[This Transcript is Unedited]
Radisson Hotel Reagan National Airport
2020 Jefferson Davis Highway
Washington, DC
FULL COMMITTEE MEETING – PLENARY SESSION
MR. REYNOLDS: Let's go ahead and get started, please. Good morning, I would like to call this meeting to order. This is the second of two days of meetings of the National Committee on Vital and Health Statistics. The National Committee is the main public advisory committee to HHS on national health information policy.
I'm Harry Reynolds from Blue Cross Blue Shield North Carolina, and Chair of the Committee. I want to welcome Committee members, HHS staff, and others here in person, and also welcoming those listening in on the internet. I would like to remind everyone to speak clearly and into the microphone.
Let's now have introductions around the table and then around the room. For those on the National Committee I would ask if you have any conflicts of interest related to any issues coming before us today would you please so publicly indicate during your introductions.
I have no conflicts of interest. Marjorie?
MS. GREENBERG: Good morning, I'm Marjorie Greenberg from the National Center for Vital and Health Statistics, CDC, and Executive Secretary to the Committee.
DR. WARREN: I'm Judy Warren from the University of Kansas School of Nursing, member of the Committee, no conflicts.
MR. BLAIR: I'm Jeff Blair from Lovelace Clinic Foundation, member of the Committee, no conflicts.
DR. SUAREZ: I'm Walter Suarez with the Institute for HIPAA/HIT Education and Research, a member of the Committee, and no conflicts.
DR. HORNBOOK: I'm Mark Hornbrook, Kaiser Permanente Northwest, member of the Committee, no conflicts.
DR. FERRER: Jorge Ferrer, VA, member of the Committee.
MS. MILAM: Sallie Milam, West Virginia Health Information Network, member of the Committee, no conflicts.
DR. MIDDLETON: Blackford Middleton, Partners Healthcare, Brigham and Women's Hospital, member of the Committee, no conflicts.
DR. LAND: Garland Land, National Association for Public Health Statistics and Information Systems, member of the Committee, no conflicts.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, member of the Committee, no conflicts.
DR. CARR: Justine Carr, Caritas Christi Healthcare, member of the Committee, and no conflicts.
DR. SCANLON: Bill Scanlon, Health Policy R&D, member of the Committee, no conflicts.
DR. STEINWACHS: Don Steinwachs, Johns Hopkins University, member of the Committee, no conflicts.
MR. SERACHE: Jim Serache(?), ASPE, no conflicts.
MR. QUINN: Matt Quinn, AHRQ, no conflicts.
MS. WILLIAMSON: Michelle Williamson, CDC NCHS.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, Committee staff.
MS. VIOLA: Allison Viola, American Health Information Management Association.
MS. THORNTON: Jeanette Thornton, America's Health Insurance Plans.
MS. WOOD: Laura Wood(?), Blue Cross Blue Shield Association.
MR. BIRNBAUM: Adam Birnbaum, Blue Cross Blue Shield Association.
DR. FRANCIS: Leslie Francis, University of Utah, member of the Committee, and no conflicts.
Agenda Item: Re-cap from Previous Day
MR. REYNOLDS: First thing, this is a kind of quick summary from yesterday. I really want to thank the Committee for your engagement yesterday, and I think all day yesterday during the discussions I really appreciate all the questioning we had of the speakers. I appreciate everybody's energy on the person centered. Remember, we just came up with that as a philosophy, so that doesn't mean it's going to fall out immediately as to what we really mean and how we're going to do it. So thanks for the energy to continue to deal with that.
If you look at the amount of change that has gone on with the administration, you listen to the significant activity that's gone on with CMS and ONC, and the things that are happening there, AHIC is changing as we knew it would, which is really a driver of a lot of the stuff that's going on in the industry.
So as we take this journey on person centered health and as we try to filter in my opinion against kind of the 21st Century vision, and I think that's one of the things we can do and I really like, and Blackford kind of kicked it off but now it's we'll all say it was our idea. The idea of really using value as a differentiator when we think of everything, and what is it that we're doing, and what is the standard and what is the data, and what is the thing, you know, really trying to get out in front and think about the value, because in the end adoption is pretty easy if you have value. If our goal is to really drive some of these things to happen and really become realities, then I think the value is a key thing.
Then obviously with all these things I just mentioned you have data and privacy and population health and standards as the underpinnings of how you actually get a lot of that done. I know from the breakouts I was in yesterday afternoon and this morning, a lot of energy, a lot of trying to build exactly what we want to do and how we want to make a difference. And I wasn't in all of them, but I think I'm starting to see some things come together where we can make a difference. That is kind of the summary of what I saw yesterday.
As we look at today, has everybody received a copy of the charges; is that correct? I know some of them were updated, so I'm not sure.
Marjorie, I guess I'd ask you, one of the things I think we ought to do I know, for example, in talking to Don a little bit ago he's got a lot of writing on his. So we talk about reviewing and improving the charges, I think it might be a good idea to send them back out to everyone and get their comments. That's why I need to kind of get some guidance from you.
MS. GREENBERG: What I would suggest is I'm hoping that each of the subcommittees looked at their charge in its current latest version, the package that Debbie put together. I think the next step is Debbie to probably get with the co-chairs and then see what she has and what they have, and try to come up with a common format for each one of them, and I think that Susan Kanaan is going to work with her on that.
Then after we get those all prepared, then each subcommittee can send it out to their whole membership and see if once we get them all in a common format. It seems to me at least when I was in the populations, but generally any turn is a little different, but sort of the scope and there should be something on the scope, and something on the methods, and those will be similar across all of them, then maybe a paragraph on the FY09, and priorities. A page probably for each one of them. Then we will also make sure we are updating the work plan or making it useful to people, that we're actually tracking specific activities. I think that those are the next steps.
Debbie?
MS. JACKSON: That's fine.
MS. GREENBERG: Does that sound reasonable? Okay. Then in February we should definitely spin off on – the executive subcommittee before then and all that.
MR. REYNOLDS: What I would like to cover next is, Justine, can you give us a few minutes on the de-identified meeting that you went to with the Center for Democracy and Technology. Then I would like maybe the subcommittee chairs since one thing on the agenda, person centered health, I think we did that yesterday, at least the base discussion. But what I would like is just a quick summary from each of the co-chairs on after we had our meeting yesterday and then you had your breakouts, just kind of give a quick update to the committee as to where you think you're going, and you may have some gaps, but just what came out of your discussions based on what we did so you could have that.
So, Justine, thank you.
DR. CARR: Thank you, Harry.
As you know, was it two years ago that we did the uses of data report, and a key area of focus that came out of that was de-identified data and whether we should comment on that. I think the more testimony that we heard the more we realized that it's a very big topic and that we were not prepared to comment in detail within the report, and the plan was to hold hearings.
But in the meantime the Center for Democracy and Technology pulled together an illustrious group for a one day think tank, and helped put together the discussion points that are included in your document. I think that they had all the right people, it was a tremendous exploration of the topic, and I think that the summary puts it together very well.
So behind Tab 3, and I think what I am going to do is just highlight a couple of things. They have broken into two parts; one is eight observations that seemed to be representative samples, and then several areas where more information and exploration is needed. So point number one is that it may be easier now to re-identify anonymous data than it was in the year 2000, that's due to a couple of things. One is that the medical information that's collected now is more detailed, and can create statistically unusual entries. Also, we're simply collecting more data. And thirdly, the cost of data storage is going down. So there is a lot of data, more detail, easily stored. That is issue number one, it may be easier.
Issue number two, patient data is very useful. We know that it's useful for appropriateness of care; it's also very useful to understand what therapy works, what evidence supports decision making. It is very useful for bio-surveillance. The way they put it is patients want privacy while physicians need data. So I think that the observation that the data is useful is important.
Moving on to number 3, data security standards are a proxy for trust. The point was that the things that are in place, such as HIPAA, the FDA, and extending security around data, represent trust, and data privacy must be taken extremely seriously in order to maintain trust with patients, physicians, advocates and policymakers and others. Unless patients trust providers to protect the data, quality issues will proliferate because patients will hold back information and the datasets will be compromised.
Point number 4 was patient access poses extra challenges. So committing the re-identification of data in response to patient requests for their own data is more difficult than simply keeping the data de-identified and secure. A persistent issue is have the de-identified data in a way that makes re-identification possible when needed. I'm not sure I understand this one exactly. I'm just reading what they say, so moving on. I think maybe the point that was made there was that data is when it's de-identified the very act of asking that it de-identified – the other points are good.
Number 5. Even if not re-identified, de-identified data still carries privacy issues. This gets back to the topic that we discussed and the uses of data, that if the data is aggregated and demonstrates an attribute of particular group of people, that that can be harmful to that group.
6. Re-identification is often easy, but there are solutions. So there was discussion of models where the data utility is maintained and privacy is preserved. So, again, that was kind of a technical discussion, but suffice it to say that as the challenges increase the technologic solutions are increasing as well.
Then another, number 7, they talked about a possible model for privacy. There they are saying it's not all or nothing, but it's how much of a risk is incurred by having your data available. No risks are greater being in a cohort than being outside a cohort, differential privacy has occurred.
Number 8 was de-identification must go beyond technical standards, must involve trusted personnel, sound policies, and they highlighted one particular, the commercial program where there is a multi-step process that layers on protections. So different levels of security or encryption keys at different levels, so no one person holds the final, the single key to get to the data.
Then they identified issues for further exploration. One was create stronger requirements for data use agreements. I think this is something that we definitely recommended in our report. The point being technical safeguards are important, but they're not enough. Require that data remain in its least identifiable form consistent with the use for which the data was accessed into – and, again, these are areas of discussion. Another one was integrate a system of patient notice which the patient is advised on what data is collected for what purpose. That is for future explanation.
Another area is fashion potent remedies for violations of data use agreements. I think this is a topic we've discussed here as well. These may include a right of private action, there should be strict penalties for re-identification of de-identified data and robust enforcement.
Other areas for further exploration include store care data and a limited access database or network. Next would be establish certified data stewardship entities, another topic we have discussed here. Then, finally, two more, develop different levels of anonymization. So when we did the uses of data we talked about that there would be only one term that we would include, which is HIPAA, de-identified. They made the point that different levels of anonymization may be important to specify, and that needs further use. Then a final area for further exploration was to brainstorm specific harms and develop targeted ways to minimize them.
MR. REYNOLDS: Jeff?
MR. BLAIR: Could you just help us understand what this organization is, because I'm not familiar with it, is it Centers for Democracy and Technology? What is this organization?
MS. BERNSTEIN: Do you want me to help? I'm an interloper at this meeting. Maya Bernstein from HHS. The Center for Democracy and Technology is a very well-respected public interest group that works on issues of, as it says, policy technology, very intent on emerging technology. In the past they have not worked on healthcare so much, but they recently hired Deven McGraw, who used to be one of the co-chairs of the AHIC subcommittee that dealt with privacy, as an employee who is now working on healthcare issues and technology.
So this particular meeting, which I also attended, and actually Justine gave a really nice presentation there, was particular to – there were people there who were not working specifically on health care but it was coming out of the healthcare context to talk about de-identification, and CDC's reputation as a public interest group and advocacy group, is a place where people with very disparate views can sit at a table together. They are considered fair players and honest brokers, and that sort of thing, so they can get the CIA and those kinds of folks to sit down with the advocacy groups, and it's pretty unusual.
MR. BLAIR: Could you just clarify one thing? Because you say it has diversity, and then you say it's an advocacy group.
MS. BERNSTEIN: Well, what I'm saying is not that it has diversity, but that it has the capability and the kind of respect that allows them to bring people to the table. It's a convening organization often, although it is an advocacy group, it is a trusted advocacy group, and they're considered fair.
MR. BLAIR: What is it that they advocate?
DR. WARREN: A public interest group is more accurate. But their byline is keeping the internet open and free. There you go.
MR. BLAIR: Okay, thank you.
DR. CARR: Also do you want me to just - Peter Swire, JD, Professor of Law at the Moritz College of Law at the Ohio State University and Senior Fellow at the Center for American Progress was the moderator, and he did a superb job.
Then there were three sessions. The first one was facts and methodologies, and there it was Latanya Sweeney, another person, Cynthia Dwork from Microsoft, and Harvey Ashman, VP of External Affairs, Associate General Counsel for IMS Health. That was a pretty smart group of people, I'll say that.
Then I had the pleasure of following them after lunch, so you can just imagine.
MS. BERNSTEIN: A lot of it was highly technical and mathematical.
DR. CARR: It was really something.
MR. BLAIR: Is this the kind of a group that we could continue to work with, or how do we work with them or engage with them?
DR. CARR: I think that we can, well, let me let Harry speak to that. Let me just say who else was there, so Linda Goodwin was a nursing program, and Duke talked about the everyday experience, restrictions, what it meant, the implications for their neonatal populations. We had the privacy officer from Eli Lilly, I think he spoke about these tiered security things. Bill Braithwaite, Mark Rothstein, and then Ken Goodman, Co-Director of Ethics Programs at the University of Miami.
So maybe Harry could talk about the next stuff.
MR. REYNOLDS: Jeff, I think as you noticed, well you didn't, but when we explained it yesterday, as we drew the picture of the person centered health and other entities that we can get information from as input, this would be a group, as input. Again, a lot of people who come speak to us we listen to what they have to say, and this would be a good group where, I think as Justine said, pulls together a broad cross-section, and that's the benefit of it that I see, is that they pull a large cross-section of people together, and not necessarily just people in healthcare, that are thinking about the subject of de-identified information, not just protecting any particular portion of the discussion about it. I think that's the kind of thing that as we try to drive what we're moving towards, especially when some of it is not perfectly clear, we would use that as a good group to keep an eye on.
DR. CARR: Let me just read you their mission. The Center for Democracy and Technology is a 501c3 nonprofit public policy organization dedicated to promoting the democratic potential of today's open decentralized global internet. Our mission is to conceptualize, develop, and implement public policies to preserve and enhance free expression, privacy, open access, and other democratic values in the new and increasingly integrated communication medium.
MR. REYNOLDS: And de-identified jumps right in the middle of open and privacy. Leslie?
DR. FRANCIS: I wanted to follow-up on a question for Justine and for Maya, who were both at that meeting. Our report that Justine referred to a couple of years ago actually says that we were going to follow-up on privacy and security issues with de-identified data, and that has been a question about whether there was a role for particularly the privacy and security subcommittee. It is something we have discussed, and currently put a little bit on the back burner because we weren't sure what value we had to add.
My question for the two of you is did you see at that meeting reasons for us to, as we're planning over the next year, move that away from the back burner?
DR. CARR: I'd like to make a recommendation that Deven McGraw or one of the folks come and do a presentation. Can I have a discussion afterwards and get a clear idea of what their next steps are and what they're going to do, and then that will help inform what we decide.
DR. FRANCIS: Could we do that at the February Full Committee meeting?
DR. CARR: They definitely want to come to this group.
MS. BERNSTEIN: They were actually here listening this morning at the privacy session.
MR. REYNOLDS: Yes, because what we may want to consider doing is walking along beside them, and learn from them and/or be part of what they're doing, not necessarily have to do it ourselves.
John has a question now. You have a follow-up? Go ahead.
MR. HOUSTON: Is it within the purview of NCVHS to simply acknowledge and support through a letter something that somebody else is doing in short form, is that something that we do?
MS. GREENBERG: After she's been here.
MR. REYNOLDS: There's been discussion, yes.
MR. HOUSTON: Yes, let's say there's been discussion, this is what we've done, we would like for you to support what we're doing, and assuming we all agree that what's being done is appropriate, I mean I've never seen a letter like that come out where NCVHS stands behind somebody else's work or acknowledges that it's -
MS. GREENBERG: If they make some recommendations that you think would be useful to send forward, or otherwise, you could even just send them a letter, I mean it doesn't even have to involve the department I suppose, and then CC someone in the department. I mean I think you have options. So what I would suggest is that we have her come here and see where they are on this.
MR. HOUSTON: The reason I bring this up is also in the context what we heard yesterday about AHIC 2, where there was some discussion about the fact that there is going to be a governance model that is going to be delivered by next summer. And speaking offline with the woman that presented yesterday, I asked whether there was interest in NCVHS being involved in some type of review of that. I'm just wondering as a result of something like that is we would make the statement, yes we believe this governance model is appropriate, or we would come out in some type of a public way either supporting her or entering.
MS. GREENBERG: You might want to have a hearing about for that matter before you made a statement. Yes, I think you have a lot of different options. We're not talking about advocacy here, we're talking about review of documents and activities and their policy implications, and advising the National Committee on Health Information policies. I think if it touches your space, you know, you will want to think about what's the most effective way to do it obviously.
MR. REYNOLDS: Where I would come down on that is we have advocated a lot of positions because we do most of our work based on testimony and our discussion. So I would consider anybody that comes in here and we listen to what they have to say and we feel that's something we can build a comment off of, it's no different than any other. I would hesitate personally to just say we went to a meeting or we heard from somebody and let's go do that.
MR. HOUSTON: I guess we did that in the ICD-10 letter that was probably five or six years ago, it was basically at some point we said we support this.
MR. REYNOLDS: In other words, back to this point, if there are key pieces, we wrote a letter, the privacy letter and a lot of stuff we did had to do with de-identified. This is but one body of work based on that that has happened. And I'll go back to the earlier thinking again. We heard the proxy of trust in the subcommittee. We heard Justine just say it, we heard it from Sallie in the discussion we just had, she called it the trust proxy. I think you switched the words. Then we talked about value. If you get value then you're going to get trust, and then how is that actually executed, and some of this discussion on de-identified and anonymized and other things.
So we need to continue to look at the layers. This would be one way if you have the right value and you have the right trust from people, then they may be able and willing to have more of their data anonymized or this or that, or more visible, if they already know that it's okay. I mean there's the whole thing about people buying stuff over the internet. Once they bought into that then they got the trust proxy, they're buying stuff over the internet.
So I think those are a continual thinking. So this is just one more in our chart, this is just one more environment, just like we talked about what did we hear from CCHIT on privacy today and some other stuff, they're just continuing, but then I think the opinion should be ours as to what we want to say or recommend, not necessarily just rubber stamping a group and say this is where we're going.
Maya, you had a question.
MS. BERNSTEIN: I was going to try to answer a little bit Leslie's question, and may have a little bit different take than just me. I found I was intimidated by a lot of what I heard at this meeting, particularly the first panel in which they talked about the latest research on exactly how to de-identify stuff, and what the advances are, and Justine described them. And I thought if we were going to take – I basically didn't feel like I was competent to support the Privacy and Security Subcommittee if we were to take that on, because the nature of the kind of work that is being done is incredibly technical.
So I was very happy to let them go on with it, and occasionally report to us, and that would be lovely. But I also thought the second panel on which Justine was on talked about the context where it comes up, where it's useful and that sort of thing, that might be useful to hear, too.
Then I thought the panel that Mark was on in particular in the policy implications was more in line with the kind of work that we do, partly because he talks about where de-identification is not necessarily useful. So he said, look there are times when even if you de-identify data there could be research subjects that would object to participating in that kind of research even if their information were de-identified. He gave an example, for example, stem cell research, people have moral and ethical objections.
There were other ethicists on the panel that talked about different angles. So I just felt like more competent in that area because it's not so technical and it's the kind of thing that this Committee and the diversity of this Committee has had in the past, really active and very engaged conversations about those kinds of issues and where we're going to come down, and so forth. I think we should hear what they're doing, but I would feel a little -
MR. REYNOLDS: Again, it's an input.
MS. BERNSTEIN: Yes.
MR. REYNOLDS: It's everything we do. And, Marjorie?
MS. GREENBERG: As Maya suggested, obviously we can't hold hearings on everything, we have broad expertise, but we have more in some areas than in others. So I think it's really good to liaise with people who are in the same space and have some of the same policy issues, and learn from what they do. We don't have to repeat what everybody else does. Whether it's useful to do like some kind of a letter is I'm thinking of a number of years ago there was a commission on quality, and the Committee reviewed that report and then wrote a letter endorsing certain aspects of the report and then saying we're planning to do these things.
That is kind of one approach, where you might see in everything that they looked at, or everything that they're reporting, or whatever, that there is some area that you want to take forward, then that has some benefit to put it in the context of work that has already been done. But I don't think you don't have to take the testimony yourself, it's like doing a literature review or something, to be influenced by it, but I would agree with Harry that you don't want to just go around -
MR. REYNOLDS: I'm going to get Don and Mark, and then I want to move on to the next subject. In other words, this was – thank you, Justine, for going, it is a good environment for us to have been seen, it's good information, continued information, growing information, and anything we want to do on the subject then we would decide and then they would be an input. Don?
DR. STEINWACHS: Mark, I was going to twist a little bit, did you want to respond to it?
MR. REYNOLDS: You on the same thing?
DR. HORNBOOK: Yes, the same thing. I just wanted to remind us that de-identification depends on where you sit. I perceive in the CTSA world, for example, very strong chasm between those who have the data, the HIPAA entities, and those who don't, namely professors who don't happen to be inside HIPAA entities. It shades their judgment in the sense of what is the adequate business risk for releasing data for research purposes.
I just would urge us to remember that there are people out there who are inside HIPAA entities who give a lot of thought to releasing data, protecting privacy, and they have done scientific analyses of de-identification processes and find them lacking. It would be nice to hear from people on that side of the fence who can talk at the same level as the informaticians who are trying to create de-identification engines that have yet to do it with a level of performance that people like Kaiser would accept.
DR. STEINWACHS: That was a great introduction for what I wanted to talk about. It seemed to me in the 20th Century we thought about health statistics data being shared with researchers and others mainly through public use tapes. There is now a sense that public use tapes may be on the death trail, indeed they are just too identifiable. And the hearings we had on data linkages it was very clear that once you linked two datasets the identifiability went up tremendously. So it would be wonderful for us to be able to grapple with this bridging issue, because I think the sense out there is that increasingly what researchers and people who want to use the data and make something out of it find it less and less accessible, because it's now becoming more and more identifiable and we aren't answering the question of how could we make something that could be more generally used.
MR. REYNOLDS: Let's start. Is this really important?
DR. FRANCIS: I think we need to frame how we should have Deven here, but we should ask the questions or the discussion to focus on the public policy questions of the kind that Don raised or Maya was discussing, rather than the technical issue.
MR. REYNOLDS: What I would like to do is get input from you and John and Maya and Justine, if you want to access her, since you guys kind of own the subject and it kinds of fits in your environment. If you would like somebody to come into the Full Committee then I would like you to be the one framing it, because that kind of is your subject rather than giving it someone.
DR. FRANCIS: We'll send you a framework.
MR. REYNOLDS: Yes, I think that would be the right way to do it, and I'd like to see us do that ongoing, so that we keep the right players with the ball and then they bring it to the Full Committee.
So let's kind of get, that's what we'll plan to do, but we'll hear exactly how they'd like to frame the discussion, because it is a continuous as we take up person centered, and that's the other filter I want to keep using, the person centered, everything we're doing, what the status of their data is and what it looks like, is an underpinning to trust, it's not going to go away.
With that, let's go just to a real brief update, not a complete discussion of the full breakout session, kind of where you're going and what you're doing.
So, Don, since you're warmed up and have the mike, why don't you go ahead with populations, especially since you're going to be meeting this afternoon.
Agenda Item: Committee Discussion
DR. STEINWACHS: So I have to work this afternoon, that's what Harry's telling me.
Just to recap quickly, because what I'm going to say you mainly heard and came out of our discussions. Certainly one of our areas is an interest in how do we look at the availability of health data that will support the analyses that will be needed as people look at health reform and health reform options. There are really two pieces of that one the hearing this afternoon pursues, and that is looking at data that helps address questions of insurance coverage and access, and certainly underlying that is also the dynamic changes and interesting where the trends are going.
But pushing that further than it has before in terms of what's the availability of state level information to make estimates on uninsurance, what do we have that talks about the duration of the period of time that people are uninsured, not just the cross-sectional shot of uninsurance. There is also the question of under-insurance, which is a sense increasing in America where we're pushing people to pick up more and more of their healthcare.
That we expect to be followed by maybe one or two more hearings, one definitely on modeling, how the data actually could be used in policy modeling and both within the government and outside. This afternoon's is really on inside the government.
The second is the medical home, and we had some discussions about work that Matt Quinn and Larry Green has also been involved in, to try and think about how we could use specific use cases to explore a data model that needs to underpin strong primary care, and using the medical home as sort of a policy relevant way to talk about it, but it is really strengthening primary care preventive services, ongoing disease management, and so on.
We quickly revisited vital statistics, and as you know, this is really trying to address the very deep concerns we have about access to vital statistics data and the quality. We talked a little bit about the coordination between that and what's being done by the BSC, the Board of Scientific Counselors, to NCVHS, and the timing of that. so certainly we see ourselves as pursuing that further.
Marjorie and Debbie gave an update to the subcommittee on revisiting the 21st Century vision, which you've heard already.
And then we just talked very briefly about follow-up to the CTSA discussion here, and that Mark and I and I think some others who may be interested were going to take a crack at sort of doing a one-page sort of thing, and then circulating it to the subcommittee and then circulating it more broadly, to begin to think about that and see how we could take some steps to initiate discussions and understandings about what the potential of building some bridges between what NIH is doing and what we are interested in in health statistics for the country.
And I think that pretty much wraps it up.
MR. REYNOLDS: Justine and Paul.
DR. TANG: We sort of framed our hearing concept from yesterday, and wanted to get back in focus on the health data that is used to assess and influence individual health that can be reused for population reporting. Then after we enumerate those data, we would need to understand novel and traditional methods of acquiring, standardizing and protecting that data, and move on to the policies that would surround those issues.
One of the areas where we sort of probed in more detail is what are the novel uses and reuses of data. One way to approach selecting this new data is to go recollect it in a sense, so have this separate way of gathering data from individuals. Alternatively, you take advantage of data that has already been acquired, and various folks pointed out well every time you use your credit card, every time you buy groceries, there is a lot of data that reveals some of your personal behavior that impacts your health. But that can be scary in many ways.
So is there a way to reuse that data in a safe way, a protected safe way, that can gather better data about an individual, that individual can use it to improve their health and, again, in a safe way we can use that data to publicly report about the health status of a population.
There is sort of two different ways. Go get data purposed only for this use, or repurpose use that has already been fathered without additional effort, but find a way to protect it and standardize it. That may be one of the things where we could use some Committee input, just input, not a decision or anything.
DR. STEINWACHS: Do we need another diagram?
MR. REYNOLDS: Do not get up and go up front.
DR. CARR: If you read our charter, in addition to monitoring health data needs, section E is “Identify strategies and opportunities for evolution from single purpose narrowly focused categorical health data collection strategies, to more multipurpose integrated shared data collection strategies.” But I think that it's, again, sort of focusing on this novel. I mean we think about it as sort of asking how old are you and how is your health, but in fact, we're answering that question in many different ways with our credit cards and so on.
As Paul pointed out, it's sort of like the 1984 when it was 1970 and they wrote 1984 it was very scary, and yet today if you read 1984 it's the reality. So I think that what we're thinking of is –
MS. GREENBERG: Not quite.
DR. CARR: No, it's not the reality.
MR. REYNOLDS: For the minutes, that would be one opinion.
DR. CARR: But the world that we're in today, there is all this data, people are getting it, and people are using it. And we see NCVHS as a platform where this reality that's going on gets surfaced so that it can then be addressed by the Committee for privacy, security, confidentiality. So, scary.
MR. REYNOLDS: John and Leslie.
DR. FRANCIS: Well, with the scary question, are you planning hearings, just as a question?
DR. CARR: Yes.
DR. FRANCIS: And when?
DR. CARR: We're thinking about February. We have to coordinate.
DR. FRANCIS: We're intending a hearing the day before the Full Committee meeting in February that will focus on the questions about the evolving role of PHRs in the space of personal health. We were looking at four panels, one on what's happening with respect to data flows in and out of PHRs and the privacy practices of PHRs. The second panel would be devoted to the legal, regulatory, and certification setups ranging from FTC to CCHIT. The third panel would be on what - ONC was also part of that second panel mix – the third panel would be consumer concerns, what concerns have been raised, what is on the table for us and so on in this area. The fourth panel would be some of the various demonstration projects that are out there, including the Medicare demonstration.
Harry reminded us very forcefully of some of the themes that had been ongoing in our work. One of those is the sensitive information, and one of the reasons for having the South Carolina Medicare demonstration project actually is to follow-up on what is and possible could be done with respect to protecting sensitive information in this area.
So I hope that's a summary. On the regulatory front, John may want to talk about the governance point in some of this.
MR. HOUSTON: Actually those are nice lumps.
DR. TANG: Just as a point of process, how are we going to decide how many days and which days are allotted to which hearings?
MR. HOUSTON: We were hoping to accomplish -
MR. REYNOLDS: Out of today we need to find out everybody with the way the schedule was everybody is now jumping in front and behind the full meeting, so you just need to submit please what you're thinking, and then let us work that out with Marjorie and figure out, we may have everybody on the same days.
MR. HOUSTON: Can I just expand on what Leslie has said? We are sort of looking also from a PHR perspective. A lot of our discussions are centered around where PHRs naturally go over a five to ten year period. Because I think where they are today and the they tether today, and who is sponsoring them, they may dramatically change as we start to see not just integration with an EHR, but rather some of the technologies related to chronic care and managing individuals in their homes, and other ways in which consumers may get engaged.
Also to a point that Harry made today, I think in large measure PHRs are the primary use seems to be in the chronic care side, but we have another class of consumers that are going to probably get more engaged with uses of PHRs over time. So the uses of PHRs may also dramatically change over a five to ten year period.
So we are really trying to set a privacy roadmap that takes into consideration not just what PHRs represent today, but where they're going to be and how they are going to integrate. And some of the issues they potentially associate with is they start to merge with the EHR space. At what point maybe to we see HIPAA and state laws all of a sudden start to cover PHRs and maybe potentially in a way it was never intended.
MR. REYNOLDS: One other thing that was discussed was to have possibly a preliminary conference call using value. Remember the discussion about some of the work that Blackford and others -
MR. HOUSTON: We haven't told Blackford yet, but we were hoping we can get him on the phone.
MR. REYNOLDS: Yes, so that basically if we're trying to do a lot of this stuff based on value, what came out of the group was to learn more about value of some of these things ahead of time, then when you bring people in to discuss it with them, and this may be something that since we can't have as many hearings it may be the kinds of things that we learn about value as people see it. Then when you have people come in you have a whole different set of questions you're asking them than you would have had without that discussion.
MS. GREENBERG: Yesterday we talked about quality and privacy kind of partnering. Now it seems like they have moved apart, which may be okay, but what was discussed this morning in the Quality Subcommittee was that from their perspective looking at the different types of data sources, they're interested in novel sources or novel uses of existing sources, whatever, they thought it was they needed to kind of define that space before looking at the privacy and security issues. But it looks like you're moving ahead to look at the privacy and security issues related to PHRs, novel or not.
DR. FRANCIS: Right.
MS. GREENBERG: I guess that's okay, but I sort of am wondering.
MR. REYNOLDS: Well, we need to normalize what everybody did their thing, now you and I and some others need to sit down and normalize what everybody's asking for. We may need to talk to some of the chairs about it. Go ahead, Justine.
DR. CARR: We're back to the question that came up yesterday. Do we have one big meeting with everybody all the time, or do we divide up the work and come back together. This is not incompatible with getting the work done. The real concern was that if we're asking the same people to come and testify twice that's a problem.
But we have stepped back from privacy, we want to just sort of say what is out there. Then I think that some of the privacy issues are going to be different from the conventional PHR, could be very different.
MR. HOUSTON: What I think we probably need to do, at least the chairs need to do, is as we're putting together these different testimonies is make sure the other chairs know who we're inviting. And I think it's our obligation to circulate -
MR. REYNOLDS: Let's go back to the picture from yesterday. Remember I drew the double-sided arrows and the circle. We are embarking on all subjects in my opinion that no longer just fit nicely in an area. There's too much going on in the industry, there's too much overlap. The chairs absolutely have to be willing to communicate with each other continually, and then we need to normalize when the hearings can or can't occur. So that would be key.
MR. HOUSTON: One other point back to privacy and security first just for a second, as I said, we talked a little bit about it with regards to governance, this is something we're going to monitor. We recognize that AHIC 2 is going to be putting together recommendations in that respect. We think probably over the next six months we're going to at least at some level engage, it may not be real substantively, but it's on our radar screen for the next six months or so, to make sure if there are governance recommendations then we are involved in at least hoping to be involved in vetting them.
MR. REYNOLDS: Judy and Jeff.
MR. BLAIR: We had a brief but efficient meeting today.
MS. GREENBERG: Sometimes that's good.
MR. BLAIR: Number one, with respect to the purpose and the questions we would ask testifiers in order to get their opinion on reflecting on that purpose. There are modifications to a number of those questions, plus the likelihood that we'll have different sets of questions for different type of testifiers, those that might be users versus those that may be part of the development or coordination process. So that was one thing we achieved. So it's probably going to be ongoing. This was really our first meeting to get together and start to chew a little bit on some of the preliminary frameworks that we had.
The main thing, we put aside a number of the elements to work on until a later time, and we concentrated on whether we had the right topics, panels, and testifiers for the four days of hearings that we were planning. There is a lot of new ideas coming out there; they will be sent to us in terms of how we might modify those panels. Kind of as a silhouette behind those panels is, at this stage I'll call it an assumption just in case it changes, that the process we go through would be to send out by email to all of the possible candidates to testify, which will probably be many times larger than those who will testify in person, requests for written testimony. From those we will select those that are mainstream, and those that are divergent, so that we can get a divergence of opinions with a limited number of testifiers.
We are then thinking that the first day of testimony would be tacked on to the very first day of our February Full Committee meeting. On that day we would be getting perspectives from government oversight and standards coordination activities, and in the afternoon we would get a perspective from the industry and the major industry forums and standards coordination and selection entities.
Then the second and third day of testimony would probably be held sometime together either in the April or May timeframe, and the theme for the second day would pretty much be SDO feedback from the standard development organizations and the terminology development organizations. Now, that may change as we begin to modify that a little bit.
The third day we would go into other feedback, probably more from users. Then the fourth day would be those that we would fill in from other perspectives that need to be covered. That fourth day would be tacked onto, it would be the day just before the June Full Committee meeting.
So that, for the most part, is what we covered. We deferred reviewing the draft letter to requesting written testimony. We also asked for folks to add to our preliminary list of folks that would ask for written testimony. We had about 15 or 18 groups, entities within the standard development organizations, there are more than that, professional associations also about 15, there are more than that, also among vendors about 15, there are more than that. So we are soliciting broader lists for written testimony.
Judy, anything else?
DR. WARREN: I just want to say, the other thing that came up in our discussion as we were looking at what questions would we ask, is we will probably be asking a question about value, again following up on Blackford's suggestion. Then Harry also wanted something about sustainability, because what we are hearing from people is that there are a lot of standards out there to adopt, some are more sustainable than others, so looking at sustainability. Then Walter wanted us to also focus in on once these standards are in place who is testing them and who is adopting them, and where are the problems in those two processes.
So we really are trying to get a handle on the whole process of standards development, implementation, and adoption in the country, because we are hearing lots of problems with that process.
MR. REYNOLDS: Again, what I'd like each of the Co-Chairs to submit immediately minimally to Marjorie and I, what you planned as far as when you're going do things, and we need to normalize that, we'll take care of it.
MS. GREENBERG: We have everybody wanting to meet at the same time. I should say that first of all we will have two Full Subcommittee full day meetings, the day before the Full Committee meeting is not necessarily doable or desirable. And we have certain resources like that.
MR. REYNOLDS: We understand, we understand.
MS. GREENBERG: And it's not pretty either. So I didn't – although the stand alone meetings are more expensive from the point of view of travel, we can't bunch up everything with these meetings.
MR. REYNOLDS: Right. Please just submit them to us.
MS. GREENBERG: But we have to do this soon.
MR. REYNOLDS: Right away.
MS. GREENBERG: Because if people are going to have a hearing even in February you have to start inviting people.
MR. REYNOLDS: By the end of the week we need what your plan is, please.
DR. MIDDLETON: Harry, just a question or a comment. I guess if this is sort of a critical juncture, and I'm not sure if it is or isn't, I wonder if there should be a request made or put before the Committee that we actually consider three days in a row. And if we were to consider three days in a row that we do it in a nice place.
MS. GREENBERG: What did you recommend?
DR. MIDDLETON: Nothing further to add.
MR. REYNOLDS: If we are going to need the Full Committee for some of this that we meet three days in a row, rather than just two days like we're doing now.
DR. MIDDLETON: We ask the Committee members if they would be willing to do that.
MR. REYNOLDS: Agreed, right. I want to do some time checks. We do need to adjourn at 11:30 to make sure we get everybody going there. How many people need a ride to the airport at 11:30? I know, John, you need a ride at 11:30. Yes, I do, too. So that's two, Debbie.
MS. GREENBERG: And how many people need a ride to the Humphrey Building at 11:30? One, two, three, I can take four people.
MR. REYNOLDS: Let me turn it over to Bill Scanlon now, please.
DR. WIILLIAM SCANLON: About the BSC, we started yesterday, about their interest in vital budget issues surrounding already that's being planned in terms of the limitations on the data. This is probably the longest standing budget issue that NCHS has had. In the past in some respects they have been using IOUs to keep things going, and now they're saying they can't pay off the past IOUs, so we are going to have to stop issuing them to keep things going.
In addition to that, there are a number of other areas being impacted by the limited funds. The National (?) Survey as of October the 1st the sample has been cut in half for this year. The National Nursing Home Survey is being postponed until at least 2011; it was last done in 2004. The Assisted Living Survey, which would be a new survey, which is critical for long term care because we now have about a million beds in assisted living, versus 1.7 million in nursing homes. That was postponed last year, I'm not sure if it's back on track yet. Either the National Hospital Discharge Survey is not going to be done this year. This is the first break since 1965 in terms of that data source.
If the continued resolution goes to the end of the year and there is uncertainty about next year, it is possible that health survey will be curtailed even more, and the NHANES sample could be reduced as well, the over-sampling of minority populations could be reduced or eliminated. And the home health and hospice survey could be postponed.
In the context of the BSC, in addition to creating the subgroup or the task force on vital statistics, they have created a second task force on the NCHS vision for the future, which partly is driven by the budgetary situation, but also driven by the idea of is there a new way to gather the necessary and important information, and what will be NCHS's role in this. It very much parallels our sense of a vision for health statistics.
They are thinking innovatively as well, how can you use the administrative data to substitute for some of the things that are currently collected through surveys. A good example might be the National Nursing Home Survey where historically there was information collected about the health status of residents, but yet every nursing home resident has a comprehensive assessment, and is collected and transmitted to a central source and can be used to provide that information. So there are alternative ways to think about getting information besides surveys.
What this task force is doing is trying to look at what are the data needs from a policy perspective, what have been the traditional, who have been the traditional users of NCHS data, and what are their needs and how can those be satisfied beyond the policy community. Then what the most efficient way to try to collect information to fulfill those needs, then what would be the budgetary implications of trying to do that in an efficient way.
Beyond the budget which is a big issue and overhangs a lot of things, they are continuing their program reviews of the various surveys at NCHS. They just completed the health interview survey review. They are starting now on the long term care surveys. That will be done before the spring.
In addition, they have had two or three discussions, two or three meetings, there have been discussions of a question of informed consent related to the NHANES data. NHANES has collected information or has information about DNA, and if there is an issue of how can potential uses of data be allowed beyond who was in the original informed consent list. There has been more than one discussion on this without reaching a conclusion. There are issues of do you go back and ask people to expand on the consent, what are the implications of that. If you change the informed consent at the outset for new individuals being surveyed, how might that affect participation in NHANES overall. There is a lot of uncertainty about exactly how to approach - what the implications of approaching it in different ways might be. This is something where they're going to be revisiting some of that again.
That's where they are at the moment.
DR. STEINWACHS: Do you distribute medications for depression?
[Laughter]
MR. REYNOLDS: Some of the budget cuts that everybody faces at home look a little different, after listening to that.
Okay, Debbie? Are you going to give us an update on -
MS. GREENBERG: Do we have anything more to say on this? No, this is the 60th Anniversary. Oh, I'm sorry, something else.
MR. REYNOLDS: First meeting of CDC.
Agenda Item: First Meeting of CDC Advisory Committee Chairs and Designated Federal Officials
MS. JACKSON: I just wanted you to be aware that this was the first ever meeting of the Advisory Committee and Board of Scientific Counselors for CDC. Harry was invited and all set to attend, and his unfortunate event kept him from coming. Marjorie was coming as the designated federal official and her unfortunate event international travel allowed me to attend. So I was very thankful to get there.
Originally I kind of looked at it as a perfunctory kind of meeting, everyone getting together, maybe talking, and meeting each other, wondering where NCVHS fits really in the collection of the Advisory Committee for CDC. But in that session I really had kind of a professional and personal breakthrough to see that CDC is more than the administrative management home base that we use, but there really are some thematic topical issues that we can plug into. So I think Harry would have fit very nicely there, in addition to being part of the nice picture that was taken on the day of the meeting and was posted the next day. That is how precise and fast they are.
The major mission was to get across the Healthiest Nation Initiative Information, and of course we knew that was going to come, and a lot of you of course are familiar with it, it came out the last 2007, and it's getting a lot of airplay where they've got a collection and collaboration of groups and entities and organizations to get across prevention, promotion, and preparedness. So you can see the CDC is trying to burst out of its silos of disease, disease, disease, and really get more into an expanded mission.
In this meeting that was interestingly strategically scheduled the week before the election, the excitement was palpable as to the state and the era of change. No matter what happened they knew they had to do things differently. And this was an attempt to do that.
We got a chance to meet each other. They prepared tremendous background documents, including a roster of all the members of the Advisory Committee meetings, so that everyone got a chance to see each other, know each other, talk, and share. One thing that we took back, we want to take back to our respective committees was to keep everyone informed of what's going on at the CDC level and their initiatives. That is why I wanted to make sure this was on this agenda, and I'll check with Marjorie and Harry to see how we can continue a kind of a cross-cutting communication, at least to see what other groups are doing. That was kind of the main intent. I thought it was very positive, and I think there will be more to come.
MR. REYNOLDS: Thanks, Debbie, and thanks for representing us.
MS. GREENBERG: I think we will need to follow whether the Healthiest Nation Initiative, which I think Garland mentioned yesterday, and I don't know, it was the in the committee on populations, whether it continues to have traction with the new administration, etcetera, because if it does then we probably want to hear more about it. It is very data driven also, probably wait and see what happens there.
MR. REYNOLDS: All right. Do you want to go ahead and cover the WHO?
Agenda Item: WHO Family of International Classifications (FIC) Update
MS. GREENBERG: I hate to be between you all and getting out of here.
I handed out the presentation. I think you all should have gotten a copy of it. I think I made a presentation maybe a few years ago about this work that I'm involved with the World Health Organization collaborating centers on the family of international classifications. Many of you were not here then, so I have provided some background information. I will just go through that quickly. I thought since I have just returned from these two meetings it would just might be of interest to just give a brief report from them.
It was very interesting, the topic of the annual WHO classifications meeting was Public Health Informatics, so it was timely. It was also followed by a two-day conference on Public Health Informatics in Asia Pacific, because we were meeting in India. WHO had done a survey of countries in the Asia Pacific region to see what the status of Public Health Informatics activities, to see what's going on, and 16 countries had responded. So it was interesting the type of things that we talk about on this over here many time zones away, but it's certainly going on there as well.
With everything on the Committee's plate, I don't know the extent to which you really want to have any kind of international initiatives, but as any of us who have been to Disneyland knows, it's a very small world. This is just a brief report from the network, just a brief introduction to the network priority, what the priority activities of the network are to see whether any of these are of particular interest to this Committee. I think many of them are. Some recent meeting highlights.
Then the synergies that I saw between the work we were doing in the classifications network and this global partnership in Public Health Informatics, and synergies of that or both of those with the National Committee. I know, Walter, I think you went to this first meeting of Global Partnership in Public Health Informatics, it was a year ago at the University of Washington. I don't know if anyone else has been involved with that, but there is this partnership that is being developed and has been supported by some foundations on the Health Metrics Network, particularly to try to bring the benefits of public health informatics to low resource countries. We may actually fit in that category at some point. Everything is relative, right now we aren't considered a low resource country, but if you look at what's happening with our vital statistics system maybe so, or the report we just heard from Bill.
Anyway, the WHO Family of International Classifications is a suite of classifications for international use basically to capture core health dimensions such as death, disease, functioning disability, and health, and related health system parameters potentially such as health interventions.
I made a full copy of this in color because you couldn't read the slide otherwise, but the main classifications are of course our old friend, the International Classification of Diseases. I can tell you that the rest of the world is happy to know that we may be finally adopting ICD-10 for morbidity. They have managed to do so in India where it's over a billion people. But anyway, enough said on that.
Of course the International Classification of Functioning Disability and Health, which I think continues to be on the radar screen of this committee, and a number of related and derived classifications. ICD and ICF, for those of you who aren't aware of it, are maintained and promoted by this network of WHO collaborating Centers, and other classifications can apply for this related or derived status, which basically is an effort to kind of harmonize these different international classifications, and avoid duplication but also see where there are gaps.
Right now there is no International Classification of Health Interventions. There was one years ago. This is not so much a problem for the U.S., where we have too many probably intervention classifications, we haven't even managed to have one, but in many countries they are trying to implement case mix systems, some things like the DRG, etcetera, you have to have a health interventions classification. So they may be adopting the Australian one, they might have adopted ours, our ICD-10 PCS, but it's taken us so long to implement it, but they've gone elsewhere.
It is a problem for a lot of countries that there isn't an international standard in this area, and it also a problem for doing any kind of international outcome analysis. If you want to look at the outcome of interventions or the cost effectiveness of interventions or the frequency of different types of interventions across countries, this is no way to make these comparisons.
One possibility is that at least they would create some kind of a template that everyone's different classifications could be mapped to, as well as providing a classification that countries could use, could adopt if they don't have one of their own. So this is still on the possible horizon.
These centers which I mentioned that promote and maintain the classifications have been established by language and geography over the past 50 years. The very first one was in the United Kingdom, and I thought it was very interesting that they have for some reasons I won't go into haven't been a collaborating center. The London center was done, the UK center, so off the map of collaborating centers about four years ago, and now we were meeting in their former largest colony, which has just become the first collaborating center, India, in the Southeast Asia region, so it's an interesting world.
These collaborating centers are designated by WHO to their regional offices, etcetera. Our mission is to improve health through the ongoing development, maintenance and promotion of an integrated suite of health classifications and related products that produce information of value and utility across the world. So I think they're saying some of the same terms that we've been talking about the last few days.
We have a number of committees, I won't go through them, I'm the Co-Chair of the Education Committee. And we also have reference groups, which the committees look at all the classifications, and particularly ICD and ICF, the reference groups drill into particular uses of particular classifications. We have a mortality reference group, a morbidity reference group, functioning and disability, and then there's a terminology reference group looking at the relationship with IHTSDO and SNOMED, etcetera, and the ontology behind really all of the classifications.
Our priorities are the implementation of ICD-10 for both mortality and morbidity data systems, the updating and revision of ICD-10. We do annual updates of ICD-10 with major updates every three years, and the development of ICD-11, with the goal still being that ICD-11 will be brought to the World Health Assembly in 2014. But this could change, it is not resourced probably adequately to maybe have that happen. But the idea is ICD-11, as I've reported in the past, could in some ways be a very major update to ICD-10, so that by our country moving to ICD-10CM, as new content is proposed for ICD-11, it can actually be built into ICD-10CM, other clinical modifications of ICD-10 and ICD-10 in general, so that we wouldn't have the kind of disrupted transition that we have had in the past both for mortality and morbidity.
Now we are finally starting the process to update and eventually revise ICF classification and functioning and the updating platform. And by the way, you may not be aware of this, I wasn't even, the platform for recommending updates to ICD-10 and ICF is open to anybody, all you have to do is register. So if you have an idea of a new code, now people haven't been that interested in ICD-10 in this country at least for morbidity, we haven't implemented it. Once we do though, but even ICF, anyone can go on and recommend it, and update and bring evidence to bear. Now, that will then be reviewed by the appropriate committees, etcetera.
We are looking at implementation of ICF in censuses and surveys as a framework for that, and that has been recommended in this country by the Institute of Medicine. Health outcomes at the clinical and service level. Right now the main outcome is whether a person lives or dies, or generally people don't get cured of a lot of chronic conditions certainly. But what are their functional outcomes, right now we really don't collect much information in that area, but that's generally what's important to I think most people and patients.
Use of ICF in administrative and clinical information. A number of countries are looking at using an ICF framework or a functional framework for redesigning their disability eligibility systems. And the whole social policy field, whether it be housing, or the whole range of what we broadly call environmental factors, transportation, housing, all the areas that are very important to influence wellbeing and health. And we look to identify best practices and provide a network for sharing expertise and experience on implementation in these classifications and training in their use.
There is an International Training and Certification Program which I have been working on for coders and trainers, currently focused on ICD, but particularly for countries that really don't have a strong infrastructure for that in their own countries.
We're developing electronic training tools, web based tools for training in both ICD and ICF. We are very interested in improving the source documents from which data are coded, such as the death certificates and hospital records.
We have developed policies on a common electronic platform for maintaining all the classifications and filling the gaps, as I said, in the family of classifications. We are collaborating with WHO and the International Health Terminology standards development organization on a harmonization agreement for mappings between SNOMED CT and ICD ICF, and then ICD-11, with the idea that actually ICD-11 would – right now the mappings are kind of artificial because you have a classification which is one construct, and you got terminologies which neither of them were developed with the other one in mind. The idea would be to actually build into ICD-11 a terminological framework underpinning based on probably SNOMED.
Our recent meetings that were held outside of Delhi, there were two as I mentioned. Some of the highlights were, as I said, the first collaborating center in Southeast Asia region was established in India. We approved 133 updates to ICD-10, some of these are more major than others, but you can see we're still updating ICD-10. We agreed that the next major update of ICD-10 is in 2010, and there was some thought that that would be the last one because of ICD-11, but a lot of people had a problem with that given that they know they'll be using ICD-10 or a clinical modification well into the next decade. We agreed that we would continue to update ICD-10, a major update in 2010, 2013, 2016, and until ICD-11 is available for implementation.
Then, as I said, we have expanded the updating platform to include updating for ICF, which will begin next year.
MR. REYNOLDS: Marjorie, I have a question. These updates to 10, as they would continue through whatever they continue through, how then do people take those and actually implement them?
MS. GREENBERG: We have been, I mean ICD-10 was the first ICD version in which we updated between major versions, so like ICD-9 we had no capacity to do that, so when AIDS came along we had to sort of stick it in somewhere, we were able to put a code in ICM but there was no code capacity to implement 9 to update 9. Now we have been updating ICD-10 since 1998 or 1999.
What we have been doing in this country since is we've been adopting those updates in ICD-10 and they have been impacting on our mortality data. So whenever, as I said, it's every three years that we put in the major updates, the others are like index changes. And then as we have been developing and updating ICD-10CM, even though we're not using it yet in this country, we've been updating it to reflect these international updates.
So when we implement 10CM it will include all of the updates of ICD-10. Many of them actually have been influenced by what we have done in ICD-10CM.
MR. REYNOLDS: But I am saying once this happens we'll need to understand whether we need to put this into the standards process to make sure that people adopt them. In other words, if you're doing that every year or whatever you're doing that's not something –
MS. GREENBERG: It will be the same as with 9CM.
MR. REYNOLDS: Oh, okay. Thank you, that's exactly the answer I was looking for.
MS. GREENBERG: We have to be consistent with the international standard, but it will go through coordination and maintenance, and just as every year we have a new version of 9CM then we'll have a new version of 10CM which will include not only what we've agreed to in this country through the coordination of maintenance process, it will also include –
DR. SUAREZ: HIPAA has actually an established process for requiring that the new versions of these coding systems be adopted.
MS. GREENBERG: It will be part of the same process. We also, as I said, the web based tools, we have developed a core curriculum modules for training in ICD and ICF.
Then we have agreed, and I think this should be of interest to this Committee, the tentative theme for the 2009 annual meeting is going to be Primary Care Classification. So I think that fits in very much with some of the interests that this Committee has with primary care and the patient centered medical home, and even person centered health information. I think we're kind of consistent, there are a lot of synergies going on.
I mentioned this Public Health Informatics survey in the Asia Pacific region. There were 80 key informatics in 16 countries; over half reported efforts to introduce electronic health records in their country. Now you have to dig down at what that really means, but it is certainly I think interesting and kind of impressive.
They thought that the most feasible areas, and I think this is true here probably too, is getting routine laboratory data, which tends to be digitized in their countries as well more. But their highest priority was monitoring patient safety, or having electronic health records that could monitor patient safety. But right below that were mortality and using informatics approaches to improve mortality and morbidity statistics.
MR. BLAIR: Why did patient safety become the highest priority?
MS. GREENBERG: I don't know. This is what the key informants in these countries said, that I guess the biggest issue that they would like to be able to use electronic health records for, and were completely incapable of doing that at this time, it wasn't that there are resources going into it, but when they were asked about a series of use cases, as it were, for public health informatics, and they were mortality data, morbidity data, there were other use cases, patient safety was one of them. And the majority said that was their highest priority.
Their major barriers are funding and lack of human resources or technical capacity, but also political will. And I think we see that sometimes even in our own country, that's interesting.
So what I saw as some of the common goals, and I think they're common goals with this committee too, is there is definitely a commitment to standards as building blocks, I mean this is a worldwide commitment including in the Asia Pacific region. And of course there is Public Health Informatics Agenda, the global Public Health Informatics Agenda, a big emphasis is on standards. We are all looking to build information infrastructure, whether it be vital registration systems or electronic health record systems, or both, and interoperability.
One of the things that was interesting, like Singapore, which has actually just joined the IHTSDO, the International Health Terminology Standards Development Organization, so they are planning to use SNOMED for their electronic health records. They apparently have a number of hospitals that have electronic health records, but they are not interoperable at all. Apparently one hospital you get transferred to another hospital, they download the whole thing in paper.
On the other hand, they have electronic health records. We heard from Mozambique where there was no one person in the entire country whose fulltime job was working on health records. It is such a diversity.
Of course, they want to improve the quality and comparability of information, make useful information available to decision makers. Build the capacity of public health healthcare and health information workforce. Capacity building was the major topic that we were talking about in all of these meeting, and commitment to person centered policies, for example, safety they saw as a very person centered issue.
Some of the common barriers, as I said, were inadequate human and financial resources, limited appreciation of the importance of the work. Why are we dealing with what we're dealing with right now in health statistics, and in the slow adoption of so many things? I think we see this here as well. It is just very interesting.
The lack of infrastructure in the developing countries, they don't even have the basic infrastructure, but the fragmentation of systems in all countries, developed and developing, the resistance to sharing data, and privacy concerns, and insufficient and incompatible privacy and security standards. I can tell you that we had a lot of presentations in that Public Health Informatics conference before anyone even mentioned privacy and security, but eventually it came out. And several of us emphasized that this is a deal breaker if you're not looking at this. I think it's a small world.
So that is my view from the field. I would just like to keep you updated, and if you have any questions I'm happy to answer them, or in the future I'm happy to as well.
MR. REYNOLDS: Judy then Jeff.
DR. WARREN: I just wanted to make two comments about this. One, I serve on IHTSDO's quality committee, and we were very pleased at the negotiations for a relationship between IHTSDO and WHO FIC. I mean from an international perspective, and it's interesting serving on an international committee, especially when I represent Africa and not the U.S. You are selected to do regions, and if there is nobody in the region that comes up then they look at the pool of experts and assign you a region.
The interesting piece is that WHO has this huge reputation for the standards of data. Again, though, most of their terminologies are classifications, or they are things like the ICF, which are really kind of metrics for evaluating things, so that it was imperative from an IHTSDO perspective that there become a relationship with WHO, so we're delighted that that has moved forward.
Then just on a personal note, there was joy in mudville, or I should say in nursingville, around the world when the ICMP established a relationship with WHO. We have been working on that for at least about 15 years.
MS. GREENBERG: And I don't know if I said, but one of the reasons we wanted to bring them in now was to get them to harmonize also with ICD and ICF.
DR. WARREN: Right. And the other thing it allows is within IHTSDO it allows us to work with ICMP to harmonize them as well in with SNOMED. Until we can establish a world value case, our rules are we don't harmonize any terminology unless there's an international case. From a nursing perspective that was very, very important.
MR. REYNOLDS: Jeff.
MR. BLAIR: You know, Judy, with your background, let me address this both to Marjorie and to you, we have been moving down a path where we've gone from ICD-9 to ICD-10, with greater clinical specificity for ICD-10 and different versions of ICD-10, onto 11. At the same time and parallel with that, there has been further developments with SNOMED, which has become international.
My question maybe to both of you is could you help us understand from your personal speculation, I'll put it that way, how you see the ICD path and the SNOMED path, do you see them converging, do you see one prevailing over another, do you see them merging? Where do you think we're going now with the direction of SNOMED and the ICD-10s and 11s?
MS. GREENBERG: I'll let you go first, Judy.
DR. WARREN: Let me tell you the SNOMED perspective. From the very beginning SNOMED was once it became multidisciplinary became a terminology to describe or to support patient description at a granular enough level to really identify what was happening with that patient, to drive decision support and all of those things. We see our relationship with ICD, whether it's 9 or 10, as using the power of the classification framework, which is not only a set of concepts, but also there is lots of knowledge information behind those concepts that a classification system provides. A terminology like SNOMED does not have that knowledge base. It is a set of concepts with heuristic ontology behind it that show you the relationships, it does not have a knowledge base.
So for our perspective, the two work in partnership in order to handle the needs, but you'll need both to do that.
MR. REYNOLDS: Mark, did you have your hand up?
MS. GREENBERG: If I could just answer that as well. I think some of you will recall when Chris Shoots(?) was here and he was talking about how if we didn't work together there was this possibility of ICD seeing the need for more of a terminological underpinning to develop its own terminology, or a ghost SNOMED, and there was a potential of SNOMED if it couldn't work with ICD seeing the need to have a classification, which aggregates, etcetera, with the knowledge, and aggregates in logic developing a ghost ICD, so we all want to avoid that.
I think that I would agree with Judy that this is now a high priority of the management of both IHTSDO and WHO. And I am involved with this development of the harmonization efforts, so I'll keep you posted on those.
MR. REYNOLDS: Mark.
DR. HORNBOOK: Marjorie, was there any discussion about preference health status measures across the countries?
MS. GREENBERG: You mean like global burden of disease?
DR. HORNBOOK: Health utility index, for example, URC.
MS. GREENBERG: Yes, a little bit. But the Europeans were not in the informatics meeting, there weren't that many of them there, interestingly enough, a number of people from the UK, so there wasn't a lot. I mean it did come up, but there wasn't a lot. But I certainly think it's on the Global Public Health agenda, you know, Public Health Informatics radar screen.
Walter, when you were at the meeting you attended was there much discussion of that?
DR. SUAREZ: I don't think that was one of the topics that was covered by the Global Public Health Informatics Initiative. At the first meeting it was more organizational in nature, and the topics were more broad issues around public health informatics.
DR. HORNBOOK: Health Statistics Canada added the Health Utilities Index to their national health survey at least two ways now, so they now have a global health score for all the respondents who have answered the questions adequately, which gives them a whole set of new analyses that were never possible before.
MS. GREENBERG: I can say that in the ICF community at least, there is a lot of angst about giving people global scores. I mean there are a lot of debates on this.
DR. HORNBOOK: Economists are comfortable with it.
MS. GREENBERG: But a lot of people are not. It certainly is something -
DR. HORNBOOK: The problem with profile scores is that you are restricted to dominant situations where somebody is health on every single scale, and the world is that's a pretty rare set of occurrences so you lose a lot of reality options that you need to deal with in your resource allocations.
MS. GREENBERG: I think there are some other international groups who are focusing on this. So it wasn't that much a part of the – but the German collaborating center has taken the lead in developing core sets for ICF. Now they are also now working on a generic core set, which could actually -
DR. HORNBOOK: Utility weights.
MS. GREENBERG: Yes, utility weights, etcetera. So they are the one group in the WHO FIC network that really is working in this area. If you are interested in being in touch with those folks I can certainly give you contacts.
MR. REYNOLDS: Marjorie, thank you. I want to also commend you for the work you're doing worldwide with the World Health Organization.
And I think at the next meeting we will need to hear more about why Africa got stuck with Judy. I would like that to be a definite Full Committee discussion as to what that would be.
[Laughter]
MS. GREENBERG: I think this is one of the good news out of Africa.
MR. REYNOLDS: It might have been better news for the U.S., but we'll see what the heck that means, but we'll go on from there.
Last bit of housekeeping, if you didn't pay for lunch yesterday, please get it to Jeanine, we're still not quite balanced.
With that, thanks everybody for your energy, your focus. I thought we had a good discussion and we have an interesting journey ahead, but I think we have the right people to do it. So thanks everybody, I really appreciate it.
MS. GREENBERG: Thank you.
(Whereupon, the Full Committee adjourned)