
DR. STEAD: Let’s just try to briefly run around, trying to stay focused on this question, of whether that should be there or not. I get that it is there. I get where it came from.
DR. MAYS: He was suggesting not putting it there. He was suggesting making it the first recommendation. I want to make sure, Erica, that we have enough time to –
DR. STEAD: Other issues related to this one question?
DR. COHEN: I just suggest, if you haven’t had a chance, look at the March/April 2006 Volume 12 issue of the Journal of Public Health Management. It is entirely devoted to the development of data query systems by states, their functionality, how they work, evaluation and measurement. It is a wonderful resource that provides context that might help you understand whether we want to be doing this as well.
DR. PHILLIPS: I almost sense that you need to figure out what is it you do and what is it you support. You almost need to set some boundaries of we work to this interface and then we support people in going beyond this. Your resources are limited and may be getting even more so. What are the fundamental things we really have to preserve that we do? What is it that we can enable that gets done what needs to get done? You are a previous resource.
DR. STEAD: That is a really good point. That is another guiding thing we can use as we try to land this airplane.
DR. ROSS: I don’t want to beat this horse to death, but where I am struggling is exactly to Bob’s point. I come back. I am listening. How far do we go with this? I am thinking of it the way I do, running a non-profit business, but a business nonetheless. I have to think about what is the value proposition here? I ask myself the question if you unplugged today, turned it off, who screams? Who cries out? What value are you creating today that puts pain on somebody who would say something about it? In other words, how do they value it?
To our chairman’s admonition at the beginning about what is vision versus what could be done, I think what Bob said says it easily. I think it is very enticing to move down the path to grow something very big. If I understand it right, you don’t actually have a legislative mandate to make this happen. It is about either, A, establishing a vision and offering that vision forward to see if the current administration or the congress, somebody wants to actually solidify that in an ongoing way to sustain it. When you start things like this, as you know, and they get bigger, the cost of sustaining keeps getting larger. Maintaining these sites gets bigger. Reworking them every couple years, which they have to be done, gets bigger.
I find myself going where Rebecca went, that is, well, NLM has a role in this. They actually have a formalized seat in a place at the table about all of this. That is something maybe we have to think about. Absolutely, I think we ought to parse out the vision part from what really is doable within resources today, acceptable, that makes it supporting immediate impact. There are two really big different things going on here.
DR. RIPPEN: This actually follows. It is going to be the big question of – one of the key things is it is inexpensive to maintain, but if you want to add how much this stuff is going to cost and then how much that is going to cost to maintain, I think that is going to be a challenge. I think the big, overarching thing is, yeah, who is going to scream? It is a really amazing, valuable sort of thing. It is the question of how do you ensure its sustainability.
DR. SORACE: I suspect two things. The first thing, I think we may just need to keep better information inventories in the HHS for our own use and then the publicly available site exposing that inventory becomes healthdata.gov. You are serving the agency and we are doing it in a manner that is transparent is electronically FOIA-able, so to speak. That is one road to go down.
The second thing is I really do think that the NLM, at the end of the day – because also, they are a massive producer of a lot of these datasets. Their expertise in ontologies and supporting complex user queries is an important player.
DR. MARTIN: So the third one is around developing a long-term infrastructure at healthdata.gov to really allow it to tap into state of the art knowledge about big data, machine learning, and algorithms for increasing customer use. One of them is about providing a foundational infrastructure to enable long-term planning and use of modern technologies to support this mission. Ensuring that HHS has the technical capacity to use machine learning, data analytics, information, and communication technologies and other state of the art scientific techniques to support user needs. We didn’t want to say you should train people, but that is sort of the idea of workforce development.
Establishing a governance process to establish standards for the datasets that are referenced in other aspects of healthdata.gov. I took notes that maybe this should be moved to the stewardship. This is actually one that California really strongly recommended that we add to it. They would not take on a data portal until they had governance in place. It was the same thing with Health Data New York. Part of why it is so successful is they had an open data handbook. They had processes in place. There is sort of a whole process for what you do. That has really kept the sites successful and high quality.
Developing some business practices to allow operating divisions to keep the data comprehensive and timely. Better understand and support user needs and meet cyber security and privacy regulations. Again, this is just about processes for doing this.
It would be useful to get input on this because this is an ask. We would be creating a Chief Data Officer, who can oversee healthdata.gov activities and routinely engage staff within operating divisions to solicit their input, provide feedback on how data has been used, and have a sustained commitment to releasing public data. Where this comes from is that California and New York have done extensive work where their senior executives have really gone door to door to talk to the data owners about what they have been doing.
This is important in their minds from the interviews that I have done because having a top-down mandate doesn’t really work if the top person tells you to do something. You really have to keep your data owners engaged because then they will start producing things. They will write blogs. They will sort of be inspired to think differently about how their data can be used.
Redefining some workforce roles to enable more active data customer support and foster health data community growth and a learning health system.
Engaging HHS in long-term data standardization activities to increase the ability to link datasets across operating divisions and decrease duplication. That had been mentioned. If I don’t have that captured correctly, please send me a note and we will make sure to edit that in the report.
MR. COUSSOULE: This is a little more of a structural question. When I look at this, this recommendation seems to create a very broad perspective on kind of what HHS should do, that data may be curated a little differently all the way up to the tools and techniques to access the data. This, to me, reads as very broad. I struggle a little bit with the point that is being made to get there.
If I look at it and I say is the value of healthdata.gov the data? No, the value is access to and information about the data to be able to use and leverage. Then this one gets into what HHS should do kind of behind healthdata.gov and not just healthdata.gov. I struggle with what we are trying to accomplish with that one. I am not saying any of it is bad, by the way. I am just saying it is a very different thing from saying here is how we want to make the capabilities to be able to understand and access the data versus here is what HHS should do behind the scenes.
DR. RIPPEN: Actually, I want to build on that. I think that there are two buckets. One is what currently exists, which is a great resource for people to access. Then another is really kind of does HHS want to have kind of a more formal process across all of its datasets? This implies a Chief Data Officer going across all of the agencies. That is a pretty significant thing, the whole governance, of making sure that HHS now actually serves as kind of using the NIST in their practices, which is a bigger issue. It is a very different sort of a thing. It might be an important thing to do anyway. It is just a different step.
MS. KLOSS: My concern is that, generally, in our recommendations, we don’t get to the level of recommending how something be staffed or how money is spent. This just seems too specific. I think if we could step that up another level and describe what the elements are of governance and data management process and leave the staffing decisions specifics out of it.
DR. STEAD: I resonate with these last couple of comments. I think we almost should set aside for – as a separate thing that NCVHS might ought to come back and look at, are the things that HHS should be considering to make its data efforts more robust, sustainable, and cross-connecting. Not layer all of that into a letter that we are trying to come to closure on conceptually in the next 15 minutes and in terms of wording, by April.
I really think if we could tease these two things apart that really might put us in a much better plane and really focus this on healthdata.gov with HHS as HHS is. I think your idea – I loved the idea that there should be an internal index that is exposed. I think that is actually within the frame of what we are talking about from HHS.gov, but all the rest below that seems to me to be a bridge too far. Just might consider that.
DR. MAYS: I think that is really helpful, is for us to think about here are the aspirational things for HHS, but it sounds like what we need to do is to really pull back and say this one thing, what can we do about healthdata.gov. Okay. I appreciate it.
MS. GOSS: I am just going to add to what everybody else was just saying in that last thread. A service that can be provided to the industry gets back to those core governance – what do you need to do, how do you need to prepare yourself. That, to me, seems like some of our toolkit and stewardship roles. They seem like they should be separated. HHS lessons could really be helpful in guiding those.
DR. MAYS: The kind of three and five, kind of pulling that together, Alex, I think – I am hoping that that will answer what you are saying.
DR. STEAD: I think we are all agreeing that, yes, healthdata.gov needs a governance structure and process. That is within the scope of recommendations related to healthdata.gov. That ties into Linda’s stewardship comments.
DR. MARTIN: Number four is a little easier to work through. So, the fourth one, again, gets at this idea of there is a lot of different users, new users to this ecosystem. We just don’t know their needs. It is implementing some mechanisms to solicit their input regularly to optimize their use of healthdata.gov and develop the development of data driven health innovations.
The first one is creating a data product team. This is actually done in industry with sites like Amazon.com. To engage regularly with diverse data customers from the private, public, academic, and other sectors to solicit their input on how to improve usability, access, and fitness for use. This is just the idea of making this a really routine activity rather than ad hoc, such as what are conversations that we have had at the Health Datapalooza.
The second one is around solicit stakeholders’ input systematically through surveys and focus groups. This is something that New York and California are really starting to do right now. They are realizing that they don’t know their users. What can they do to try to meet the needs of diverse users? Is it possible that not all needs can be met so they can use this to help target who would be the right groups to be focusing on and how?
Then the third one is that just using web analytics to improve our understanding of healthdata.gov customers and their use patterns. This can facilitate how we tailor content to users. We can evaluate dissemination strategies. We can evaluate how we are marketing data to specific groups. We can use this to increase the findability of data and other benefits.
What I mean by this is that you can take something like Google Analytics, which is already there. You can test out different features on your website. So, with AB testing, you could say let’s try today change it to this other version. You can actually use the web analytics to track things like do people stay on the site longer.
All of these things are doable. The recommendation is just to do it as a recommendation and the same thing with the surveys and focus groups.
DR. RIPPEN: I think one of the things that we have to just figure out is positioning. There are two different views. One is, hey, we have this really great data from a federal government perspective and also from the states that we are providing that we want to use, an internal mechanism. We are just going to do two for one. We are focusing on what we need, but then we are letting others use it. The other is, no, this is a service to the public. They are two somewhat different because the argument and the positioning is very different. If we are saying it is a service, then we have to identify the demand. People are demanding this from the government and they want it. Then we can go and say, well, how do we make sure that demand is met.
Which way, I guess – what is the recommendation? Is it that we are trying to say, hey, this is something the government is going and it adds value so not only is it meeting the mission of the federal government, it is actually providing an opportunity for innovation at low cost? Or there is a demand for this and we need to do work and spend money?
DR. MAYS: I think it is the second. Data is collected for a purpose. The purpose is to give it back. You kind of make these agreements with the public, in terms of collecting it. I think there is a demand that is there, as you can hear from Pop. I would say I think it is the second. Other people may have a different opinion.
DR. RIPPEN: We just have to make sure that it is included as a very clear demand.
MR. COUSSOULE: I think it is a really important point to create a feedback loop that does kind of two things. One is to – you know, the efficacy of healthdata.gov, itself. Also, potentially a feedback loop into the original data source owners to understand the usefulness and potential either adjustments or improvements into that data source. I think this is a really important point to create that feedback mechanism. It doesn’t have to be particularly complicated, but it lets you tie into, yeah, healthdata.gov can be better used this way and this way and this way and, oh, by the way, this data source is really useful to me. Here is what might make it better or might make a difference.
DR. MARTIN: I completely agree. This is certainly what New York and California have been doing. New York had regular meetings with data owners, where they said this is exactly what we have been doing. California actually has a regular working group every month. 25 people keep showing up for this thing.
Do you think that that should be something that is moved to number four, as a bullet point?
MR. COUSSOULE: I was not sure if it was trying to get to that or not. I do think –
DR. MARTIN: It was somewhere else, but maybe it could be moved in here.
MR. COUSSOULE: Again, I am probably less about the actual data, itself, and more about the can I get to what I need and is what I need useful for me and what would make it better is I can provide that feedback loop. It may be slightly different than what you are addressing here, but that is my thought process.
MS. LOVE: This is just my education. I would assume that each of the data owners are soliciting that feedback anyway, right?
DR. COHEN: That is a bad assumption. No.
MS. LOVE: I guess I am really shocked. Having run a data system – I mean your users are always clamoring for more or more timely.
DR. MAYS: We usually see them at a meeting and that is when we tell them things that we are upset about.
MS. LOVE: So, we are really hitting a more fundamental issue.
DR. COHEN: Informal and anecdotal.
DR. MARTIN: Good point. That is sort of why this was in there. This is why the feedback loops were particularly helpful in New York and California.
So, then the fifth one, I guess would now also be revised to be about sort of a little bit more governance, as well. It is making sure to implement data stewardship practices to keep the commitment to the public to protect health data security and confidentiality. The idea behind this is just to make sure that it is sort of well-known and understood and that we keep this commitment to we are not just posting data willy-nilly. We are actually making sure that it adheres to these important standards.
One of the recommendations include assigning responsibility for data stewardship. Designing processes for monitoring data recombination and the risk of the mosaic and other similar effects that would permit identification or stigmatization of individuals or small groups. The idea that as more data are out there, it is easier to combine things and match them up. It is easier to potentially discover individuals. Designing processes to monitor downstream data transfers and use to avoid reidentification risks or violation of other principles of data stewardship.
DR. MAYS: In term of these five recommendations, I just want to recap things that I heard. First, I think we need to separate out kind of the things that are general and really focus on this is healthdata.gov and that is what we are really focused on.
Some of the other things we might give back to the committee for thoughts, in terms of HHS data, in general. I think we will try to do that. The other thing is, in terms of raising these recommendations up, I think we need to think more strategically, in terms of exactly part of what you are saying is what is the goal here. Are we talking about this is yours and you are making it available? I think that is going to be important for the administration, in terms of thinking about it. We are talking about something that is a public service, that is a public good. It has a lot of demand. We need to be able to make sure that we are meeting that demand and in meeting that demand, that it does all of these other things, such as make health better, create jobs, et cetera.
I think I would like to hear, just in the few minutes we have left, your thoughts about that. I think that is an important direction and framing we need to consider.
DR. ROSS: I think the articulation of the public good is a very critical step. We, ultimately, have to differentiate what is inherently a governmental responsibility or need that the marketplace just isn’t going to meet. Connecting and service HHS agencies better is something nobody else is going to do. It needs to be done. Somebody has got to do it. There is a real need for that.
Moving farther out to the communities and to the broad universe of users who really have to stop and say at what point does the commercial marketplace have an opportunity to service this versus us, the government, becoming a business entity for this? Certainly, if you go forward for appropriations that would sustain a significant investment, congress is going to ask that question. We might as well ask it ourselves.
Back to what Bill said originally, let’s get the vision part of this separated from some more immediate action. I think a big piece of what we are talking about here is a potential restatement of a vision. That next vision is going to have to be bought into by the current congress and the new administration, I think, if this is going to go as broad as this is being described.
DR. STEAD: To not mislead, I think the – suggesting a potential vision mission statement actually fits within the context of a recommendation that is related to healthdata.gov. I think the draft you have got is very good. The – I think Dave’s frame is another lens on how we narrow the best we can in on what is actionable to healthdata.gov now. One lens to apply to that is the marketplace is not going to do it. I think that your clarification that HHS – healthdata.gov has shown how powerful an internal HHS index would be. That is almost the knowledge bridge between healthdata.gov out here and the stuff back within HHS that we are going to save for our next round of work of the full committee at some juncture when our white space comes up. I think that is maybe a way to think about parsing it.
We are really out of time. Let’s let Helga have the last word.
DR. RIPPEN: I think that also just making sure that there is something that summarizes the process and the input. I mean I think about all the people that actually have been involved in all the conversations with everybody and the outreach, just as far as – this is not just something that just a few people in a room came up with. There was a process. Very, very many people who represent many different organizations and perspectives were included.
DR. STEAD: I just want to thank Vicki, Jim, Erica, and the workgroup for adapting to our requests and, in essence, managing this as a mainstream committee activity. I hope the robustness of this engagement with the committee helps everybody understand how important it is to mainstream. We could never have gotten this level of conversation any other way. I hope people feel really good about it. It was a big shift. I think – I just could not be more grateful for everyone making it happen.
DR. MAYS: Let me just say thank you, in particular to Erica and Jim, in terms of – Jim has had kind of insights that we would have been kind of falling off in a hole without. I really have appreciated his guidance and for Erica’s knowledge in this area, knowing who to go to to talk to. It has just been very helpful to us.
DR. MARTIN: I know it is your last word, but I just want to solicit any written comments or emailed comment. My email is email@example.com. If you have written comments, it is very helpful if you put on your full name and your email address. That way I can contact you just to kind of share back what I did and also make sure that I have captured your ideas correctly.
DR. STEAD: Thank you. What we want to do now, each of us should have receive by email in the wee hours of the morning, the revision based on yesterday’s comments. We are going to let Linda walk us through a process in which we will see if we are comfortable with those edits and see if we are in a position to approach the letter.
MS. KLOSS: First of all, I would like to thank Mia and Rachel for taking all of the discussion yesterday and going offline and working yesterday afternoon to get it all incorporated. As I looked at it last night, I changed two periods and one semicolon. That is the quality of the work they did. They nailed what we were looking at.
We added Honorable Thomas E. Price, M.D., Secretary. Had no further changes to page one.
Page two, we accepted that insertion in the first paragraph under major findings. Remember, yesterday, we moved in that once deidentified PHI is transformed. That insertion was accepted.
Continuing to scroll. We include the highlights. You will see what changes were made.
MS. HINES: Let me read Barbara’s suggestion on seven. She says, at line 113 and again at line 544, it refers to redisclosure risk assessment. Just wanted to make sure that was what you intended as opposed to reidentification risk assessment.
DR. STEAD: I think it should be reidentification risk assessment.
MS. KLOSS: Correct. It should be.
MS. HINES: She also has some other things, but I think if you just look –
MS. KLOSS: They are editorial. All right. We will most certainly accept her change in recommendation number seven, here, and at the end, reidentification risk assessment. Correct.
Recommendation eight, we changed that to –
MS. SEEGER: From patients to individuals with the understanding that covered entities also include health plans and clearinghouses. When the Notice of Privacy Practices is rendered, it is not only rendered to patients. It is also rendered to beneficiaries of health plans, to their members.
MS. KLOSS: Okay, recommendation nine, we changed deidentification data holders to recipients of deidentified datasets. Eleven, we softened that by stating –
MS. BERNSTEIN: Can we go back to 8 for just a sec. There was another change. We made the technical change that Rachel described, but also the language of 8 was changed to accommodate the committees’ concern about how we wanted to – what it was in the Notice of Private Practices, what you really wanted to do, which is to tell people that their information might be deidentified and re-disclosed, not about the process of deidentification, and so forth.
MS. KLOSS: Eleven, we softened that to include investigate the feasibility of requiring covered entities and business associates to track disclosures. We removed the investigation with regard to EHR technology.
There were a couple of changes to the narrative. This is – you added this. Did you want to describe that?
MS. SEEGER: We did. Maya, we are now on line 164, the nod to NIST. One of the only efforts in town with respect to deidentification, is the HIPAA privacy rule. Just acknowledging that many different types of organizations look to the HIPAA privacy rule, deidentification standard and the absence of their own specific standard for which they have a business need.
MS. BERNSTEIN: This was responding to Bruce’s comment that we talk about states and so forth. We said that we would put in something general here to explain that we recognize that other kinds of entities that are not covered use these standards as well.
DR. COHEN: Thanks. That’s wonderful.
MS. BERNSTEIN: Also, I dropped a footnote there about showing that NIST had started doing this in 2015.
MS. KLOSS: We removed that straggling sentence about the OCR data. No complaints?
This is the header sentence that makes sense of that list of issues contained in that paragraph. There are multiple significant risks that the DID standard does not address, for example, first. We might want to spell out deidentification, but that is editorial. We will do that in the final.
MS. BERNSTEIN: Yes. I will do that.
MS. KLOSS: So, first, second, third.
DR. STEAD: Much improved.
MS. KLOSS: These are just numeric changes in reference to standards and leveraged.
MS. BERNSTEIN: That is a word that arose during the conversation. I liked it better.
MS. KLOSS: Yes. I liked it better, too. Here is the competency guide. It would be a resource to define for both covered entities and deidentification practitioners, the capabilities, proficiencies, and aptitudes that practitioners should have in order to properly apply deidentification methods to PHI. I think that – then that final sentence there, further, a written competency guide could be a first step in defining competencies necessary –
DR. STEAD: Change that to to a for. Necessary for any voluntary certification program.
MS. BERNSTEIN: I see it.
DR. STEAD: End of line 506.
MS. BERNSTEIN: I have it. Oh, 506. It might be 506 over there. As we make changes the line numbers change slightly.
MS. KLOSS: And I think that is it. Is it not?
MS. BERNSTEIN: On 544 –
MS. KLOSS: The change to recommendation number seven, the reidentification.
MS. BERNSTEIN: Right. The others are just – yes. They just mirror what we did above. In the discussion after recommendation eight, use the word transparency.
MS. KLOSS: We are going to scroll to that next.
MS. HINES: Maya, just make sure that Barbara’s edit on 549 gets implemented there, too.
MS. BERNSTEIN: I got it.
MS. SEEGER: And recommendation eight, the removal of patient and bringing it to individual with respect to the notice of privacy practices.
MS. KLOSS: Okay. Keep going. There, we made that conforming change in the description.
MS. SEEGER: And then rather than educate, we changed that language to, create greater transparency for individuals.
MS. KLOSS: Recipients, we looked at that earlier. The final, under the narrative on 11, under the 2013 Omnibus Rule. That makes clear that accounting for disclosures guidance is outdated, but be more specific on that.
MS. BERNSTEIN: And also specifying that we meant the Omnibus Rule, when we said recent changes.
MS. KLOSS: Right. That is much better.
MS. BERNSTEIN: People will know where to look for that. Just tagged recommendation eight there.
MS. KLOSS: Keep scrolling. We have looked at that. We talked about that. That is it. Sincerely, William W. Stead.
DR. STEAD: Warm regards. I see lots of nodding heads. Do we have a motion to approve the letter?
MR. LANDEN: So moved.
DR. ROSS: Second.
DR. STEAD: Any discussion? All in favor?
(Chorus of Ayes)
DR. STEAD: Any opposed? Any abstained?
DR. STEAD: That is awesome work. Okay, we are going to start back on the report to congress. Let’s pause as we are shifting to that and ask if anybody has joined us on the phone that has not announced themselves.
DR. STEAD: I believe that we had completed the discussion of the Pop Health section and themes in section three and that, therefore, we are ready to go to work our way through data access and use and then the section on next steps. Is that correct? Do we have additional work to do? I know Bruce had said he thought we were done.
DR. PHILLIPS: Bill, I had offered some additional language in the Pop Health section. Was that good?
MS. KANAAN: What you are looking at up here and I have just put it in double space and I am going to see if I can add line numbers, is people’s inserts, which have just been dropped in, we haven’t edited it yet. Also with notes about new contributions that are still coming.
DR. STEAD: I think, just from a process point of view, what I would like to do is pick up where we were, get to the end, and then if we have time, we can come back and go through where you are in dropping stuff in. I don’t want to loop back until we have made one pass all the way through.
DR. MAYS: Can you give us page numbers.
MS. HINES: On page 14, yesterday, we, in one of our final discussions –
DR. MAYS: In the e-agenda book.
MS. HINES 35. We changed theme one and we have that. That is where my notes end in terms of – now, Bob, you did email us some minor insertions into – was it number three?
DR. PHILLIPS: In two and in four.
DR. STEAD: Two and four was what we had mentioned. For process points, did anybody have anything else they wanted to put on the table related to Pop Health theme two? Okay. Pop Health theme three?
DR. MAYS: The only thing I struggled with is the title. I know what we are trying to do, but this says a collaborative approach to health data collection and dissemination across sectors can maximize improvements. I am not sure I understand exactly a collaborative approach to health data collection. Are you really talking linkages more so than – we don’t collaborate on health data collection, unless you really want something very radical, which is that NHANES is collecting data and part of its data goes to Department of Justice or something. It is just the title.
DR. COHEN: Would integrated be better, an integrated approach? The idea is that there is data collection throughout the data streams, as opposed –
DR. ROSS: Can I argue why collaborative? I am seeing a piece of this in my own work closely with CDC that by collaborating within the agency, they have now been able to put to rest several data collection systems because they have helped one another recognize that if they did one particular effort very well, they would totally take care of three or four other groups’ needs as well and, therefore, they can eliminate several other vertical systems they were doing. You can collaborate. It is difficult, I admit. If we start talking about collaboration at the breadth we are talking about here, it is not easy, but it could have a lot of power.
DR. MAYS: I don’t think that is what this is talking about. That is why I am confused. When you go in and read it, they – it is like they are working on integrated schema for data on health and its determinants. Up here, it talks about a collaborative approach to health data collection and dissemination. One of them just doesn’t match. I don’t know which.
DR. COHEN: Vickie, if it is the word that bothers you, would something like integrated or comprehensive rather than collaborative?
DR. STEAD: How about coordinated?
DR. MAYS: Okay. Yes.
MS. HINES: Honestly, to me, number three is the lead in to number four. It is – number three is what led to number four in some manner. I don’t know whether that is helpful in thinking about – I don’t know.
DR. MAYS: I am good with coordination. Then the stuff that is in here, you get a better sense of it being multisectoral and short of using the word also linkages.
DR. STEAD: So, one thing we may need to make an explicit statement at the end of the last sentence in theme three that then says, four is that consensus set.
MS. HINES: To me, three is the lead in. It is like and here is number four.
DR. STEAD: So, one question would be would we simplify without losing anything if we made number four the theme, and if we took the paragraph under three, and made it a first paragraph in four?
MS. HINES: I think that is better.
DR. STEAD: Does that help? Bruce and Bob, are you comfortable with that?
DR. PHILLIPS: Yes.
DR. STEAD: You got it Susan?
MS. KANAAN: No.
DR. COHEN: Delete the title in number three and move the title of four to where three is.
DR. STEAD: And put the text under three as the first – you’ve got it. That simplifies it.
MS. KANAAN: That would take care of it. Thank you.
DR. RIPPEN: Just one quick thing. You may want to replace the word verify. Verify seems kind of like an interesting word to use there. It might just be confirmed or reinforce the need of.
DR. KANAAN: What would the word be? Confirmed?
DR. RIPPEN: It could be confirmed.
DR. STEAD: Let’s keep trucking. That means we have actually done Pop Health. Let’s move to Data Access and Use. Rashida.
DR. DORSEY: So, I have a general comment. I shared this with Bill yesterday. In reading the report, it seems as though the first like half of the report is really – I could see the connections to HIPAA very clearly. Since this is a report to congress on implementation of administrative simplification provisions of HIPAA, I could see that. In the background, the introduction, like the last two paragraphs, two and a half paragraphs talk about how the – how we also make that linkage to the pop and community health and the other portions.
I would suggest for the committee to consider if there is a way in the overview that you make that connection again for the reader as you are going into the pop health section and moving forward. It reads as two different reports. At the end, it brings it back. The connections are there. For the reader, that might be something for you to consider.
DR. STEAD: I think that is a really good suggestion. In essence, this is part of the context in which the HIPAA implementation is taking place. As we have converged, the – we have to think about the implementation of HIPAA within this broader lens to be successful with it. That actually is how we make the connection – also the connection to Beyond HIPAA, where we are talking about the multiple levers. We will need to figure out how to capture that in the overview. I think it is a really good point, Rashida. Thank you.
DR. COHEN: That would be in the overview of Pop Health and in the overview, I think, of Data Access. Just to add another sentence there in the overview to remind people how this fits into HIPAA. Great suggestion. Thanks.
DR. STEAD: Data Access and Use. We will, again, want to figure out how to make that tie in. Other comments on the overview? Okay. Then theme one. People good with theme one? Theme two.
MS. HINES: Given the discussion we had earlier, we might want to also refer to some longer-term forthcoming work or focus.
DR. MAYS: I was just going to ask the question, in terms of timing, when this is going to go in and how far along the healthdata.gov report will be and what we should or shouldn’t put in. Is this what we have already done?
MS. KLOSS: This is as of the end of 2016.
DR. MAYS: I forgot.
MS. KANAAN: I did wonder – another timing question, though, Vickie, related to that. We use the verb, are forthcoming in number two. In fact, I think they will be coming out at the same time because they are both presumably going to be approved in the April call. Do we want another verb here?
DR. STEAD: What I would like us to try – we have this challenge both with this and with the DID report and to a degree with the Pop Health Workshop report. All things that are relatively done, but will technically come out in 2017. We will, in fact – they will be the starting point of the 13th report, whenever we elect to do it. We have tried to have the bridge. I think we can, in fact – if there are things we think are important to say, we can say, to the degree we already know them. We have tried to walk that line pretty well. Those are the three places it applies.
MS. HINES: You are right. Forthcoming is not the right verb.
MS. KLOSS: We referenced in early 2017, with regard to DID.
MS. HINES: We could say in the first half of 2017 or whatever.
MS. KANAAN: I guess that is a small editorial change.
DR. STEAD: At the end, we say NCVHS expects – we can say on healthdata.gov in 2017.
DR. MAYS: Are you also asking like any of the recommendations that we think are high level that we want to mention here? Are we going to say something like the workgroup has brought issues up about HHS data, in general, that will be pursued? I am trying to walk the line to see what we can give you in here.
MS. HINES: I actually think it is short and sweet. The only question I have is, given the conversation we had this morning, do we want to add about in 2017 and maybe 2018 – I don’t know how far in advance people want to talk, really.
DR. STEAD: We just don’t know because we haven’t yet – we have made a note. Okay, there is a bucket of more work that we should put in our workplan someplace. We don’t yet know where it is going to land. That would be risky. I think we need to keep this around. The workgroup has identified opportunities for further enhancing healthdata.gov – successful implementation, early use. We expect to issue recommendations on the topic early in 2017.
MS. HINES: My only question is should we pull some of the detail from Erica’s introduction in here, just because this is really broad and generic, to the point that the input we got from Ankh, maybe we could just add a couple sentences.
MS. KLOSS: I think it is more important to tie this to the HIPAA report than to add more. I think not only do we need to do that kind of in the header, but we need to reinforce that a little bit in what we –
MS. HINES: We are not disagreeing.
DR. STEAD: Suggestions?
MR. COUSSOULE: I might actually take the second one and expound a little bit on the actual bullet there to figure out how the tie-in goes. Further recommendations to do what? That might help you tie it in a little bit as well.
DR. COHEN: The effectiveness and efficiency and use, both internally to HHS and for external users of healthdata.gov.
MR. COUSSOULE: Focus it a little bit and it actually helps make the tie-in.
MS. KLOSS: With proper governance and stewardship.
MS. KANAAN: Can someone tell me, using line numbers, where we are exactly?
MS. HINES: We are at number two, further recommendations on healthdata.gov are forthcoming. What Nick is really helping us do is see we need to say, to accomplish what to do what. Linda, you said something about – well, what I heard was to make it more effective and with proper governance.
MS. KLOSS: For use within HHS, as well as externally.
DR. MAYS: Bruce had it. Say it again.
DR. COHEN: Are we being recorded? To make more efficient and effective internal use at HHS and for external users. And then add something about stewardship and relate that to HIPAA.
MS. KANAAN: So, this isn’t for the title. This is within the report.
DR. COHEN: No, that will be a sentence in two in the text, not in the topic.
DR. STEAD: I think the stewardship piece will be another nice way to tie it back in to HIPAA. Do we think we have captured what we need to get at here? I know we haven’t captured language.
Let’s move to next steps. Does the overview paragraph in some way need additional tie-in to HIPAA or are we good in this part?
MS. GOSS: It struck me, when I was reviewing this, that we may want to say something to the effect of through our deliberations – during the time period that we are reporting on, we got a lot of input about what the industry wanted related to advancing HIPAA and garnering further efficiencies.
MS. KLOSS: Do we want to say anything specific about the current implementation timeline associated with HIPAA changes, the two years post notice of final rule? We grapple with a lengthy built-in timeline for revisions and a dynamic time. Yet, nobody is suggesting that it can necessarily be shortened, but it can be clarified. It can be laid out in a way that the industry is aware of –
DR. STEAD: That is what you are trying to do in the Predictability Roadmap, I believe, not in the overview.
MS. KLOSS: Okay. I understand.
DR. STEAD: Were your comments, Alix, related to the Predictability Roadmap?
MS. GOSS: No, it was in the overview, although I put my card up because I do have a comment on the roadmap part.
DR. STEAD: Let’s try to know where we – anyone who has cards that are up related to the overview?
MR. LANDEN: I like Linda’s thought. I suggest that we put a clause before the first sentence that says, based on lessons learned from several migrations or updates including 5010 ICD 10. Just an introductory clause as the first part of that first sentence.
DR. MAYS: I was going to suggest something similar, but I don’t have the wording. We need to say why we are doing each of these, not separately, but it is almost like we have to give authority as to why we are doing this and relate that authority back to our charge in some way. Otherwise, it just seems like we are just on a path.
MR. COUSSOULE: I think instead of saying significant areas of priority, I would say areas of opportunity and then specify what those opportunities would be and then it ties into the action.
DR. MAYS: Yes.
MR. COUSSOULE: That gets you into the outcome-based model.
DR. STEAD: You got that thought, Susan?
MS. KANAAN: Yes. I think so. Rich, did you have a specific place or I guess I can figure out where to put your clause about based on lessons learned.
MR. LANDEN: Specifically, I was saying beginning of line 529.
MS. KANAAN: Right at the very beginning?
MR. LANDEN: Very beginning. Do the insert that I talked about, put a comma, and then keep the rest that is there.
MS. GOSS: In regards to the predictability and burden on the healthcare industry, we will be having an extensive discussion about this roadmap in a little while, but I am wondering in the list in the second paragraph – so I am down like line 546ish or after – it struck me that WETI is also an advisor. I am thinking maybe we need to mention them because they are specifically cited as an advisor on HIPAA and they have a role in a variety of aspects. I wanted to include them because I think that they would be part of the organizations in the first sentence, starting 546 and ending on 548. I would insert them in that place if people agree that they should be noted.
DR. STEAD: That is fine. Got that, Susan? Okay. Other comments around the predictability roadmap?
MS. GOSS: I have one question not on the document. Of the members who are here, who participated in the June 2016 discussion on the roadmap? It will help me gauge the background level I need to get into. Not everybody. That is helpful. Thank you.
DR. STEAD: Linda, would you still advocate for an addition to the –
MS. KLOSS: I would. I am trying to figure out how to word it. I think it is not –
DR. STEAD: Is it best for you to maybe email some wording?
MS. KLOSS: It is. I would just make the point that we have gotten very used to using predictability roadmap. I don’t think somebody coming into it cold is going to quite get that concept. I think we may need to define it a little bit more.
MS. GOSS: We will talk about the topic in the 11 o’clock discussion. The document we are going to review may give us exactly what you are looking for.
DR. STEAD: Then let’s move to Beyond HIPAA.
DR. RIPPEN: I didn’t know if you wanted to just add in the listing of items that kind of fall into that, the whole question that we had talked about earlier, which was around individuals being able to see who used their deidentified information.
PARTICIPANT: Might be too specific.
MS. KANAAN: Could you say that again?
DR. RIPPEN: It is the key of tracking use of deidentified information for patients. It could just be a list of –
MS. GOSS: So, are you suggesting, Helga, we want to modify this?
DR. RIPPEN: No, not the title. I am just saying there is a list of things that fall under the Beyond HIPAA. I consider this Beyond HIPAA. The question is is it too specific as an example. Given that we probably want to follow up on it, that is the only reason.
MS. KLOSS: I think I probably would raise it up a level and refer there to ensuring appropriate transparency rather than get into the mechanism.
DR. STEAD: Will you do that then?
MS. KLOSS: I will.
DR. STEAD: The next generation –
DR. MAYS: Wait, I just wanted to ask – it says, to name a few, these new drivers include big data, analytics, blah, blah, blah. Shouldn’t we also include technology there? Technology, big data?
MS. KLOSS: I always think about analytics as technology.
DR. MAYS: Oh, it is like they are developing all kinds of wireless and there is transmission of data. Technology is developing lots of things that we are worried about, in terms of HIPAA. Is it a friendly amendment to add the word technology in front of big data?
MS. HINES: To that end, Ankh sent us comments, which we haven’t incorporated, about APIs and mobile devices and IT.
DR. MAYS: Oh, so they have it.
MS. HINES: They sent us that comment, but we haven’t incorporated it. I am just echoing through Ankh that someone else saw that same –
DR. STEAD: We have talked about that before.
MR. LANDEN: I think API is our significant –
DR. MAYS: I agree. I think we need to hear it.
MR. LANDEN: I think the opening sentence does a good job of describing what the providers and the health plans look like, in terms of paper pushing at the time of HIPAA, but we don’t have anything in there about what the technology was. I suggest we add, toward the end of – that would be line 555, right where your cursor is, two sentences. The first would be pre-HIPAA electronic data processing was largely mainframe-based, storage was expensive, and telecommunication was tape to tape, period. Internet, interoperability, and standards were all non-existent, period.
DR. STEAD: That is good. That will tie into the technology. Are we ready for next generation vitals?
DR. ROSS: Helga and I had a vigorous sidebar yesterday about the second sentence. I think we would all agree that the statement the vital statistics system is a national asset is right. How we word key component of identity protection, not certain. On the last part of that – and informs aspects of national security. Something like that. It isn’t per se. It was ever established per se to be a national security information system. We run the risk of alluding to something that it isn’t.
MS. GOSS: So, part of the aspect that needs to be incorporated is that it has taken on new roles and wants to be leveraged in new ways.
DR. ROSS: The information derived from vital statistics is broadened. That gets us into some slippery territory. Probably Helga needs to say a few words to.
DR. RIPPEN: Because I can’t help myself. So, I think even the overall kind of title and also some of the sentences that follow it, next generation vital statistics. In theory, I thought we were going to be having hearings to actually assess the current state of vital statistics, the challenges, if there are issues. I think we can probably predict that it might lead to a next generation vital statistics. I think that we haven’t had the hearing yet. Again, we can call it vital statistics, but next generation implies that we already think that –
DR. ROSS: Doesn’t NCVHS actually have an initiative funded underway, Next Gen?
MS. HINES: I don’t see an issue with that. I really don’t. It is kind of like saying it is a problem to say the sky is blue. We all know the sky is blue.
DR. RIPPEN: That is fine. I am just trying to be agnostic here.
MS. HINES: That is much appreciated, but I don’t see that as an issue in this particular case.
DR. RIPPEN: Then the other is going back to the component of identity protection and our national security. Again, we recommend our – I mean, our charter is for HHS. If we want to have hearings that include the other various departments and agencies that are not part of HHS and discuss that, that is one thing. That is different than saying – I guess I have just concerns about how it is positioned.
Perhaps, again, going back to the end, as far as and it is used by other agencies, such as social security and others to do whatever they have to do is fine, but I am just a little bit concerned about some of the wording. I can actually provide some recommended changes instead of having a continued dialogue about just the concerns of positioning and how it might be interpreted.
MS. HINES: You suggested some wording that I think we ought to look at, which is that it isn’t per se to be a national security information system, but the information from it –
DR. ROSS: Here is the challenge, it informed – these data inform a whole host of other agencies’ responsibilities and activities. Helga is just drawing the question about where is our authority?
DR. COHEN: Let me try to answer that. So, I think a missing word is establishment. It is identity establishment because that is what a birth certificate does. That is what a death certificate does. Those are within the purview of health systems. Recordation of birth and death initially was a state responsibility – actually, a parish and village responsibility and church responsibility. The evolution of this system to a cooperative health statistics system is a Health and Human Service activity. The establishment of identity using a vital statistics system is the key. Protection of that system is sort of an important aspect –
PARTICIPANT: Protection of the system, but not the identity.
DR. RIPPEN: Again, it goes back to I still have to use my birth certificate when I get an ID. It is not electronic. Again, going back to its focus as it relates to HHS and health, I guess I am just –
DR. STEAD: Let’s parse – let me try to – so, one, NCVHS’ charge is to advice the Secretary. It also includes convening other stakeholders. It includes providing advice to others. Our primary focus – we prioritize our work around advising the Secretary, but our charge, from a point of view of convening and advise, is actually much broader. I decouple the legalities of what we are supposed to do, the charter, from the positioning in this statement that best makes the point we are trying to make. I think you are right. We can position the statement better. I would figure out what that is. I don’t think our charter is the problem.
DR. RIPPEN: I think, again, it is up to this committee to decide the scope and the tone and the focus of the meetings. I think, though, if you expand the scope, you will have to make sure that the testimony that you obtain from different perspectives broadens. I am really having a conversation about scope and focus.
DR. STEAD: Let me give you one more maybe enabler. This is the report to congress. We will have the discussion in a bit around vitals. We are going to, in that discussion, need to be clear on the scope of the hearing and that we have got the right people there.
I think most of us believe this hearing to be the start of a journey and that there are likely to be more steps. There are possibly going to be more steps before the next report to congress. We don’t know. We will be informed by the hearing. This needs – this can be at a higher level than the hearing. It needs to be directionally correct with where we think we are going. I am just – it doesn’t have to exactly – it needs to be consistent, but not tied to the hearing, if that helps figure out how to word it. I am just trying to – I am good with I think where you are trying to go.
DR. MAYS: Exactly. I think the distinction between the hearing stuff and making sure that it is an open hearing, from what you are saying and what we are saying to congress.
One of the things I think that maybe would help this, since it is going to congress and congress doesn’t quite understand exactly how birth and death statistics are used, that maybe instead of just centering on this issue of identity protection and national security, can you just broaden it and add some other functions? That would then, I think, help – I get your sense of it seems like we are just doing it for security issues, but I think what might be helpful is to say why we need to overhaul this and that it has many uses. If we could give some of the other as well, I think it would be more balanced.
DR. RIPPEN: I think the other thing is you need to spell out what vital statistics is. I don’t know that people would know it is birth and death records.
DR. MAYS: It is just a little more here would help, in terms of explaining what vital statistics are, how they are helpful to the country, and then, you know, this is the rest.
DR. STEAD: I think that is pretty good consensus about how to handle this. I like – I think there is real synergy between what the two – really all three of you are saying. If we say it is a national asset, it is a key component of establishing identity and the information derived from it is used for many purposes, including national security, blah, blah, blah. I think that gets at that heart. If we get the – right.
DR. ROSS: Which rationalizes why we have really got to take a time out and think about how do we modernize this.
MS. HINES: And there was Bruce’s thing – one last thing, Susan, is informs a whole host of executive agency activities.
MS. KANAAN: Where did you suggest putting that? Is that in the same spot?
DR. STEAD: We can do that offline. Okay, team, we are about seven minutes behind. Could I advocate that we come back at five after by my watch, which would be in seven minutes. That would give us 12 minutes or 11 minutes. Let’s come back at five after by our telephone’s time. Maybe someday we will reset that clock as part of improving HHS’ data capabilities.
DR. STEAD: We need to get under way. I will introduce Alix and Lorraine to walk us through predictability roadmap.
MS. GOSS: Good morning, everyone. We are going to go ahead and be talking about the predictability roadmap. In advance of this meeting, you received a document of the roadmap and this draft discussion document. It is dated February 12th. It is reflective of conversations that started a couple of years ago. We really vetted as a committee in June of last year. We are very excited to have Lorraine joining us, especially considering her extensive background in understanding players and HIPAA and the framework and the challenges we have had over the years.
What we would like to do today is provide you with some context of the players in the HIPAA industry to help have a more fruitful conversation where we will present to you the idea of what a roadmap is about and more especially what it is not about. We want to have a very specific scope on this. We will talk about our anticipated timelines, so you can follow along in the Word document that was provided.
The Roadmap, just to kind of set some 50,000 foot perspective, is we need certainty. We need businesses to be able to schedule and anticipate what is coming down the pike, whether it is in standards development or operating rule development. Whether it is commenting on the proposed drafts, ready for commenting on the idea of adoption and promulgation, and then the implementation effort to garner efficiency and the goals of administrative simplification.
We would be better off if we had a better, more predictable timeline and approach to all of this. That is ultimately the goal. But before we get into talking about how we are going to accomplish more predictability, Lorraine is going to give us the alphabet soup and some context to help us in the discussion.
MS. DOO: Good morning, everybody. It is always good to have a common vocabulary. You all have been in this world for probably as long as I have. We were kind of trading war stories yesterday about our history in this. None of us will admit to how long actually any of us have been doing it. But we all have familiar faces for it.
I am going to try and tame the tower of Babel. It will be familiar, but that is okay because then we will make sure that we have the same terminology. Actually, I realize some of the names that I use might be some of my own. But my kids always said that I made up words anyway. That is okay. We will have the same ones.
This is from a presentation actually that the first word that you will get familiar with, and Denise Buenning actually knows this, but what I call the mother DSMO, because it is all capital letters. That is from the Designated Standards Maintenance Organization, which was created in regulation and is the founding group of standards development organizations that are responsible for coming to NCVHS every year to give a report on the change request that they receive. So what I call the mother DSMO because it is capital letters. Moms are always about capital letters because when they yell at you, it always sounds like capital letters.
They are responsible for receiving change requests. So this is the process of how changes come in and then ultimately come to NCVHS for recommendations. What you will see at the bottom, though, is for operating rules, which is this box at the bottom of the slide, is separate and apart from the DSMO process. I am just going to start at the far left or right for you guys.
But if the industry finds that they need a change to a standard or a code set, not an identifier, but presumably it was supposed to work, that they can either make that request of the individual standards development organizations or they can make it to the mother DSMO. That request would go into that entity. There is actually a DSMO website. It is HIPAA/DSMO.org. You can go to that website and submit your change request. They get assessed by the group of SDOs. You can go through the SDO process. You can go through a code set process, but separate and apart from the mother. You can also go through an operating rule process.
The DSMO would review it and make recommendations. In the original MOU, which was again done by regulation, and HHS is party to that memorandum of understanding, once upon a time, there was actually going to be a cost benefit review of whether those change requests were going to be good for industry or not. That actually never got implemented, this was at the very beginning back in 2000, in part because there wasn’t funding provided for it.
The one big cost benefit analysis that was conducted, and HHS did pay for that, was when we implemented 5010. We have paid Gartner to do an analysis. It was an official cost benefit analysis that really did look at what the improvements could be when we implemented that. So there is an official study that was done for all of the transactions. That is posted on our website. You actually can look at that.
But the DSMO will come forward to NCVHS as they do each year and they will make recommendations. Then HHS uses that data, as we did with Gartner, and uses that to make recommendations. With the operating rule entity, they don’t go through the DSMO. They are not part of the memorandum of understanding. But of course, they do come through NCVHS, and that is CAQH CORE. Then WEDI, as Alix has said, is also advisory. They will come through, also.
MS. GOSS: Great introduction. I just want to clarify that the DSMO is made up of, you said SDO. Right, but what I wanted to do is we tend to have slang when we talk about SDOs. Some of us think very specifically like NCPDP or HL7. But data content committees are also sort of a larger bucket of SDOs. If you think about the National Uniform Billing or Claims Committee, so it is kind of the double dip.
MS. DOO: Actually, the next two pages, just to give all the abbreviations, so the standards development organizations are X12, HL7, NCPDP, NACHA, which is responsible for the electronic funds transfer standard is not part of the DSMO, either. Even though we adopted their standard for electronic funds transfer, they didn’t become part of the DSMO. That is interesting. They never got incorporated.
The Dental Content Committee because of the dental codes, and then as Alix was just saying, the National Uniform Billing Committee for the institutional claims and then the National Uniform Claim Committee for professional claims. Those are the code set maintainers because we adopted those codes.
Then you will hear about this, and it does have a lot to do with the predictability roadmap and what is a hurdle for HHS is something called the Administrative Procedures Act. Those of you in government know that fairly well. That has requirements for us, for HHS, because when we publish regulations, it means it is due diligence that we must give the industry the opportunity to comment on anything that we adopt. We have to give ample time to industry to comment on any proposal. That includes the proposal to adopt standards and code sets and identifiers.
That means either interim rule or proposed rule. That is part of what takes the time. That is what we have to look at is what kinds of opportunities we have under the Administrative Procedures Act that will afford us a way to expedite any rule-making processes. There are some that other agencies have used to do that more expeditiously.
Those are the words that you are going to hear. I guess I will go back. Just so everyone has the same language, and it will make this next part easier.
MS. GOSS: Before we switch over to the other Word document, is everybody comfortable with sort of industry has requests? There is a set of processes that have been adopted to help manage that. It is executed before it all gets to us who the players are. Any questions on that? Okay.
Hopefully, you all have this document available. Just grab the document purpose. I think that is really where we want to talk about it. In June of last year, we had some conversation around what was it that we wanted to do in this big concept of a roadmap. We decided that through some lengthy discussion that we didn’t want the kitchen sink in it. We wanted to be very narrow to tackle what we thought was feasible and what would have a great amount of impact, and that would maybe also lend itself towards subsequent consideration efforts on roadmaps that we may want to pursue as NCVHS.
So we narrowed our thinking down to really standards in the sense of X12, NCPDP and operating rule standards that are adopted currently under HIPAA. There had been a tremendous amount of work on this topic. When I sat literally a long time ago on this side of the table in 2003 to 2006, my role in X12, along with NCPDP and HL7, and subsequently we also engaged in the DSMO partners, in trying to figure out what did we learn from implementing 4010, the first HIPAA version of X12, and also the NCPDP standards? And how could we make this all better for ourselves? That work has been brought forward again in 2009, and there has been some other work that has been done by HHS to try to figure out where we might be able to go in improving some of the processes around Administrative Procedures Act to get more predictability.
We need to make sure that we are current in understanding what is happening today in both the SDOs and the operating rule authoring entity, CAQH Core. We need to understand the current state of the regulations. We need to understand what has been done before in trying to advance the improvements. We need to figure out where we play and the NCVHS role in all of this. Then we need to create those recommendations and then pursue them. The document is designed to help us get at those objectives.
Is everybody comfortable with why we need a roadmap? Any questions on that? Yes, Linda?
MS. KLOSS: I think this is outstanding. I like the way you have really defined it clearly. I was going to suggest one small change in the summary. You were right earlier saying that we have the words here that we need for other document.
Can we consider adding a degree of certainty in the timing and sequencing of the adoption? I think the sequencing is as important as timing.
MS. GOSS: So when you say sequencing, you are talking about do we want to do the big bang theory and adopt all the things at once? Maybe you want to think like an 837.
MS. KLOSS: What order must that adoption occur in? We need to adopt 5010 before ICD 10. That was a classic example of where we were kind of backwards.
MS. GOSS: So the interplay of standards and code sets, or transactions and code sets, and how those work together.
MS. KLOSS: Timing and sequence.
MR. COUSSOULE: It is also important from a sequencing perspective because many of these are on different paths, parallel paths. Changing one without changing the other could create a sequencing, so that is what you are talking about.
MS. KLOSS: A lot of the problems we have had have been sequencing problems, rather than when it is adopted, but in what sequence.
DR. RIPPEN: Even from IT infrastructure perspective, some things just make more sense to do first, even though from a regulatory thing, you wouldn’t notice that. It is more from a mitigating cost and risk.
MS. GOSS: It is a good thing we have the MBI code value within a transaction or else the SSNRI would be in a world of hurt.
DR. PHILLIPS: I am really ignorant here. This is like a 101 class for me. This is really helpful. How does certification of EHRs and Meaningful Use fit into this? It does not? Why should it?
MS. GOSS: But it should, and that is part of the issue. In the document, if you scroll down in it, you will see that ONC is included in there. For that very specific reason because when you think about some of the conversations we have had, Bob, around convergence of administrative financial and clinical standards, this is a great place to start crafting those, especially when you think about the flexibility that ONC has had with their advisories, standards advisories and being able to move their regulations and the standards and certification much faster. That is a lessons learned that we definitely want to look at.
I think that we don’t want to be short-sighted about just how to fix what is broken in HIPAA adoption. We want to think about where that convergence needs to go for the whole industry when you have got the data, when you have got different standards and how they fit, I mean general standards in that sense, how they fit together and how we are really responsive to the industry needs. I think that this is a portion of the larger conversation. But beyond HIPAA, and the vocabulary conversation, and even that CIO roundtable conversation that is on our strategic roadmap work plan list, all fit into that larger conversation that you just opened the door on.
MS. DOO: If I may also, with respect to, I think, in terms of meaningful use and the quality payment models and alternative payment models and the quality data, that is coming from that information rather than using claims data, if it is coming from the electronic health records, then you are going to have a much better tool for any of those payment models. If we are getting away from fee for service payments, then again the data should be coming from there, not from claims data. There is that convergence. That marriage should be happening and not from an EDI transaction.
MS. GOSS: Any other questions before we continue? So, going down the document purpose, which sort of already got into, so now I am going to scroll down to the end goal. Ultimately, what we want to do is develop a roadmap instead of recommendations, philosophy or an approach for more efficient development, adoption and implementation of standards and operating rules. That means that we really need to take a look at the existing processes and figure out where their synergies for maintenance and changes to both standards, transactions and operating rules.
We want to look at the operating rules processes and naming convention to determine whether there are further opportunities there. We want to request that HHS evaluate best case opportunities for adoption efficiencies. Evaluate other mandates such as certification of EHRs, et cetera, meaningful use and the convergence aspect. So great question because you pulled all of that together for us so far.
I am not seeing any questions. I am going to skip the detailed fodder in the middle.
DR. STEAD: Alix, is this roadmap limited to the administrative standards? We have talked about the fact that we need to figure out how to align administrative and clinical. My sense is this is around the administrative. I think we should call that out, so it is very clear. We are going to try to get a predictability roadmap around administrative. Then with that locked in, we are going to come back and work with ONC and others to figure out how to get alignment across.
MS. GOSS: I feel like there are two parts to that. First of all, it is administrative and financial standards and operating rules. But we can’t do a deep dive on that effectively if we don’t bring in those clinical partners who have had more flexibility in their lessons learned. But you are right. It is not about trying to get to convergence.
DR. STEAD: Let’s try to clarify. We need those other partners at the table as we figure out how to get the predictability roadmap around administrative and financial. But we are explicitly listing as out of scope alignment of clinical administrative.
MS. GOSS: And we have a section on out of scope, so alignment.
MS. KLOSS: And by administrative, we specifically mean transactions, code sets.
MS. GOSS: Code sets is a kind of an interesting dynamic. The code sets, you can talk about external code sets, the ones that are adopted in and of themselves by the federal government and regulation. Or you could be talking about code sets that are internal or used within a transaction.
MS. KLOSS: You said HIPAA. Is this the administrative only? Or do we leave that?
MS. GOSS: I think that we decided to not pull in, yes. I agree with you, Linda. It was determined last June through conversations that we had to be careful about what we could actually accomplish. So those things that are adopted as code sets under HIPAA promulgated regulations I would consider out of scope for this initial round. But our lessons learned in the process, they benefit that.
DR. STEAD: We will add them to the explicit out-of-scope lesson.
MR. LANDEN: I am just concerned that it may not be possible to use terms like exclusive and something may be out of scope, or it may not be out of scope. It may be out of our jurisdiction, out of our control, but if say with Meaningful Use, there is a CMS component that is regulatory, an ONC component. Yet, as we converge, we mention the ISA, the Interoperability Standards Advisory from ONC.
There are code sets that are HIPAA code sets, which isn’t our purview. But there are other things that aren’t HIPAA code sets that could be done that could impact what has to be done by industry. I am thinking too bright a line around what is exclusive, what is out of scope, may be hard to peg down. Some of those answers we won’t really understand or know until we get far in the process. I am fine with drawing scope around it, but as long as it is pencil.
DR. STEAD: Well, let me be clear. I tend to always want to work down from a whole system perspective. What I think we need to do here is to tightly demarcate this first round of work to generate a first ever predictability roadmap that, in fact, crosses certain dependencies. We only will get that done if we really narrow the scopes. I am not concerned about our jurisdiction. Again, I think our jurisdiction actually can be pretty broad.
I think if we can make this first effort and have a hearing, and then based on that hearing, get a draft roadmap, get feedback, and have one piece, one example that is as sufficiently narrow in scope that it is real. Then we can come back and say, okay, how do we scale this thing? How do we take it up a level? How do we bring in the other dependencies, which are probably going to work out with a set of detailed predictability roadmaps that are explicitly interconnected up one level of abstraction? That is sort of where I have become comfortable. Four years ago, I wanted to start at the top. But I have wanted to do that four years, and we have gotten nowhere.
MS. GOSS: The problem is we are going to drown if we try to drink the ocean. We really want to be able to keep ourselves highly focused. I really agree with Rich that they are going to be some unanticipated curves and twists in the journey as we explore this because there appears to have been some progress in a variety of the players, processes, that we need to better understand and put together to advance it. So I think a general out of scope in soft pencil, and then we can kind of fine-tune that as we further down the journey.
MS. KLOSS: I think that is a wise scope. We probably should just call it predictability roadmap for administrative standards and operating rules.
MS. GOSS: Start it clear at the top. I like that. So what I would like to do is could we go down to E for milestones. I want to talk about the big picture process as opposed to going into the various pieces of the roadmap document, activities and players and all that stuff.
All right. So what we wanted to do was to start at this meeting with a fresh updated plan for how we want to approach the predictability roadmap for administrative simplification transactions and code sets, which I am now just going to call roadmap, so that we can make sure everybody is on the same page and understand how we were proceeding. Our goal is to have a lot of activity on this in the next couple of months with the caveat of we are going to proceed towards an HPID conversation/hearing hopefully in the April timeframe, presuming that we can prepare ourselves for a successful event and management of the process.
So we would definitely update you at the next meeting, whether that is April or June, on our progress. What we want to do is develop — thrown off by number two, develop timeline for public posting with comment, okay. So I don’t know that we have to do number two. That was your question. I think we have to explore that separately, so let me skip that step because that is internal process.
We need to develop background material. What we know is that there were documents that were created previously on the process modification processes that were recommended by the standards community. What they have done in the document is not only put a narrative together to say these are the problems, but they have actually created a set of recommendations and a set of detailed process improvements, like timing improvements, that we can start with. So we need to validate that information that was updated 2009 is still accurate or adjusted.
We also need to ensure that our NACHA and operating authorizing entity partners who are not a part of the DSMO, but are a part of the ecosystem, we have an understanding of their processes. We also need to understand what activities have happened inside the federal government on exploring opportunities for enhanced efficiencies because they have been thinking about this since probably 2003.
So once we get all those materials pulled together and agendas, we would pause, so Bill could ask a question.
DR. STEAD: I am struggling to get up one more level to what the work plan is for the roadmap. I am still carrying around in my dimming memory an August hearing and then something else that eventually results in something. If we could maybe get that big, and then this sort of becomes how do we get ready for that.
MS. GOSS: All right. We need to get prepped. We need to have a conference call with the players. We need to have an August workshop to vet those at a detailed level. We then need to figure out where we are at, come back to NCVHS, decide what we products we are going to do and then act on those, which could be either formal recommendations to HHS or it could also include to the industry and our partners because this is not a process that is focused just on what the feds can change. This is a look at holistically what we can do.
DR. STEAD: From my overarching perch, I will carry in my head that we are calendar hopefully soon, an August workshop. We will work out these milestones, so that we are on the right glide path for the August workshop.
MS. GOSS: Yes, and part of our ability to do that is to get the feedback to make sure that everybody is on the same page.
DR. STEAD: Absolutely. Then I am assuming since the September meeting will follow pretty quickly on the workshop, at that point, you will have had some experience. You might, at the September meeting, have a block with NCVHS, in which we discuss how you want to tackle the work going forward from there.
MS. GOSS: That is very reasonable. I cannot commit to it, but it is reasonable.
MR. COUSSOULE: Structurally that is correct. We just have to lay out all the sequencing.
MS. GOSS: Let’s be mindful though also, that we actually now have this new HPID. We want to make sure that we can deliver on that and not derail any of the workshop activities to August.
DR. STEAD: Totally understand. I do better if I know.
MS. HINES: Alix, when we talked about this a couple of months ago, and I want to make sure either something has changed or I am not getting something, that after the workshop, there would be some proposals. Then there would be a hearing on those proposals.
MS. GOSS: You are right. But when we actually started specking this out, we anticipated that we would have had this workshop already, and that we would have been factoring in the recommendations or the suggestions of where we should go from that workshop, so that we could then solicit a larger set of industry feedback under what was going to be potentially the review committee hearing. So the original thought was, let’s take the deep dive and then elevate it. Get some feedback, and then enable us, as NCVHS, to make some hardcore recommendations with a broader set of input.
MR. COUSSOULE: Your point, the first workshop may not be enough.
DR. STEAD: The mental model is we have a workshop that helps the industry and us figure out where we are headed.
MS. GOSS: I want to clarify. It is a small segment of the industry. I think what we need to do is not limit. I think we need more input as NCVHS from the larger industry.
DR. STEAD: Absolutely. So there will be at least one hearing, but that wouldn’t be before either Novemberish or first quarter of 2018.
MS. GOSS: I think that is all very fair. My crystal ball is a little foggy on what the timing could really be. But yes, I think your point is spot on that we need to have a workshop and then some sort of a hearing or some solicitation of industry input.
Is everybody with us so far? Anybody lost? Okay. A lot of what I think I was going to go over I think we sort of got at in that dialogue, thank you, Bill. So I think the point is that we are expecting to include a number of individuals from across the standards community. We have tried to identify some of the usual suspects in our thinking. But it is really the designated standards maintenance organization members, NACHA, operating rules, ONC, that is the players. You can kind of link back to the level setting orientation that we got from Lorraine.
We are thinking it will be sort of an off-cycle in-person meeting. It would be ideal for us to likely collaborate on the venue with our federal partners. So NCVHS could host a meeting that would enable participation by the necessary federal partners who may want to listen or be involved in it. There will be the forever challenge of attendance and effective management to get to some kind of effective, real output. We are going to have to worry about some size and how we handle this.
Having been involved in some of these discussions, we definitely will want to have like one official voice per organization. We want to make sure each organization has the ability to provide input on the background materials, make sure they have got the right building blocks. Then also have a preliminary discussion with the attendees to make sure everybody has the ability to influence the agenda and kind of how we approach this, and set us up to have a much more productive in-person meeting.
Hopefully it only needs to be one day. Depending on what we need to tackle, it likely could be more. Just based upon some of my experience previously, I spent way too many days at the Hilton and Chicago Airport negotiating stuff like this. Then have the in-person meeting produce some sort of results from that, bring it back here for some consideration in determining what really makes sense for the next steps.
MS. HINES: Alix, just to help me out so I am tracking with you. So the conference call under the milestone V, is really to do that agenda setting and to make sure that is what you were just referring to.
MS. GOSS: Yes. I really think that we have got to bring everybody kind of on the ground floor. They have been listening. They have been hearing a little bit about this along the way. We really want to make sure that we do the same kind of one-on-one level setting. What are we trying to achieve? What are we not trying to achieve? How are we going to manage ourselves? What are we going to talk about?
Make sure that also there is a cross-pollination on processes and understanding. If you haven’t dealt with the APA and you haven’t been schooled on that, it may not make sense. Now, I anticipate that a lot of these players have that background, but we don’t want to make that assumption.
MS. HINES: So that is what the conference call is going to help accomplish? Thank you.
MS. KLOSS: The obvious issue is the month of August to do something like this and the urgent need to get a date out there, so people can plan their vacations in the summer around it.
MS. GOSS: I think we were talking about September, but with the NextGen, nobody wanted five days of meetings that week. I think you are making a really good point. We are going to be challenged with finding a date. We need to get going on this. So while we are also getting HPID going, which I think we have a good launch point for questions and stuff like that, which we included in the overview document already for folks that we presented yesterday, I think that we can get that going very quickly in the next two weeks. Hopefully get executive committee feedback on those questions and the scope, going through the management process, enabling us to get that out by the middle of March at the latest.
We are going to have to pick dates. There are a lot of logistics for us to pull this off. I think we said April, but I don’t know honestly if that is feasible. We are probably going to have to be thinking more like May. But we have to step back and have some quiet time to work through those.
We have some logistics, but we don’t need to dive into those details right now. We can handle those offline.
MS. HINES: Again for the workshop, for that hearing, the April, May, whatever hearing, HPID, we weren’t necessarily aiming for CMS as a location. Whereas the location, we were thinking it would be good to have that in Baltimore. Okay. Just helping us identify space.
MS. GOSS: And those, I had some preliminary conversations last summer, but those players are gone. So Lorraine and I will work through those as we get our arms around all of this.
MS. HINES: But the April/May would be maybe in this building, for example, whereas the August could be in Baltimore?
MS. GOSS: Sure. I think that some of our planning in November has really changed. I am also open to if we need to think about the larger picture of when we meet and when we do things as a full committee and how we co-locate from a budgetary efficiency perspective. Any other questions? Any concerns? Any feedback on the rest of the document that we provided because I didn’t want to go through the details the first couple of pages? Okay.
I thought we would talk a little bit about HPID after we got done with that, but we wove it all together really well. With no other questions, I turn it back to you ahead of time.
DR. STEAD: Do we want to roll in the NextGen vitals and see if we can then –
MS. HINES: That is a public agenda. You may lose people.
DR. STEAD: We can pull forward working on the April. Let’s pull the 1:45-2:00 block forward and get it out of our way, and then break for lunch. We wanted to take time to talk to the group about how to handle the April Full Committee meeting. In my tenure on the committee, we have had one telephonic full committee meeting. It was simply a phone call, a conference call, in which we approved some things we needed to approve.
At this juncture, what we are hoping to approve at that meeting will be the Report to Congress, and it will be the HealthData.gov, that I am now actually beginning to think it will be a letter. We will need to sort of book our head through that, and a Pop Health letter. So in essence, there would be three items that we would be dealing with that, in theory, we have already had the detailed discussion of.
But as we think about how to handle the meeting, remember that now we are going to have a Full Committee meeting in November, a non face-to-face. I am just trying to connect the planning challenge.
MS. HINES: Let’s just remind everyone that we have been advised not to have four in-person meetings a year due to budget issues.
DR. STEAD: Which was part of our discussion in November, I thought. This is not new news.
MS. HINES: It is. It has got to be absorbed into what does that actually look like on the calendar.
DR. STEAD: I am just trying to have us begin to think how we are going to work when we have virtual meetings. We can take a minimalist approach to April. We have taken a minimalist approach to the length of the April meeting, two hours. We have already locked it in.
But I am assuming that when we have a virtual meeting in April, that it will be a fully technology-supported virtual meeting of some. I understand all of the challenges. We are in the process of doing the contract. We don’t know who.
April can, if we want to be, be a baby step toward November. By November, we are going to have to have worked out how we are going to get the various reports and updates we want virtually. We are going to have to work out how we have the technology that allows us to be as effective personally as we were in the last two days. I think we have actually now got a pretty good visual of what an effective committee meeting looks like. Our job is to figure out how to reproduce that virtually. That is the November task.
The only reason I sort of want people to begin to think that way is I want to know if there is any aspect of that that would be helpful as we are trying to land these three airplanes in April. Maybe we are good with simply a two-hour call and praying that we know where we are. Or maybe we need something that is a little better in terms of collaborative technology. I don’t know. I just wanted to put the question on the table.
MS. HINES: Collaborative technology meaning there is a WebEx, but that is mostly for the public to view things and to be able to hear us, which they will have.
DR. STEAD: Right. That means if we need to make a change in something we are approving, somebody can be doing that and we can see that.
MS. HINES: That is what I don’t know. Unfortunately, we are not allowed to use zoom, which as you know is fabulous.
DR. STEAD: The reason we put this topic on the table was to have the conversation, how we wanted to handle this meeting. What order did the tents come up? We have got four of them.
DR. RIPPEN: So going back to how to handle online meetings, one approach could be that when you have the in-person meetings, you do the bulk of the deeper dives. The shorter meetings or the phone meetings where you approve and you have a lot of work ahead of time for people to provide comments.
If you want to have a longer meeting online, where you are actually doing work work, where you are actually making changes, you will probably need some sort of technology, be it Google Docs, where people who have authority can make edits real-time or flags or something like that. Otherwise, you can’t really have a robust conversation.
DR. MAYS: In the same ballpark, I think we want to think about the type of work we want to do and the type of technology that we need and what makes the meeting run well. I am going to suggest a few things because we do this with our community partners all the time. We have had full day meetings with them.
MS. HINES: Without federal rules.
PARTICIPANT: Can you tell us the federal constraints, just so we know.
MS. HINES: There are many and a lot of them have to do with security. I am not an expert on technology procurement. I know that every time I have asked for anything, the answer is no, seriously. I wanted to use software that allows us to share our work plan. No, it doesn’t meet federal security. I am just being honest with you. I don’t find much flexibility. My boss took it all the way up the chain for me. We tried.
I don’t know for April because we are in a certain set of resources. After September, we will be in a different set of resources. I can certainly, as we develop the next round of resources, try to put in some language and see what shows up. But I wouldn’t count on too much right now.
DR. STEAD: I think rumor on the street, at least, is that ONC does this pretty well. ONC is federal, a piece of HHS.
MS. HINES: They have a different procurement office than we do. I love that suggestion. What I would like to do is talk with the right person there, find out what they have done, and then come back to our people, our admin people, and see if we can replicate it.
DR. STEAD: I think April, if we have a WebEx and a telephone system, we have done enough work on these things. We are probably okay. Even though we are doing everything we can to get more efficient, I think it is unbelievable what we have gotten done since November.
I think we have done a pretty good job of limiting what we are doing. Even so, if we are not going to have a meeting between September and February or whenever, we have got to have a real honest to God meeting. We have got to do that virtually. I think this is where we prove we can do it. Maybe ONC can help us.
MS. HINES: I like that. I really like that. The other thing is we could change the schedule, so that we have the meeting in person in November, but we don’t have the February one. It deserves no stipulation as to when we have the in-person meeting. I just want to put out there that if you really feel like we are on a role in November, and that is when we need to meet in person, maybe we have the February meeting.
DR. STEAD: Do we have the two November dates still held on everybody’s calendar? We need to make sure everybody keeps the dates because presumably we would use them either virtually or in person. We need to not get the calendar where it is a problem for us, while we are working out what we can do and what we need to do.
Let’s go back to Vickie and then Denise.
DR. MAYS: Part of what I was going to suggest was the issue of determining whether or not you can meeting in a building that is different. We do have the UCDC building here. There is a different set of technologies. My understanding, and this can be checked, is if you are in a state or federal building, and you are meeting there. You can employ the technology that is already in the building. That might be a possibility.
DR. STEAD: I hear that as what do we want to do if we want to better enable the technology in one of the meetings where we are, in fact, face-to-face. That is actually a different question. I think if we can solve the problem of how to meet when we are not face-to-face, that will probably increase our capability of linking in, of having better public participation, et cetera, when we are face-to-face.
Every bone in my body is we should take this constraint to improve not just our efficiency, but our effectiveness, and how we are able to engage people. I am viewing this as a good challenge, not a bad challenge. I just think we have got to figure it out. I understand that is not simple.
DR. MAYS: The other thing is before the meeting, it is going to help if what you have is a set of instructions of how you want people to participate. So we have a list that we give people. We tell them about webcams they can use if they don’t have them. We tell them we want them on headphones. We tell them how to set up their computers. We give them the whole list before the meeting.
The other thing we do is offer them the capacity, using the day before, and they need to be where they are going to do it. We have a whole set of things that then just makes it run flawless. We are not having it fall apart during the meeting. I would suggest really planning those things. Also making sure that you use the software. People are really into emoji’s, but there is software that will allow you to have people make little faces and happy claps.
It sounds silly, but it really allows you to get the sense of the room during the time that it is happening, so that you know to stop or that you know to do certain things. But it needs to be everybody can see it. It can’t be you just have one person that can see it. I would just make some of those as suggestions.
DR. DORSEY: I will reach out to some other advisory committees. Rebecca will reach out to ONC. But there are other advisory committees who have used web platforms to have interactive meetings. They have really gone more towards having virtual meetings than having in-person meetings. I will look into what some of those options are and what they have learned, so that we can have some of that as we consider approaches for our virtual meetings that might be useful for us, that are currently being used by federal advisory committees.
PARTICIPANT: However we choose to do this, we want to dry run it before the meeting maybe in one of the exec com meetings or something like that.
MS. HINES: That is a great idea.
DR. ROSS: You are just wanting us to sort of enumerate all of these functions, so like online polling. In effect, we have to design the discussions and understand which points are polling required, which times sense of the meeting is required, which times small huddle groups. I would also ask that the public openness part of this, does that mean any discussions that go on, like the sideline?
The thing about what happens in person, we will stand and talk. Part of what helps us move along is some sideline discussions. Nobody says, well, is that on record or off record because we are all just standing around. But that is really what makes this stuff effective is that we think through that. That means we have to kind of design the whole flow of the two days and then block it into the different functions.
DR. STEAD: This has been a very helpful discussion. Denise?
MS. LOVE: I really just want to say that we have been doing Standards on these calls. Four-hour calls on WebEx or Citrix with breaks and what not. It really is effective for a small working group when you are working on a document. I am not sure on a wider view on technical, how effective it is. But for working, it is actually very —
MR. LANDEN: Very quickly with apologies, I don’t think we have to overthink this. The technology that ONC uses is proven and tested. We can get into all sorts of bells and whistles, but we are not so big of a group that we can’t get by with just the basics.
DR. STEAD: What I am taking away from this is we actually want to use some non-trivial part of our bandwidth in the June meeting and in the September meeting, making sure that we are ready to make. If we are still on target for November being our first full-scale virtual meeting, make sure that we are really thinking through how to do it. Not just from a technology point, but from today’s point, from the process about how we actually designed the meeting. I think that could be very good.
For April, what I am hearing is we are okay with basically WebEx-style support, so that we can all see a common screen. I think we are talking about being able to see a common screen and to be able to talk either on our phone, it depends on where people are and what their set-ups are, either on our phone line or not. That will get us through April.
DR. COHEN: Will someone be able to interact with that screen to put edits up that we will be able to see? So the edits would have to be done?
MS. HINES: We have had problems, but Janine is saying we can do it. Okay.
DR. COHEN: That would be very helpful.
DR. STEAD: We have a plan for the April meeting.
MS. HINES: We do. We are going to use WebEx Live to make edits. People can either access the meeting through their WebEx or on their phone. We will have both options. It will be open to the public. The main thing will be to approve the three items we just outlined, the report to Congress, the pop health letter and the HealthData.gov letter.
DR. STEAD: I think I have been sort of making a little list. Vicki and Eric, you will want to work out the timeline of how we are going to bring the HealthData.gov piece. I think that report to Congress, we have already sort of talked about how we were going to get that. We had a draft and comments back on that draft in a new draft before the end of March exec committee call.
Alix and Nick will co-conspire to get the health plan ID piece. I presume you will figure out whether you need us to deal with that exec committee by mail in advance of that call for timing, or whether we can do it on the call. We will need to have the pop health letter out in advance.
MS. HINES: I am assuming we would have an e-agenda book just the same as we have always had. We will do as much pre —
DR. STEAD: I think we asked people to do it this time, and we got a fair amount of it done. I think this call is only going to work if the drafts we are discussing on the full committee call, if any comments people, basically all comments have submitted electronically before the call. So we know which ones are nit. We know which ones need to be talked about. We are really going to have to have that discipline. That means we are going to have to keep the number of items small, so we can keep it on those three.
MS. HINES: Even better, if our timelines work out, and we have got almost a month, almost two months, we could even have enough time to have drafts out with comments back and incorporated, so that when we get to the call, hopefully there will be almost nothing left to be done.
MR. LANDEN: For those of you who are not familiar with the ONC FACAs, the next scheduled webcast for them is March 8th. You can access their calendar at HIT.gov. If you want a comfort level of seeing how it works, I would suggest clicking in for at least a part.
DR. STEAD: Are we at the point where we can break for lunch, and you get an extra five minutes for lunch? We have cleared the April topic up. Our strategic conversation can pull forward to 1:45. We will be back here at 1:15.
(Break for lunch.)
A F T E R N O O N S E S S I O N
DR. COHEN: We are on to the discussion of Next Generation vitals hearing. Everyone should have received a hard copy of this outline. Based on the discussions we had this morning around the report to Congress, we have developed some new language that we think will work there and also work here as a broader introduction to describe the context of what it is we are trying to do. Do you want to read that, Susan, or do you want me to?
The new introduction, and think of this as language that would also work in the last section of NextGen vitals in the Congressional report. The vital statistics system, and this does need some wordsmithing, but this is the general broad overview. The vital statistics system is a foundational national data system. Its primary components are birth data and death data, with additional data systems for fetal deaths, and marriage and divorce statistics.
Vital statistics are used for health research and demography. Individuals use birth certificates for routine activities, such as obtaining driver’s licenses and registering children for school. Commercial uses include establishing the basis for insurance benefits and filing claims for deaths.
Vital statistics are used by the Social Security Administration, Medicare, and other major programs to determine future revenues and costs. Vitals are a critical component of population estimates produced by the Census Bureau and are used by the Bureau of Labor Statistics as a basis for protection of the future labor force. Furthermore, vital statistics are a key component of identity establishment and protection as inputs for protecting individual identity and informing other national security needs.
As currently constituted, however, this federated state-based system is cumbersome and at risk. It has considerable unrealized potential for enabling public health surveillance, increasing effectiveness of health care and health financing, and being useful to business and commerce. To address these concerns, NCVHS is making plans to convene a two-day hearing to obtain input from key stakeholders, including HHS, other federal sources, state jurisdictions, researchers, health care providers and other experts.
So this is our attempt to be more inclusive and describe vital statistics and what it is we are really trying to do. Let’s talk about this. Tents. Yes?
MS. KLOSS: Could we add to one of the very first sentences a statement about the legal foundation? Later on, you say federated law. But I think it would be vital statistics system is a foundational national data system established by what set of laws?
DR. COHEN: It is not established by any federal law. It is actually a jurisdictional system established by 57 jurisdictional laws and regulations. I mean, that is a really important point that we should add.
DR. ROSS: What we have as a national system is really a compact built on practice over time of states to agree to work with NCVHS to compile state-based data into a national data set.
DR. COHEN: It is called the Cooperative Health Statistics System. That is a great point. We will add something about established by the 57 vital statistics jurisdictions. Can anyone name the 57?
MS. KLOSS: Just one other suggestion, I think in the very last sentence, don’t you want public health leaders, where it says who is going to come? Right now, there doesn’t seem to be any reference to public health.
DR. ROSS: On the sentence toward the end, I guess next to the last sentence, it begins with it has considerable unrealized potential. I think the it we are talking about here is the collective national dataset. There is a system in the sense of the compact, the agreement to practice of every state working with NCVHS to submit data. That is one thing that you would call a system. But it is a system of agreement. But what I think you are talking about there is that collectively, these data have unrealized potential.
MR. LANDEN: I would like to ensure that we include an important both technical and political buzzword of interoperability. The way to do that, I think, would be along Linda’s line in that very last sentence, include as among the invitee’s interoperability, health data interoperability, subject matter experts.
DR. COHEN: Thank you. Denise.
MS. LOVE: As I read through this, it says the system is cumbersome and at risk. My brain went at risk of? Could we elaborate what that means? I don’t know what the risk is.
DR. COHEN: Should we say fragile rather than at risk?
MS. LOVE: What we say in my world is that we have huge state-to-state variability. So collectively, the system is shaky. I don’t know how to couch it exactly.
MS. HINES: Precarious?
DR. COHEN: Any suggestions?
MS. LOVE: The collective systems at risk, but some state vital records systems run pretty good. They are pretty strong. Others are just not.
PARTICIPANT: I think it is at risk of not being able to be relevant. In an era when timely information is what matters, the system is so clunky from the collection of data to the curation and transmittal and et cetera.
MS. HINES: We just made the case of how relevant it is. I don’t think it is at risk of not being relevant.
MS. LOVE: Just say that – can we take the risk out? I am just saying and underfunded.
DR. COHEN: Do you want to change that risk to fragile?
MS. LOVE: Not meeting the needs of public health, at risk.
DR. COHEN: Does fragile work?
MS. LOVE: I think it is easier for me to read through that than risk. Risk means different things. Then when I read through, now, this is just me, so don’t take it personally, but I don’t see how vital records, it says effectiveness of the health care and health financing system. I mean, maybe I am just missing a function of the vital records.
PARTICIPANT: Or our understanding of most things, that is all.
MS. LOVE: I will let that go. I just was curious. I am in a health care financing sort of world, but that is fine, as long as it makes sense to everyone else.
DR. COHEN: When I think of, for example, Medicare-paid claims, need to establish, when somebody dies.
MS. LOVE: I think if vital records went away, you would still have a pretty robust health care financing system. That is critical. They are a tool used, but it is not integral to the system like paid claims.
DR. COHEN: Increasingly effectiveness of these systems. That is the subject of that clause.
DR. STEAD: If I could play agenda management, my sense is you have made huge strides. This is very close. I suggest we stop trying to finalize it as a large group and that we move on to working through the rest of what you need to do in terms of the hearing, if that is okay with you and the others.
DR. COHEN: If you have any specific suggestions as you review this language, please send it to me, Susan. And the workgroup that has really been leading this are Dave, myself, Vickie and Helga with huge assist from Kate and Rebecca. This is the planning team for this event.
The event is a one and a half to two-day hearing, although we can talk a little bit about what we mean in terms of hearing. Is it a classic hearing where we have individuals come and give us testimony based on pre-developed questions? Is there going to be facilitated discussion at some point or any group activities to brainstorm about where we go from here, which is less traditional? Vickie, do you want me to run through everything first? Or do you want to ask me a question now?
DR. MAYS: I wanted to ask you a question about that because it will make a difference in terms of what we do. That is, is it impossible for us to engage technology, so that we can have testifiers from other places or people participating from other places? It will enrich what we can do. If that can be considered, I think it tells us kind of what we can design.
DR. COHEN: Can we?
MS. KLOSS: We have had remote testimony at remote Privacy hearings.
DR. COHEN: So some of the structural questions we have, we would like to discuss. But generally, we thought the two days, there would be four discreet activities. The first set of activities will be hearing from experts around establishing the need for vital records. Again, not necessarily in this order, but vital records are used, as we have just discussed, for an enormous variety of activities, both at the individual, local, state and federal levels. Dave?
DR. ROSS: Yes, I have thought about this. My biggest concern is trying to jam too much into two days. On item one, would it be possible, and I really mean this as a question, to write a brief, several-page paper that explains this? This is really, as I look at this, it is didactic. It is explaining what this is all about. Why not put that in a paper and give that out to everybody’s reading, and then we start with the issues and challenges and things like that? Does that work?
MS. HINES: Just to play devil’s advocate, we discussed in earlier planning sessions that assumes everyone will have read it. It seems to me maybe we started off in an unconventional way with the 30-minute didactic on like what you were saying in an earlier conversation today about how a lot of people, including very important people with important roles, don’t quite have all of the jurisdictional responsibilities and authority. Level set, yes, if we could start off with a level set, then I think we could cut out a lot of that.
MS. KLOSS: I would just echo that the last three Privacy hearings, we have started out with an overview. We have done an expert just to level set the topic before we have gone into the specific hearing?
MS. HINES: How long were those?
MS. KLOSS: They were generally an hour.
DR. COHEN: Do we have resources to hire a consultant to do that?
MS. KLOSS: We didn’t hire anybody. We just invited an expert.
DR. COHEN: Okay. The person I was thinking of to do that was somebody who knows vitals, who we have hired as a consultant before. I don’t know. Let’s discuss how to do this. I am trying to think of somebody who would be an appropriate person. But we can work on that level setting. I think that is a good idea to bring everybody up to speed on the basics for vitals.
Then the major core of, I guess, that morning’s activity will be testimony from a variety of perspectives about the current status and challenge and fragility of the existing system. The perspectives from the jurisdictions and I think from NCVHS, some of the concerns around funding and legal requirements, issues around data-sharing and transmission, and content, timeliness and quality. These are the major topics that we could think of.
One of the things that we want to do as we prepare this outline and move to the next version is I think really circulate this to DVS, at NCVHS, and to NAPHSIS and some of the folks who are in the trenches to make sure that we have covered all the important topics with respect to the current status and challenges. Vicki?
DR. MAYS: You were asking a question on whether it would be a workshop or something else. I am wrapping my head around it is a workshop then or hearing. Okay. Don’t we usually start with a set of questions, so that we can get everybody to land? Is it that each of these ABCs will be turned into a question, and the question is what are?
MS. HINES: I think the Privacy model and the APCD model were really excellent. Each section, and Denise, you can remember this fun we had, has a set of questions that get sent out to all the people who come in. They prepare written testimony, which is usually much more thorough than we have time for. Then the juice is when you have a panel, so let’s say you have a topic like data-sharing transmission and protection, and then you have three people testify for maybe 15 minutes each.
And then you have an open panel for them to take questions from the members is one way to do it. But you give them questions well in advance, so they can prepare written, as well as have time to speak and then have a dialogue. It works very well because it is a mixed model, if you all think that would be a good way to go.
But it is a both and. You get some time with them giving and downloading their best thinking, their expertise, and you get time for dialogue between them and with the members. It just means you have to be very careful. You can’t over pack it. We have made that mistake before. You have to be very careful with how many people you invite for each section and how long the dialogue goes.
DR. MAYS: I am fine with that. I thought when you were saying to distribute this that you were distributing it to get comments from people that then would be our questions, as opposed to sending it with a set of questions that we are getting comments about.
DR. COHEN: I think once we finalize what the agenda for the day will be after we get another round of feedback, then we will begin identifying individuals and preparing questions for the different topics to send to the panels or the individual testifiers. Does that make sense as an approach? You look a little confounded, Vicki. That is not what you were thinking?
DR. MAYS: Let me listen a little longer. It just is this round that keeps going out. That is why I am confused. It sounds like you are saying we are going to send it out now as this. I am not sure to who for what.
DR. COHEN: I guess we need to discuss this more in the planning group. I want to make sure we haven’t missed anything significant that should be here. There are others who know the vitals landscape, particularly the folks at DVS and at NAPHSIS, for just a quick review. Does this make sense? Is there any glaring omission? That is all I want to do. We are not asking them to do anything more, at this point in time. Then we will develop questions and identify all of the folks who we want to be involved.
DR. RIPPEN: I am just going to link the two and the three. It just depends on ultimately how we think it will run the best. If you are asking people to testify on the current status and challenges, that is already an issue. You could also consider asking them about what do they think might be a solution to be discussed later.
So what you have then is a deeper area of expertise that already has thought about it in their kind of domain. You can accelerate the process. So that might be another way for us to get kind of a robust discussion going on. Just a thought.
MS. KLOSS: Just the thought process that we have gone through in designing hearings in the past has been to think about it in terms of panels. If you have got a day and a half, roughly you will want to reserve a couple of hours at the end for the subcommittee to digest what is heard and to do its work. You can back into how many panels you can comfortably have and allow enough of the dialogue, such as Rebecca is suggesting.
So as I look at this, if you covered the overview, which is essentially the first topic, the importance of vital records would be your overview topic. Then number two is current status, and you might give some thought to current status at the state and local level, current status at the federal level. So that could be actually two panels. Then emerging best practices and barriers or something could be panel three and four.
Then as a practical matter, what happens is that everybody touches on all the topics. But at least you have sculpted some. I think you probably need four panels and introduction, and then two hours at the end to digest. Then you will develop questions around this outline for each of them.
DR. COHEN: Great. That sounds like a really nice approach and a very efficient structure. I need to think about it a little more because when you talk about the current status, the federal status, they are going to be very different perspectives from a variety of federal agencies. NCVHS who collect the data and Social Security. There is not one status, but a panel can. Good. Excellent idea.
Let’s move on to three. Three again, as Helga and Linda mentioned, it is sort of the continuation of the discussion of two. Here are the issues. Everybody who has thought about the issues has thought about what they would like to see done. There are other people who might have different perspectives on what they would like to see done to improve the system from the local and state level to the national level in terms of making this system more robust, more consistent and more timely.
The list of issues, and three is similar to the list of issues in two. But as we move towards maybe a panel configuration, they might align even better. Again, everybody has different levels of expertise in the field of vitals and different perspectives. We want to cast researchers’ need for vitals and worldview of how to use it will be very different than an insurance company executive who wants to be able to clear the death files immediately once someone dies to make sure they are not paying claims. There are lots of important perspectives that we need to figure out how to address. Rich?
MR. LANDEN: As I am understanding this now, section one is level setting, section two is identifying pain points and known, but unobtained, opportunities or needs. Three gets more into possible solutions in all or in part.
MS. HINES: So my notes say Linda said that after the level set, the first would be current status challenges at the federal level. Then the next panel would be current status challenges at the state and local level. The third panel would be emerging best practices.
MS. KLOSS: Or flip them. Sometimes we found it works best to do the barriers and then —
MS. HINES: Do the barriers, okay, and then emerging best practices. Then the last piece would be time to integrate and discuss. I have written those down, so I can email it. Then you can play with it.
PARTICIPANT: Then you get the diversity of opinion by who you invite to be on the panel.
MR. LANDEN: In my mind, though, best practices is not synonymous with solutions to problems or ways, methods to obtain opportunities that are envisioned, but not realized. As I am thinking at this more globally, I am not sure that we have got a real solid outcome as far as solutions. I am wondering. Question number one is wondering what you are envisioning is what is going to be the substance of the recommendations that we expect to come out of this.
Number two, totally different thought, is are there similar industries outside of health care, outside of vital statistics, that have had similar problems and have indeed come up with solutions. One that occurs to me is college transcripts. In my day, they were all paper. It was a pain in the butt. Now, it is all electronic. I have no idea how that happened or who did it or how well it works, but it is there.
DR. COHEN: So, my response would be I heard Rebecca read what the panels were. I would say the last several panels would be existing model practices and options for future improvement that don’t already exist. Then a discussion amongst the committee considering all the testimony of the day. So it would be all of those things in the latter part or the second day, the middle and latter part of the second day.
One of the challenges here is I can’t think of another federated system that is developed using state and local regulations, but has such overwhelming national impact that we rely on. So the model, I don’t know whether it is unique, there might be some other kinds of systems that transmit information. But you know, the legal responsibility and the development lies all in these jurisdictions. But there is an overriding national interest to use the data. I think that tension is something that we will need to explore and really balance.
One of, I think, our challenges in developing this will be we certainly don’t want to threaten this federated model that relies on jurisdictions to be in the trenches to collect these data. At the same time, it is really important that we achieve. I think we all feel that we can achieve more national coherence and benefits. That will be part of the discussion. I don’t know whether that addresses your comment.
DR. STEAD: Let me play agenda police for just a second. We have used the allotted time, but we have 15 extra minutes. We can elect to move. If more time will help this, we can elect to continue to use more time. Or we can bring this to closure and move ahead of time into the next round of discussion, whichever you prefer. I am comfortable with you doing that. We have reached that.
DR. COHEN: We have gotten really good feedback so far. If there are some other strategic or generative comments that will help move us forward in redesign of the day and agenda setting, I would really appreciate them now.
DR. ROSS: I think that Rich gave us a little bit of a homework assignment that I think would be useful. Just for the record, that we look at some comparator systems, whether it is reportable disease data. I think of the highway trust fund. We have reached an agreement in this country how to set standards and fund a national purpose, but in collaboration with the states. Are there comparators that we could bring forward and help people think about this? That is homework.
DR. COHEN: Thank you for clarifying that for me.
DR. MAYS: A couple of suggestions and then a couple of questions for clarity. Given that this is a federated system and, as you said, we don’t want to threaten people or anything, it might be useful for us to think about whether to set up a mailbox and ask everybody to respond, to get comments as widely as we can. We then just build into, again, I don’t know the timeline. But if you could get comments from everyone, and then make sure that those issues are covered in some kind of way, so that even is part of the question setting. I think we would have greater happiness among everybody that had a chance to comment.
The other is I want to make sure that I understand two things. We are talking not about like problems at the level of individuals, but we are talking about things that fix this as a system, correct? This is a systems, a redesign almost of systems. I shouldn’t say a redesign. I think throughout, that has to be like said, so that individuals don’t kind of feel like their questions or issues are not being heard.
Then my third thing is, I want to have a sense of exactly in vitals what all we are including. One of the things I brought up before is the National Violent Death Reporting System. That is part of mortality. That is part of what helps mortality.
But if we are talking a system, I don’t know whether then that is connected or not connected to it. Are we ever going to talk about other things like marriage, divorce, those kinds of things? If you can help me to make sure I am understanding why you are landing where you are landing and what it includes, I think it will help the clarity.
DR. COHEN: I like your concept of system. I think that is where, for this first set of hearings, we need to be and focusing on births and deaths, rather than the other vital statistics data collection mechanisms. I think the systems will move us away from considering specific content and uses right now. But it is entirely possible as we move forward that this will lead to more consideration and recommendations in that area.
DR. MAYS: So you would or wouldn’t include the violent deaths? It gets into law enforcement officers.
DR. COHEN: I wouldn’t include any specific examples of uses right now. I think we need to focus more on the general systems, what is working and what is broken and how do we fix them in general, and then move on to that later.
MS. KLOSS: Just a comment that is maybe more of a mindset comment. I think the purpose of the hearing is just to kind of throw it out there and learn as much as you can without trying to think through too much about what recommendations or what you are going to end up with. I found the secret sauce based on the good staff work we have had from Mia and Rachel and others is to get the right people plugged into the panels and the magic will happen.
Then just kind of, I don’t know, don’t worry so much about what will come out of it. Just approach it as a learning and not over orchestrate. Then you will see you may need a workshop to follow it. This is a broad, quick way to learn what is very cross-cutting at a lot of different levels from the experts.
DR. COHEN: Great advice, thank you. If anybody has connections in the vitals world and can think of people whom we should be considering now that you know we are looking at the systems of vitals, please let us know. But that was a great way to end. This has been great feedback. Thank you all.
DR. STEAD: So we will switch gears and step back and think about our work plan. What I am hoping we can do, we have got two very important topics that we are hoping to work in earnest in 2018, unless something comes out of the administration that suggests we should do something different. Those are the ideas of beyond HIPAA and the terminology of vocabulary work.
What I am hoping we can do is think a bit in the next hour about how to use time in our June, September meetings to refine our understanding of what we want to do and what the state of play is and how we want to tackle it. So that we would then by our November meeting be in a position to build it more concretely into the work plan for 2018, if that still seems appropriate. That is the purpose of this discussion. It is really trying to look out and figure out what advanced work we need to do, to know how to frame what we are trying to do in this space.
Both of these spaces are, as most things this committee does, bigger than a breadbox. That is the purpose of our conversation now, if that is comfortable with people. I think it will influence who we ask to come to play. I do think, from my perch, this meeting has shown that the conversation we had with the evidence-based practice commission, the conversation we had around the Social Security number change, those kind of conversations are much more useful than general updates because they are really targeted.
I am hoping we can derive things we need to know from these two things into the next couple of meetings and minimize the amount of general updates that we are doing. That is another think that is going into it. We have asked Linda if she would maybe help us think through this, since she is helping us guide both of these scoping topics as they are at this juncture.
MS. KLOSS: Our purpose is really to generate some discussion around each of these topics. Maybe we can commit ourselves to 20 minutes of good discussion on the beyond HIPAA and then 20 minutes on the vocabularies and classifications topic, and then sort of pull it together. Is that fair?
So beyond HIPAA, as you will recall, has morphed some. We went through the kind of detailed scoping document at the last meeting. But we started by thinking of it as the future of privacy. It has changed a few times. But it still is primarily privacy based. How we look at this space outside and beyond the boundaries of HIPAA. We have touched on it in a lot of our recent work. It just keeps coming to the boundaries.
The real question for us is to advance this, how would we frame the questions? Then how would we move at trying to get our hands around them, our arms around them? I wish we could pull up that framing document. Do you think there is any way we could find it?
MS. HINES: Yes. It is in the agenda book. It is on that computer. Right, Marietta? It is the one – 7.1.
MS. KLOSS: Bob, you worked on this, too, and Barbara. When we look at the space that this is, we are really broad. We are really strategic. We really have to stay at a pretty high level, or we will get buried in a minute. The background, all the issues we are bumping up against, big data in analytics, cybersecurity, balance between individual privacy and socially beneficial data access, certainly things we have touched on in the DID letter.
Recurring calls for individual ownership of data, population, community health, personalized medicine, all of these uses for data that continue to expand. And the technology, mobile technologies and bio tools and so forth. Could we scroll?
DR. STEAD: Let me just pause at one, at the risk of taking us in a rabbit hole. Just maybe give an example that we could begin to engage in. So take cybersecurity challenges. Right now, the policy framework for security in HIPAA is built around an annual risk assessment and management plan done by covered entities. In essence, I think that is the policy framework.
I would pause it that framework is necessary, but not sufficient, for the current rate and pace of cybersecurity threat. So does there need to be some other lever that is affecting, is changing the playing field that individual organizations are working on when they do their internal risk assessment and management plans? If so, that would be a lever that was beyond HIPAA.
I am just trying to get an example. We don’t need to go into that example. But I think we have got to get. I really do like this idea that when we design the 2018 work plan, our intent will be to really have a good draft of a framework of principles. Right now, we just know that is a concept, I believe.
MS. KLOSS: We said that really as an outcome, what would be a reasonable outcome would be the output would be policy guidance relating to framework of principles and so forth.
DR. STEAD: I guess another could be what policy levers do we need around individual health information or individual information that is outside of covered entities that is not a policy lever around making them covered entities, but is actually a different kind of policy lever of some. If we get this predictable standards and operating rule process to where it is coordinated and aligned and planned for, it will still move a small fraction of what we have got to move to enhance interoperability using non-transaction based approaches to how we make information, et cetera.
I don’t know. I am just one person trying to think about current policy levers and how we begin to think about what those next steps and pieces are. And what do we need to learn that would help us have this conversation. I am seeing a lot of blank faces. I know we are after lunch. But this has to become real to us, or we are not going to get anywhere.
MS. KLOSS: One of the starting points that occurred to me, and I will just throw it out as a discussion item, we tried to tackle this topic through the framework for community use of data, that letter from the secretary from the end of 2012 with principles. I think that is probably not a bad starting point for thinking, trying to articulate what the principles for the world outside of HIPAA might be.
I think that is supported by so much of the work, for instance, the privacy tiger team that ONC did. A lot of the recommendations that our subcommittee has had over the years, it always comes back to what are the principles. We go back to that same set of principles of fair information practice. It seems like that is a starting point. But it is abstract. It is a hard thing to work on.
We heard testimony in the DID relating to let’s get back to principles and look more broadly at data management practices. So we can start by doing some work ourselves to understand how these principles can be used. I mean, that may be not a bad starting point. What are they? What are the different sets? How are they used? Where does the guidance come from that always says when you are trying to grapple with these new things, go back to what the basic principles are? We could do a session on that at an upcoming meeting.
DR. STEAD: So basically review the principles in the 2012 letter.
MS. KLOSS: Well, we could even go a little further. I recall at that time what Leslie Francis did on our committee, was she kind of went to all the iterations of fair information practices and pulled together a grid of what they were. We kind of used that as a starting point. There are certainly experts that spend their lives studying this. We could even have somebody come and talk to us about it.
DR. STEAD: And do those practices leverage modern technology? So if part of our problem is that our current policy framework is being overwhelmed by technologies that are much more ubiquitous and may capture all sorts of new stuff about us, practices that were designed for the era that Rich just described, no matter what we do, aren’t going to scale to that.
One possible answer here is we would actually discover that framework. What would a new framework look like? Could you end up with a traditional and next generation table that put the levers side by side? I don’t know.
MR. COUSSOULE: Are we trying to focus this more specifically on security privacy? I think it is riskier as a scope issue.
DR. STEAD: I think at a minimum, we have got to do two things, and maybe we do them in sequence. We talked about the predictability roadmap. I think we have got a huge thing in the space around privacy security. I think we have a huge thing in this space around what the standards look like in a rapidly-changing world. We may want to make those two separate topics both tackled under the broader umbrella of beyond HIPAA, but as two separate. That would be an advance to even tease it out that way.
MS. GOSS: Building on that because I am struggling with the fact that we have made this really big. It gets really difficult to produce something concrete on the other side. But there are also, to Linda’s point about principles, there were very specific principles around the standards adoption and initial HIPAA rules that came out after the public law 104-191 whatever that led us into actually having the HIPAA regs.
I think that starting there, and I am almost envisioning this. We had these guiding principles in an era that EDI transactions look like big new stuff. Now we have got fire. I think we have to almost look at the world as to what were the principles we have been using, what do we think the principles are. Then really, the principles may still change, but the way we look at them or the nuances may change. I think there is some goodness in having some conversation around that.
I also am hearing this idea of our CIO roundtable like maybe it is important to hear from the industry as a whole about what they think around this topic. Maybe we could put them in two buckets. It has to be collaborative.
DR. STEAD: Let me ask a question. Could we envision a first step toward a CIO roundtable being a panel at either our June or September meeting that is part of the strategic block where we take an initial dive into it?
MS. GOSS: I think we need to do some homework first and have them react to some things. I think we need to get the pulse of how they are viewing these things. A lot of what was developed 20-some years ago, was based upon industry status.
DR. STEAD: I understand. One of the things I am asking is what a useful thing for us to do is to bring three or four together into a panel in one of these meetings, as an initial step to take a pulse that would actually help us design how to do and how we want to do something bigger or not. This was originally your idea, I think, Rich.
MR. LANDEN: I am looking to our resident guinea pig here.
MS. KLOSS: We could actually put some questions to them and say how would you be working through these issues today? What would you look to, to help you figure out how to deal with things that really aren’t covered by law or regulation?
MR. COUSSOULE: One little twist on that. It becomes a distinction on what we have to do to meet our statutory allegations and what we choose to do to be better. I think that is where we get into some fuzziness. I think it is worth exploring. When you look at the existing rules and regulations, and we say, this is what we know we have to do to be compliant. Compliant is a floor. It is a not a ceiling. Then it is a question of what is next and what else is there that either new regulations or changes could help, what is missing? That is the way I would view it. I think that could be very useful. Does that make sense?
MS. KLOSS: You could set up a few scenarios and have a discussion around that.
DR. COHEN: I would like to step back from my perch, which is different than the standards and privacy perch. When I think about beyond HIPAA, I think about envisioning what data access and data exchange will look like 20 years from now. We are talking mobile health. We are talking about whole new streams of information and exchange that aren’t confined by the ways we have traditionally looked at data and information in this space.
It would be wonderful if we want to be truly proactive and think to sort of opine and get the experts to help us. What is the scope going to be like that? I think the principles that we have talked about are constant. But they are going to be applied to a different paradigm of information exchange and data access. To me, when we talk about beyond HIPAA, we can’t be bound by what we have seen so far. It would be wonderful to get strategic thinkers to help us begin that kind of vision.
MS. LOVE: What has changed in 20 years, I think the principles and the boundaries around HIPAA then were around protection and how to guard and save the information, which is a good thing. But I think 20 years now later, we are struggling with how to shoot information back and forth, and how to open the information. To me, that is a huge paradigm shift.
I lived through the first HIPAA. It was how to protect. Now we are saying, how to unprotect safely. But Bruce, you said it. Data policy at the state level, it is just not an industry conversation. I kind of was hearing CIOs, how important they are. But you have a lot of folks doing intricate data policy out at the frontlines of the state and federal levels, as well.
MR. LANDEN: Yes. A lot of what they said, but also going back and looking at why HIPAA. How did that come about? HIPAA was passed in ’96, but actually work on it began with the WEDI Report of 1992, which was the result of several unsuccessful efforts for the industry, providers and payers specifically around claims standards, to come together and figure out. To acknowledge that how they were working was inefficient, was expensive, it had, what is the buzzword, a lot of dissatisfaction attached to it.
It could be done better if the industry came to a consensus on how to do it. But they were unable to do it voluntarily. Out of the WEDI reports in ’92 came legislative drafting in ’94 and the legislation in ’96, which essentially took the ideas developed by the industry and codified them in legislation. Now that was all predicated on how payers and providers did business back in 1992. Well, they do business very differently now.
So the fundamental question in my mind when I think of a CIO beyond HIPAA roundtable is, okay, here is how we did business in 1992. Here is how we are doing it now given HIPAA. Where do we need to be 10 years from now? What does NCVHS need to do to start the process to enable us to get to where we want to be in 10 years given everything HIPAA has enabled us to do that was not part of what was built into HIPAA originally because the industry just wasn’t that far.
MR. COUSSOULE: I think Bruce raised a very interesting point in that it depends on how far out we want to look with what our recommendations are going to be. There are issues and opportunities today that are either gaps or new things that have happened in the last 10 years. There is also stuff coming, whether it is the internet things or the data access, those kind of things, which may not even be quite here yet, but that are coming as well.
So I think we need to think through about how, if we want to focus on what I will call kind of today issues that we know are real versus what we believe is coming and what those impacts might be, or both. I think they are both valid. But it will become an interesting scoping exercise to figure out how we deal with that from a short-term recommendation versus a more structural long-term recommendation perspective.
DR. STEAD: In my gut, which may be wrong, is that we want to take a long-term vision, and in that process, if we can identify near-term recommendations, craft these recommendations. Otherwise, follow the vision with a more focused work product activity that would develop recommendations within a scope, but living behind the vision. I find the way the pop health work always goes back to the 2002 figure. That figure is current. Yet, it is 15 years old.
I think at the level of beyond HIPAA, we are looking for the 2002 figure or maybe a couple of them. Then step back and make near-term recs to the degree we can, if that makes sense.
DR. MAYS: One of the things that when you started talking about beyond HIPAA is, in the start, we were worried about entities. We worried about the health care system. We are starting to worry about people engage their health outside of the health care system, but it is still in relationship to the health care system.
What happens is that we ask them to do things. We ask them to use things. One of the things we had talked about, this is very early in the workgroup, about the ways in which, for example, people use apps. They sign to use the app. It may be collecting their blood pressure or something like that. But they don’t know what the terms or conditions are, and whether or not there should be some kind of fair practice reporting that you understand what it is you are doing. Then there are these analytics that are using your data. Those analytics then are big business.
So the issue is there is a growth in like wireless monitoring. There is a growth in these things. HIPAA was really about kind of the health care setting. In terms of when we are asking people to engage in well-being and to ask them to engage in these personal health activities, do we have a responsibility at all to help them be more educated about the use of their data and protecting their data and what fair use is of that data?
DR. PHILLIPS: I like how this is shaping up very much. I think to Denise’s point about HIPAA being about protection early on, we have had so much legislation just in the last couple years. It is about lubrication. It is about movement of data, whether it is the 21st Century Cures Act or whether it is macro.
The pressure on physicians to let data flow out of their EHR is so high that they are behaving like patients who sign their HIPAA thing because that is how I get my care. I sign this, and I am in. If physicians now who are singing BAAs with the number of people, they have no idea what the content is. They don’t frankly care because they have to move data. I think we need to figure out how do we get it to catch up with where we are now. What are we looking to do down the road that it may need to anticipate?
MR. LANDEN: I talked about what were the business practices back in the late ‘80s, early ‘90s. But she brought up the new technologies, created a whole new set of business practices that didn’t exist then. Policy-wise, the language has shifted from system to conversations to person-centered in the health care delivery system. We are talking about patient-centered. If we go into the wellness pop and pop health, we are talking more about wellness, both individual and population.
Then we have got these fantastic apps that are doing stuff out there. Those apps can be part of the programs both by employers or by health plans. We don’t want to get into the difference between those two and HPID. But we have got those as part of business model. That is an example of business models that have been enabled by standardization, that did not exist in the early ‘90s, that nobody thought of in the early ’90s, and are yet a fact of life now.
So yes, in the mid-term, we have got to address that. But long term, we have got to anticipate what new apps have we been enabled that we don’t know of now, but will be commonplace by the time we come out with our recommendations.
Finally, we have also got to think about NCVHS as an enduring institution and recognize that in a few short years, in our place will be sitting a lot of gray-haired millennials. What do we have to do now in this planning process to help enable their success?
MS. KLOSS: I think it sounds like our consensus to go along. I like that a lot. I think that is kind of where we started thinking that maybe there is a 21st century kind of diagram that could emerge, that could guide a lot of our work and help us set priorities.
Then the question is how do we begin that go-along process. We would really need some sort of visioning exercise, I think. Not necessarily a roundtable, not necessarily a hearing, some sort of visioning.
DR. STEAD: My sense is we are thinking we are going to move forward, maybe thinking of this as a two-track process, one focused around privacy and security, and one focused around the next generation of standards and so forth. I am hearing that we want to have some discipline in characterizing the practices that were in place. The world when HIPAA was done, the practices and principles that came out of that, and set beside them the world of today and what those practices are beginning to look like.
I think it might be useful to have a session, Rich, I am looking at you. I think it would be very interesting to pose to a few, maybe like three, select CIOs. Just like you said, we got here because the industry developed practices that were in time reflected in legislation and regulation. Tell us the things you are doing that you are not required to do, that are examples of the things that those could be things that became embodied in downstream stuff.
Then tell us about a few things that you cannot do because there is no enabling. Things you want to do, would do if you could, you are not required to do, but you can’t do because they are not in the enabling practices. Something like that might begin to bring in. I think we are going to be helped if we can begin to bring some external voices into our internal planning. That might help us understand how to do something bigger. I don’t know. Would that be possibly a useful way to pull in some information?
I like the idea of going back and maybe making a list of what were the key characteristics of the world in ’92. We might begin to build some little building blocks that could be short.
MS. KLOSS: Another little interesting angle on that, if I recall history well enough, the privacy piece of this got added to the law because we weren’t going to be successful getting a preemptive privacy law as a standalone. It got rolled in. So we have always had this kind of a little bit delink. That is one of the challenges now that the privacy and technology need to be more tightly knitted. That history does tell us something about today’s challenges. Well, we will have to do a next iteration of thinking through how to proceed before.
DR. MAYS: One of the suggestions I was going to make on the technology side is one of our workgroup members is Wendy Nilsen, who is at NSF. They do all this training. Some of these very issues are what they are doing. Every year, well, they do probably about four or five classes where it is about technology app development. But they have lectures where they worry about some of these issues of privacy and confidentiality. They try and help the engineers to understand a lot of this stuff.
I mean, she is one of ours, so asking her to come and talk about this and what kinds of things they are worried about and what kind of advice they are giving people might be also early on some kind of useful information.
MS. KLOSS: Do you want to talk about where this falls in the work plan now or go onto —
DR. STEAD: My sense, I think maybe we go to terminology. I think what we want to reflect on from this is what would be the most important things in this space to think about incorporating in the June meeting.
MS. KLOSS: It seems to me that the committee could do a refined description of what we want this project to look at in the June meeting. But that we would need to anchor some kind of in-person work in September because that would be our last opportunity for in-person. I don’t see taking up this topic on a conference call.
DR. STEAD: I think that is right. I think one action is we will need to update this to reflect what we have learned. I will be glad to work with you on doing that. Does anybody else want to be engaged in that little team? Okay, we can flip it back and forth.
We can have another round. Once we see what we do, we may also want to get a first external. We can decide do we want to do some things in this realm like begin to go back and do a draft before and after table to the degree we can do that. We just need to see how this goes.
MS. KLOSS: There may even be some interesting readings that could precede the next discussion.
DR. STEAD: I think taking this time now and spending similar time as we work through the fall will really let us set up. We will be in a much better place when we are trying to really land next year’s plan. So we are going to morph to technology and vocabulary?
MS. KLOSS: It is not in there.
MS. HINES: It is in there. Marietta, do you want to pull that up.
MS. KLOSS: So we have gone in a good sequence because the terminology and classifications is a little narrower than the big universe of beyond HIPAA. But you will recall our discussion there was to step back and take a broader look at our national approach and capability to update and disseminate code sets and classifications, taking advantage of technology and better meeting the need for timely updates.
This is a good time to step back and do this because we have got ICD 10 operating, and we are still some years out from ICD 11. What could be learned by looking at how we prepare for 11 differently and also look at the whole universe of code sets and classifications that we use, and kind of the disharmony in how they are maintained and disseminated? Is that a fair recap?
DR. STEAD: From my perch, the sound bite is, okay, we have updated ICD 10. We then had a period of time, and we have lived through a big bolus update of all the stuff that had to set while we work through the go-live. Unless we do something different, we will be doing the same thing with ICD 11. We will always be very far behind the curve.
The question is whether we can come up with some more continuous process for updating our vocabulary and terminology. The rate of pace of biomedical concepts is dramatic. I think that is the nut of what we want to do. I don’t know if we have got that captured well in the purpose.
MS. KLOSS: Well, in scope, current terminology and vocabulary mosaic, how this compares to approaches taken by other countries.
MS. HINES: I think what Bill just said is so clear.
DR. STEAD: I am really thinking about what we want in the purpose statement.
MS. HINES: Basically, the rate of pace of changes and updates is dramatic compared to previous years. Can we develop some more continuous process? We need to say that.
DR. STEAD: That is in essence what the question is we are asking. We have got that on the tape, I presume, or Rebecca has got it on her fingers or under her fingers. We will capture that and can bring it back into it.
I think what is in scope was actually pretty — yes. Let’s look. You are right. This is bigger than a bread box, but it is not a 20-year vision statement. It is actually posing the question, can we modernize the approach. Is there a way to do it actually in time that might affect the way we get to ICD 11? Because if so, it would really be a highly-leveraged activity.
MS. KLOSS: I do think there are some things that we could do this year to prepare ourselves for 2018. I think it would be very useful to have a panel at a meeting where we invite the National Library of Medicine to describe how it functions as our link. We need to understand how ICD 10CM and how 11CM is getting developed, where it is at in its development and how it is similar to or different than 10.
We need to learn more about ICD 10 PCS, which is developed by CMS and how that maintenance of those two systems progresses. I think those are pretty fundamental.
DR. STEAD: I keep asking about RX Norm, and you keep shying away from it. I think RX Norm, I don’t know if either — I think RX Norm is a weekly update cycle, if not more often. I think so. We can find that out. I think it may be pretty close to something that is now organically maintained and developed as new drugs are approved and so forth. I think it may be a fairly normal thing. If so, I think that would be a nice example to include in these.
PARTICIPANT: Are you suggesting we ask them how do they do that?
DR. STEAD: I think in what would be four of these, that what we might be doing is trying to get the four people in a block. We have had some drop out here. I know Bruce at least, did not grasps that fact that we were changing. We have deliberately decided to bring this meeting to closure at 3:00, so people can get out. But that means we have got to have people there until 3:00 if practical.
If we are going to have like these four things, that would probably become something like a two and a half to three-hour block. More like three, if we are going to have Q&A. I really think somebody would lay out how they are currently doing it. We would ask questions. It would be spinning us up and beginning to build the fact base to let us decide how to tackle this.
We could tell them the reason that we are interested in this. We are asking the question in our purpose statement. That might be something that we could organize for June. If we did that, that would let us do as you suggested and more slow walk the other towards September.
DR. MAYS: We are bringing the National Library of Medicine to the June meeting. I think they are doing their strategic planning. I don’t know if it is going to be the same person, but at that meeting, it might be useful because we could then feed in the things, particularly if there are things we didn’t put in the letter to the Secretary. But we could actually then begin to explore a space with them.
DR. STEAD: I think I would keep these two ideas apart. We might could still do it. There are four strategic planning panels. One of them is around the work we do around data for public health and so forth. I am on that panel.
The NLM is doing the strategic plan the way they have always done it. They have four different planning panels, which that doesn’t mean like a presentation panel. That is a couple of months cycle that produces one component of the strategic plan.
What I am hoping to do is actually try to insert some of this directly. We will see whether I succeed. That is another reason I want the health data framework thing in a form that is insertable. So let’s treat that as a separate problem because these actually would be different people. This would be different people. We would actually probably need two different people, one around RX Norm and one around the connection to SNOMED. Those are, I think, different activities.
MS. KLOSS: Then one from NCVHS and one from CMS.
MR. LANDEN: In response to your earlier question, Mr. Chairman, according to the National Library of Medicine website, RX Norm, the full set is updated monthly. There are updates weekly for newly approved drugs apparently. The comment I really wanted to make, though, is the cycle on which a set is updated, is not necessarily synonymous with how often it is consumed and put into production. You might have weekly updates on a given set, but the clients might only update once a month, once a year, if ever.
DR. STEAD: We can find that out.
MR. LANDEN: If you are asking can we find out how each and every client updates, no, we can’t find that out.
MS. KLOSS: This is one of those trajectory things. It is not saying we are going to recommend absolute change.
DR. STEAD: Lorraine, posing as Alix.
MS. DOO: This has to do with the last discussion. But it was if you had not already, related to the privacy and the discussions about mobile apps and access to information, how many of you were already familiar with the health IT play book? Okay, good. I thought that should be referenced in that discussion because a lot of what you were saying is actually in that playbook. They talk about all of those bits of information and the relationship between ONC and the FTC and some of those criteria.
I didn’t know if you wanted to make a reference to it, or if you wanted ONC to come and talk to the group about how that had been developing. Because a lot of what you are saying could fit in really nicely there. I just didn’t want to lose sight of it for this group because of what the HIT policy and standards committee had done that was connected to what NCVHS was talking about.
DR. STEAD: Other comments? We have got a rough plan. We have got a way to translate that plan tentatively into both June and September. Then is there anything else we need to do around the work plan? Are people pretty comfortable that we know we can sort of sketch in where it is? Maybe we look at it reflecting this conversation, make some edits from our perch and flip it back out to the executive committee to make sure we have got it right.
MS. KLOSS: I have just one follow-up item that I would like everyone’s concurrence that we should do this. The issue came up about the need for current guidance on accounting for disclosures as an area where we still need HIPAA guidance. I think we mentioned yesterday that we thought it might be a good idea to talk to the Office of Civil Rights and see if they would appreciate our looking into that.
Now, that would be taking on another project, probably the size and scope of de-identification. But with your blessing, we will have that conversation and see whether they think this would be useful.
DR. STEAD: You have our blessing to have the conversation and to bring back the answer, without committing that we will do it. So that we can have the conversation and look at how to sequence.
MS. HINES: Can you just synopsize again what the topic is?
MS. KLOSS: HIPAA requires that covered entities provide an accounting of disclosures to patients or to individuals who ask for that. It is just an area of practice that probably isn’t as mature as intended to be.
DR. STEAD: The other thing I ask, and I would like to capture as we go through these meetings, things we want on our pick list of future topics we may or may not get to. I asked Rebecca to put on and follow up to the workgroup discussion, a project around ways to approve HHS’ internal data infrastructure and things like the possibility of chief data officer and so forth. That whole bucket of stuff that we can almost pull out of we are not going to need it for what we are doing right now, pull it out into a version 0.0 of a scoping document of a task that could be linked to this. I think that would be another thing we may want to look at downstream.
Let me just summarize what I see is the immediate next steps from this meeting, and see if people think my list is right. First, we need to finalize and post the Pop Health report after making some titling and so forth changes. We need to finalize and send the DID letter. We need to complete the edits to the RTC. Our goal is to have them really in pretty final form by the March Exec Committee call.
MS. HINES: I think we were going to send out a next draft for everyone to look at and then a next draft for the March 29th. Okay.
DR. STEAD: Correct. We want the next draft before the March 29 one to go out with a few days’ notice, so that people have a chance to provide email comments that we then discuss on the call.
MS. HINES: Now, do we send the next two drafts of that to the whole committee since the whole committee provided input?
DR. STEAD: I would wait and send the version. I think I would probably wait and do it just to not pepper. What is the group’s call? Would it be best to go to the whole committee since we have been through it at that level? Would you like to be in each of these loops? I am trying to protect people’s time? Because all the inputs came from everybody, I think you are right. Given the level we went through it here, let’s send it to everybody.
MR. COUSSOULE: I think you started with the whole committee, and then you skimmed down the next iteration.
DR. STEAD: Then the HP ID hearing plan, we will likely be asked to do something before the September call maybe by email.
MR. COUSSOULE: Our plan right now is we will start working on the mechanics of that to be able to bring up at least some thought process to the next Executive meeting.
MS. KLOSS: You mean the April call, Bill?
DR. STEAD: I am afraid the health plan ID hearing plan is going to have to be approved before the March 29 Executive Committee call because of the timing. If we were actually going to do something in April, we will have to approve that by email. If it is more like May, we can approve it on the March call. We will just have to see.
The Health Data Framework sort of process timeline for closing that down, you and Eric might sort of say where that — no, I am saying the HealthData.gov.
DR. MAYS: Rebecca and I are going to do it relative to the deadlines that we have internally.
MS. HINES: We just know we have to make the April —
DR. STEAD: I am hoping we can actually get these timelines externalized in a way. This bucket of work is going to be fast-paced enough. If we can get back where we have one timeline for all of this, all of our lives will be a little simpler.
The Pop Health letter is the other thing that we will be trying to have in essentially final form by the March executive committee call. Okay. Then I am asking Bruce and Bob to take a look at a draft I did. I think Vickie will want to be involved in that, also, of taking our health data framework, convening document, which we have had for a good while now, and stripping out everything but the data framework and the data structure and the methods. So making it a white paper that we would just post to the website and cease convening around. But see if we could actually maybe get that done.
I have already flipped the draft. If we could get that done in time to be able to get it to the EPC before they lock the door because I think the data cube might be helpful to them. You will need to run the traps if we are talking – Rashida, you may need to pay attention to this – if we are talking about simply posting something to the NCVHS site as a white paper, is it okay, and this has been extensively dealt with over years by the Full Committee.
But what we are doing is just pulling out the white paper to try to get it in a form that is accessible by the community, in particular, so we can point the EPC to it. Can the Executive Committee approve that without us going to the Full Committee? It will be a thing to think about. You and Rebecca, I would think we could do that if what it is as a white paper with no recommendations, no themes. Then work plan update to the Executive Committee. I think that has got it.
MS. KLOSS: No, I think we have to rework the Beyond HIPAA description. We have to rework the classification vocabulary and design the panel.
MS. HINES: Are you going to rename that to classification vocabulary?
MS. KLOSS: That I don’t know.
DR. STEAD: Let us keep it terminology and vocabulary for the moment.
DR. MAYS: I just want to make sure either it is like a pick list or something. We have all these things that are left from the Workgroup that also need to be finished. We pulled together the draft of all the best practices material. I think there should be some plan as to whether we want to turn that into something.
DR. STEAD: Let me have the sidebar that I did not have time after I listened, and see if it computes for you. I think the idea of writing a letter that has what we have talked about is in the report.
DR. MAYS: This is separate from that.
DR. STEAD: I know it is. If we constrain ourselves to a letter, then we have the opportunity to create a couple of additional resources that are much more like a white paper. One of which would be the best practices. The other of which would be the Google Analytics stuff.
If we get those two down, I think those actually could become resources that we could put on the website. And they could become resources that we could hand off to the people responsible for HealthData.gov. But it would let us keep a compact letter. If we write a report, we will have the tendency to want to put those in the report. Then we will end up with a separate letter for recommendations. I think we will discover we have too many moving parts to hit our target.
I was thinking that might be the way to get the letter in a form. We could get it out in a way that is actionable over the next 6 to 18 months. Then these resources, which will be actionable for a much longer time, but they are actually actionable to the larger community. That is sort of the way we were parsing it, if that feels good to you.
DR. MAYS: Yes.
MS. HINES: We have APCD?
DR. STEAD: Yes, we do have APCD. I have forgotten about this.
MS. HINES: So ideally, we would have that ready for April, or at the latest June.
DR. STEAD: So you and Bob and Denise will need to collaborate around a timeline and tell us what is possible.
MS. HINES: So Bob, I propose the three of us set up a call and develop a timeline.
DR. STEAD: My own gut is do the work when it makes sense. I think trying to get that done in a two-hour call when we are approving these other three things might prove to be a bridge too far. If it is ready, we can see. If we can get it ready, the group has just not discussed. Well, not in a while, and not in a form that was driving toward common thinking about where we were going to go. I think that is going to be very hard for a two-hour call.
MS. HINES: Could we have time on that call to discuss a framework for describing? That way, then we have the input from all the committee members on our framework for the hearing themes. I think we can put together a letter, instead of having recommendations for what HHS should do around APCD, rather have a letter with recommendations of potential areas of focus that either HHS or the committee or both could do.
DR. STEAD: Believe it or not, it is three minutes for three, and it is about public comment time. Are we in a position that we can open the phone for public comment? That also is from the audience. I think a bird told me that there was a photographer that was going to want to make some public comments.
MR. FLOTOW: First of all, I just want to give greetings from the Board of Scientific Counselors at NCHS. Also congratulate you all on what a terrifically productive meeting this has been, having sat in and listened to some of this. One of the most exciting things I think I can report back to the board are things surrounding the NextGen vital statistics. I would love to be able to share some of the vision about that with leadership related to the board and so on.
As you finalize sort of the hearing document and so on, let me know, and I am looking at Rebecca, when I might be able to share that or some version down the road. I have to tell you, this is like the third meeting that you all have had before the BSC has met. The last time we met as the Board of Scientific Counselors was mid-September. The January meeting got cancelled, so our next one is May. I will have a lot to really report to them in person when we get to May. But it is your productivity versus sort of our schedule.
Now, just a couple of minutes, and I am going to make it as brief as I can, I am going to totally switch hats. I am going to put on hat as me as former director of the Illinois Center for Health Statistics, chief vital statistician and demographer sort of a sequential thing. This is really kind of directed towards the community health thing. There were other states besides my state, which is Illinois, that we went through a period in about 20 years ago in the 1990s about doing community health and letting local health departments do their own assessments.
I was having this sort of déjà vu feeling as I was listening to everything that you all were discussing about that. There were things that we had considered in the ‘90s, building the database, having it beyond health, environment, census and so on. Sort of not only could you browse through the database, look at your area, compare to other areas, there were canned reports that would give them on certain things. They wanted something on MCH. They could find out the measurements for that for their area versus the state versus the nation and so on.
Let me just give you the four lessons learned from that. Bill may call it scar tissue, but it is what it is. We found that even though, in some cases, we could provide a lot of county-level data, they often wanted the next lower level. I know it has been 20 years since, but I also hear you all sort of struggling on that. That was always feedback that we got from the local health departments was can you get it smaller, can you get it smaller.
Now having said that, what the local health departments did was they would get together in the community, people that were interested in determine what they wanted to work on for the next five years in terms of local health. We were trying to get them to have that database or evidence base, if you will. And we always had trouble finding if they had people that could interpret the data.
Now, we would help. We were on call. We had consultants who would do that. We built more tools into the database, so that they could graph a map and so on. We even did hand-holding level training in terms of going through the statistics and so on.
On the flip side, I was also someone who would review their evidence-based decisions. Now, if the community groups said, well, the evidence doesn’t really show this, but we are kind of interested in working on this, that was okay because the data was just one component of what they were after.
But often, their conclusions on what they came up, after all the training and so on, what they came up on the data, sometimes I had the opposite interpretation of the same data that they were looking at. So what I am saying is they just did not have that expertise. Even some of the bigger health departments did not necessarily do a good job out of the 80 or 85 or so local health departments in Illinois. There are usually like three or four that really did a nice job. There was quite a number of them that really, in terms of data, did not very well.
One other thing, as a goal for this program, was when they came in with what they wanted to work at on the community. We were going to try to match funding streams for it, to see if there was funding at the state or federal level that we could make these connections for. That was really difficult. There was a lot of communities that came up with some nice ideas and so on. Some of them, data-based, but that goal we couldn’t really fulfill in terms of trying to back that up, so that they could work more effectively on that. Now again, this was 20 years ago.
The other thing we talked about here in the last day or two was sustainability of this. We started out with five full-time folks. We had contractors not only for the database, but also to help with the training. After a couple of years, that went down to three. Then it was one and a half. Currently, last one I was at the department a few years ago, it was down to a half.
It isn’t that it got easier and easier to maintain, but it was sort of a cost thing. I just wanted to share some of those lessons learned as I sort of listened intently to what you had to share about going down that path. I hope some of the tools are a little bit better and so on. Really, finding the people that are able to interpret the data are just really key. Thank you again. I am sorry it took so much time.
DR. STEAD: Thank you very much. Other comments from people in the room?
MR. LANDEN: During the meeting today, I got an email that I thought was helpful enough that I want to make sure, for the purposes of transparency, it does get included in the record because I think there are some ideas that the workgroup that is working on the predictability roadmap can use.
This email came from Dr. Joe Burmell(phonetic). To paraphrase a lengthier thing, it is talking about the predictability roadmap. He points out that there are many other instances of processes that are in place that affect the US health care system both governmental, international, ISO and private sectors, some corporate programs that are very good.
One specifically he calls out is the National Quality Strategy, which is similar to ours. But it also incorporates some processes about priorities that our presentation did not address and which affected administrative and clinical code sets used for the quality reporting. So bottom line, as Dr. Burmell points out and essentially challenges us that two-dimensional thinking as we approach the predictability roadmap may not be sufficient for the purposes we are trying to achieve. Thank you.
DR. STEAD: Very good. Other input from the room? Input from the phone? Then I will say we are adjourned. I thank everybody. This has been an awesome meeting.
(Whereupon, the meeting adjourned at 3:08 p.m.)