[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
FULL COMMITTEE MEETING
NOVEMBER 19, 2009
National Center for Health Statistics
3311 Toledo Road
Hyattsville, Maryland
Proceedings by:
CASET Associates, Ltd.
Fairfax, Virginia 22030
TABLE OF CONTENTS
- Call to Order, Welcome, Review of Agenda – Harry Reynolds, Jr.
- Department Update – Mr. James Scanlon
- CMS Update – Tony Trenkle
- Quality Letter on “Meaningful Measures”-ACTION – November 20th – Drs. Justine Carr and Paul Tang
- Department’s Work on Comparative Effectiveness Research, Briefing – Patrick Conway
- Enhancing Health Information Capacity for the 21st Century – Drs. Don Steinwachs and Bill Scanlon
P R O C E E D I N G S 9:00 a.m.
Agenda Item: Call to Order, Welcome, Review of Agenda
MR. REYNOLDS: Good Morning. I would like to call this meeting to order. This
is the first 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 am Harry Reynolds, Chair of the Committee. I want to welcome Committee
Members, HHS Staff and others here in person. I would also like to welcome
those listening on the Internet and would like to remind everyone to please
speak clearly into the microphones. 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 introduction. Now
before I start that, I have no conflicts of interest and I work at Blue Cross
and Blue Shield, North Carolina.
Another thing I would like to do before I go around, is I would like to
welcome our new member, Tony Rodgers.
(Applause)
If those of you that remember when we had the Meaningful Use Hearings, he
was one of our star testifiers and as Jim Scanlon and I were doing, I have
actually quoted him a number of times on Data Stewardship because of his
presentation and the idea, as he stated, “They stepped up to be the
steward of the people’s data,” was an important statement to us and
really, I think, one of the key things. I know Justine and others have also
referred to that. So it is exciting to have him join us when we already know
that he knows what he is talking about and then we cannot tell him that he
ought to listen to us.
(Laughter)
So that is good. We have got another player that is already coming in full
service, so that is pretty good. So do not act like he is a new member, guys.
He knows what he is talking about, so do not dance around over there. So Tony,
thank you. We are really, really glad to have you. So with that, Jim, please.
MR. SCANLON: Good morning. I am Jim Scanlon. I am the Deputy Assistant
Secretary for Planning and Evaluation at HHS and the Executive Staff Director
for the Full Committee.
DR. CARR: Justine Carr, Chief Medical Officer, Caritas Christi Healthcare,
Member of the Committee, no conflicts.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, Member of the Committee,
no conflicts.
DR. FITZMAURICE: Michael Fitzmaurice, Agency for Healthcare Research and
Quality, Liaison to the Committee, and Staff to several Sub-Committees.
DR. MIDDLETON: Good morning, Blackford Middleton from Partners Healthcare,
Brigham Women’s Hospital and Harvard Medical School. I am the Corporate
Director for Clinical Informatics Research and Development, Member of the
Committee and the Population and Quality Sub-Committees and no conflicts.
MR. LAND: I am Garland Land, National Association for Public Health
Statistics and Information Systems, Member of the Committee and no conflicts.
MR. RODGERS: I am Tony Rodgers. I am the newbie. I am now one of the
principle consultants at Health Management Associates and I have no conflicts.
DR. GREEN: Larry Green, University of Colorado Denver, Member of the
Committee, no conflicts.
DR. STEINWACHS: Don Steinwachs, Johns Hopkins University, member of the
Committee, and no conflicts.
DR. SCANLON: Bill Scanlon, National Health Policy Forum, Member of the
Committee, no conflicts.
DR. WARREN: Judy Warren, University of Kansas School of Nursing, Member of
the Committee, no conflicts.
MR. BLAIR: Jeff Blair, Lovelace Clinic Foundation, member of the Committee,
no conflicts that I am aware of.
MR. HOUSTON: I am John Houston. I am from the University of Pittsburgh
Medical Center. I am a member of the Committee and like Jeff; I do not have any
conflicts that I am aware of.
DR. HORNBROOK: Mark Hornbrook, Kaiser Permanente, member of the Committee,
no conflicts.
MS. GREENBERG: Marjorie Greenberg, NCHS, CDC. I am the Executive Secretary
to the Committee and I would like to welcome you all here to NCHS and welcome
our new member.
MS. KHAN: Hetty Khan, CDC, NCHS, Staff to the Sub-Committee, and Privacy and
Confidentiality.
MS. WILLIAMSON: Michelle Williamson, CDC, NCHS, and staff to Standards.
MR. BURKE: John Burke, College of American Pathology.
MR. FRIEDMAN: Dan Friedman, independent consultant.
DR. PARRISH: Gib Parrish. I teach in Vermont, independent consultant.
MS. JAMISON: Missy Jamison, NCHS, Committee Staff.
MR. EZMONGA: Don Ezmonga(?), AHIMA.
MR. QUINN: Matt Quinn, Agency for Healthcare Research and Quality.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, CDC,
Committee Staff.
MR. REYNOLDS: Are there any Committee members or staff on the phone? One
thing that I would like to do before anybody else joins is -– Jorge Ferrer
will be probably joining us by phone. I spoke to him yesterday. He is going
through some tough times right now physically, and I think it would be good if
you spend any time with him, it might be good to give him a call and wish him
the best. He is doing great, but I think that would be good. He is a good guy
and makes a lot of difference in lots of ways, and so I asked him if it was
okay for me to share that and that is all I will share. But I think you know,
really good spirits, really things look great, but I think it would be good to
reach out and touch him if you get a chance; good guy. So, I just wanted to
share that with everyone so that when he gets on, you will understand why for
the next few times we talk to him, he will probably be on the phone. He
probably will not be available to us.
MS. GREENBERG: He has been very dedicated to this Committee, I must say.
MR. REYNOLDS: Absolutely. Well and dedicated in his field, too.
So we are going to move into the agenda and I think we have some exciting
things. We are going to have Jim first. Tony Trenkle is going to join us. Tony
is at a special staff meeting and so he will probably dial in at 9:30. I asked
him when he dialed in to let me know he is here. So if Jim is talking and he
interrupts you, it is not because he wants to.
(Laughter)
MR. REYNOLDS: And then, you know, we are going to get into the Quality
letter and I would like to thank everybody on the Committee and the Executive
Sub-Committee and others, that have worked diligently on this. One thing, I
think we have an excellent synergy going on with Paul Chairing the HIT Policy
Committee for ONC. So this idea of how we work together and what we do, so I
think what we will continue to hear today is this idea of -– this is a
journey.
And so as we continue to put out installments in this journey of how we make
a difference and can pass things along, I think that is going to be key as we
do this, so I want everybody’s attention and let’s try to make something
happen. And then we are going to hear a little are going to lead a discussion
on Enhancing the Health Information Capacity for the 21st Century, based on
what we have said at our last meeting. Then later, we will have our breakouts.
Then tomorrow, hopefully, close out the letter and we will get updates from ONC
and then Jeff and Judy are going to talk about the HIPAA Report that we are
moving on and then we will have our update on the 60th anniversary
again and then an update on Vital Stats. So with that, Jim, why don’t you go
ahead and proceed.
Agenda Item: Department Update
MR. SCANLON: Good morning everyone. Let me report on a number of things that
have occurred since we met in September. I am going to update you on leadership
petitions at HHS, on our budget, on the Recovery Act briefly, and then we will
have other folks, Patrick Conway will take us more deeply into the Comparative
Effectiveness Research part of the Recovery Act. And then a little bit of the
work within HHS and then looking ahead to health reform and the Committee’s
involvement.
So let me start with personnel changes at HHS. We continue to fill our
leadership positions. Since we met in September, we have named a new Chief
Technology Officer who is Todd Park. I think many of you remember Todd. He was
associated with Athena Health and he has now joined HHS. His job will be to
serve as a catalyst and a leader and to support our mission through the latest
in technology. So he has a number of projects and I think at a future meeting,
we will invite him to come and talk about where he is as well. He clearly has
been drawn into all of our Health IT as well as our Internet initiatives and
our other planning as well. In addition, we have had a new Surgeon General,
Regina Benjamin. She was confirmed two weeks ago and she has arrived and
hopefully she will get right into the swing of things on Public Health. We have
a number of other folks who are slated for hearings in the Senate. They would
require confirmation in the Senate. So hopefully, before long we will have all
of our positions filled.
In terms of budget, and this is what the federal agencies worry about
obviously, we are now in the first quarter of fiscal year 2010. We do not have,
nor do most agencies have, an actual appropriation. We have a continuing
resolution which basically allows us to continue spending proportionately, as
we did last year, a very highly controlled kind of an arrangement, through
December 18th. Basically, we cannot start new activities. We basically have to
continue the operations. It is basically continuing the operation level as was
done last year.
The thinking seems to be that we will probably have, after December
18th, an omnibus appropriation which usually means a number of
agencies and departments are combined together in one overall spending bill.
Because of the number of legislative days winding down, that seems to be the
outlook though, of course, anything can happen.
We had some good news in the President’s budget for 2010. We just have to
see if it materializes. On the population health statistics side, we had an
increase for NCHS. A sizeable increase would have helped us get back to at
least current services levels there and for MEPS and some of our other
population data and statistics, as well. So we are keeping our fingers crossed
so that we will see those increases maintained when we get the actual
appropriation. And remember, it is a very large budget. Our total is $800
billion dollars, but only 10 percent of that is discretionary. The rest deals
with entitlements and we have relatively little discretionary control over
those.
Let me turn to the Recovery Act -– always in the news, and obviously
the Recovery Act Fund supports a number of Health IT activities. Remember that
the Recovery Act included for all of HHS over the 10 years, about $167 billion
dollars. Much of that is associated with increases in the federal match for
Medicaid and for some of the incentive programs associated with Medicare and
Medicaid. But we did get about $22 billion in discretionary spending and there
were three funds that, again, were aimed at supporting public health and in a
sense, a down payment on health reform. The first was the creation of a
Prevention and Wellness Fund at about $1 billion dollars. Second was the
establishment of the Comparative Effectiveness Research Fund which is $1.1
billion. Dr. Patrick Conway will be going into that later today on where we are
with that.
And, of course, the third major arm was the Health IT, the HITECH Act, $22
billion dollars. But if the incentives for Medicare and Medicaid really took
off, it could easily grow beyond $22 billion to $44 or $50 billion over the
years of the incentives. And again, the aims of the Stimulus Package were state
fiscal relief, build up the healthcare safety net to support biomedical
research. A lot of construction programs –- shovel ready construction
programs and obviously human services investments.
In addition to the normal healthcare and public health and scientific
contributions that those investments would make, the thinking was that they
would create an economic stimulus as well. They would create or retain jobs and
they would have multiplier effects at the local community level. Just this as
an aside, I remember seeing that NIH estimated that for every $1.00 in their
R01 funds, it generates a $1.35 in the local economy. And it supports, and you
know better than I probably, six to seven positions. So, again, just to show
what the local multiplier effect might be.
Let me turn to the HITECH Act because that is probably the most directly
relevant. Chuck will probably give us much more detail on where we are
specifically tomorrow. You will remember that there are two parts of the
HITECH. One is the $2 billion dollars that creates a lot of capacity grant
programs to the states for health information exchange, extension centers to
assist in the adoption of electronic health records, health IT workforce
grants, some public health infrastructure programs and standards and the
opportunity for research and development centers of excellence and so on in
Health IT. But another big part, and even the bigger part of the HITECH Act, is
the Medicare incentives. So beginning in 2011, eligible providers in hospitals
who demonstrate meaningful use of electronic health records will begin to be
eligible for incentives. Again, this is a voluntary program. Depending on the
uptake, it could be anywhere from $19 billion to $40 or $50 billion dollars
over the period.
We are and Chuck gave an update on this, we have been working on the
regulations necessary to implement that program. What exactly is meaningful
use? How do we measure it? Paul and the Advisory Committee have given us a lot
of very specific criterion about how to do this. The regulations, basically,
are operational on how this would be done. I think before long, hopefully
within a week or two, the meaningful use regulation will be available for
public comment. There are other regulations and policies that would be needed
that are required by HITECH and the Department is working on those as well.
A minute on the Prevention and Wellness Fund, and again, I think these three
elements here with the Recovery Act, I think, were viewed conceptually at
least, I have heard people refer to them as a down payment on health reform in
the sense that these were seen as necessary improvements in the healthcare
system that would help with or without broader health reform. These were things
that were needed that would help us down the road. But on the Prevention and
Wellness Fund, the total I remember was $1 billion dollars; $300 million
dollars goes to states for general support of the immunizations grant program.
Another $50 is directed to states for prevention of healthcare associated
infections, which gets to be a big issue every year as you all know. But the
biggest part is an initiative that provides funding on a competitive basis to
communities. It is called Communities Putting Prevention to Work. The idea here
would be that communities would apply the funds and at the community level
again, they would initiate evidence-based public health interventions,
community sites now, in several areas. And then there are a fair amount of
lessons learned and evaluation components for this as well. And the areas of
emphasis seem to be increased levels of physical activity, so again
community-based emphasis here, improve nutrition, decrease overweight and
obesity prevalence, decrease tobacco use and exposure to secondhand smoke, and
then a fairly large component for evaluation and data. So even though these are
supposed to be evidence-based community interventions, there is a fairly large
evaluation component to see what happens when you actually apply them across
large settings. So the grant announcement has gone out for those and the
communities are developing their grants and will be submitting them as well.
Comparative effectiveness, I will wait until Patrick gives us a better
report later. The spending plans and the reports on that initiative have pretty
much been made public already, but Patrick will bring us up to date. And again,
you will see that data, data infrastructure, data standard and so on, are a big
part of all of these initiatives and you will see in comparative effectiveness
particularly there are a number of initiatives that would in addition to
conducting the comparative effectiveness research itself, would focus on data
infrastructure and data improvements to support newer kinds of comparative
effectiveness research. And again, I think the thinking was that this is not
only intended for comparative controlled clinical trials. There are other ways
to look at comparative effectiveness research as well.
I will just say a couple of words about health reform because any of the
bills would involve this Committee, more work for the Committee, and again no
one is trying to get ahead of the Congress obviously, but at a minimum we want
to monitor what is occurring. The Department has been providing technical
assistance to the Hill and we used our modeling capacity and actuaries and
other technical capacities as well. But there is, in both versions, a provision
relating to administrative simplification, among other things, and assuming
this passes this would require, and again Tony may want to report more on this,
but it would require operating guides for some of the standards.
We have already adopted and it would require us to push ahead on the
standards that were not developed, claims attachments and so on, that were part
of the initial package, but were not developed, they probably would require the
health plan idea as well and it would require a non-governmental institution to
basically develop those operating guides and keep them up. So we will be
looking at, however this materializes and the specifics, we will keep
monitoring it.
In virtually every phase there are steps for folks to come to the Committee
with proposed standards and proposed operating guides. The Committee here would
be kind of the opportunity for the public to participate, to hear competing
views and so on and recommendations; the Committee would have to make
recommendations to the Secretary for almost any standard or operating guides as
well, before the Secretary would adopt. So again, we will share some language
on that; we have the House language, but we have not seen the actual language
on the Senate yet but again since it is appearing in all of the bills, some
form is likely to happen. I think that CBO actually associates a fair amount of
savings with the implementation and the widespread adoption of all of the
transaction standards as well. I think I will stop there and take any
questions.
DR. TANG: One, in terms of putting leadership in place -– any word on
the CMS Administrator and the second with respect to other kinds of standards,
one of the things that we have had a little bit of email discussion on is the
unique health identifier because it is a different time. It is thirteen years
later. I think we have different privacy rules in place and there is a better
appreciation of some of the costs in terms of errors that occur. And then the
other thought was that in the HIPAA language it was mandatory, whether there is
any organization on the Federal Government that can at least look at this and
explore this.
MR. SCANLON: Well Paul, I was interested in that yesterday. There is no
mention of a unique patient identifier in the administrator simplification
provisions and I am not surprised, but it tends to be a third rail as you know.
And I could ask more formally within HHS if you would like. But whenever I have
asked the question about are there any plans, are there any requests to go in
that direction, the answer is typically no. And I think we still have, we will
see what the appropriation brings for 2010, but typically we have an annual
appropriations prohibition on even thinking about it. It is not simply a
developing and adopting, but I think even the developmental work. I think if
the Committee feels, I can ask. I do not think that is an area that you would
want to go unless you were sort of asked to go otherwise you will just run into
trouble right away. But within the Department, I have not heard about any
request or interest and it is not in Health Reform as well.
DR. MIDDLETON: Two quick questions following up Paul’s comment. You may
recall I did mention actually this issue in my own invited testimony to the
NCHS when we were gathering testimony for Meaningful Use and I think actually
the time is different and the issues perhaps might be revisited in a more
enlightened way that could be extremely useful. Perhaps what this Committee
might suggest is a phased process to look at standards in a manner which is
politically acceptable and tractable and useful, obviously, so it is not viewed
as an all or none but perhaps a phased approach to the next relevant sets of
standards that will be important for all the things that we have been talking
about. That might allow us to paint basically the road map in a way that we
talked about before. I would be very supportive of their being a more formal
inquiry at this time about that kind of process.
MR. SCANLON: I will take that as something I can bring back and particularly
in the context of health information exchange. I mean obviously it is not a
global kind of an issue so much as it is how do you actually exchange and
identify patients in the context of exchanges and whatever else might back up
the exchanges.
DR. MIDDLETON: The part two if I may, thank you Mr. Chairman -– I think
if I understood you to say correctly Jim the proposed CMS measures will be
coming out in a week or so?
MR. SCANLON: Yes.
DR. MIDDLETON: I understand the usual process and again to be the notice of
proposal, making a measure review and public comment and what not. I wonder
actually if we might want to suggest at some point sort of test bed
environments, wherein standards or measures could be actually applied in a
practical way or a test bed type of environment to test their suitability and
utility for the purposes intended. I am not sure if this is something that has
been discussed before or if there is anything like this which exists already,
but I know what happens in practice is that you make comment to your heart’s
content and then actually try to implement something and find something even
else that you had not considered before.
MR. REYNOLDS: We may want to hold that. Tony is going to be joining us and
since he owns the regulations pretty much, I think that would be a great one.
MR. HOUSTON: As I recollect, back whenever the national identifier was able
there was a fair amount of politics associated with that and I guess the crux
of it was concerns I recollect about like issue of privacy and things of that
sort. Am I correct in that?
MR. SCANLON: Absolutely. And there is still. And I would not want to
underestimate how much the concern remains with that. It is quite significant.
MR. HOUSTON: And the reason why I asked that, I wanted to confirm since it
has been such a long period of time, is that – I think that in my own thinking
there are some practical things that we can do that I think would have the
effect of depoliticizing that. It would practically allow us to move forward
with that while still taking the edge off of concerns over privacy. And I will
just say one of my thoughts is that one of the biggest concerns that people had
with social security numbers is that it is used for other purposes. It is
rampant. One way to depoliticize a national identifier is to say it is a felony
to use it for any other purpose than, and then innumerate the purposes that are
permitted and say everything else you just cannot use it for. I think there is
absolutely a compelling case for a national identifier and I think that if we
can think of ways to remove barriers, I think that is a value.
DR. FRANCIS: I agree with the point about there are enormous risks,
including privacy risks of not having an identifier, having a way that is a
good way of linking patients. My own sense is that there is another politics
issue about government giving me a number and the role of the government that
underlies the third rail point here and actually what I think we ought to do is
we ought to scope out the landscape of what are the various possibilities to
solve the problem of making sure that Joe is Joe and what the different risks
are including risks to patient care of these different methods and try to see
whether if we scope things out in that way we might not figure out something
that is a pretty decent end run without the third rail problems.
MR. RODGERS: I would just like the Committee to consider that at the state
level the focus of the Office of the National Coordinator has been to create
state level health information exchange. Maybe that is the opportunity to start
with. That is the base on which you build the standards or eventually a
national ID, but initially if you can gets the states to cooperate, especially
through the grants and come up with a common set of standards for development
of state level patient ID, especially as it relates to health information
exchange. You can then build on that eventually, the national idea. I think it
is just a cultural issue about people having a national ID and concerns, but I
do not hear the same thing at the state level. People seem to be more
accepting. So that may be an opportunity for us.
DR. TANG: Well I just wanted to go back to what Jim presented which is we
are being asked to share data, exchange data for care coordination. That is the
positive and the up side and this is really to find a way that would reduce the
errors when we do that. And so that is a different context and I think that is
the way to approach this, or one of ways we can.
MR. REYNOLDS: And I would echo that and again I would temper the Committee
that it is very clear when appropriations are not allowed for certain things,
it is very clear that leading with that as the first two words puts us in a
position of not necessarily helping. I love the Paul presented it and the way
that we have heard it in a couple of things to where there are going to be
exchanges. In the past when we talked about this everybody was kind of talking
about their might be. There is now money, there is appropriation, there is
action, there is state action, so what could we bring to the table about what
are the best practices in the states that are putting together HIE’s to make
sure that the patient gets identified. So I just ask officially from the
Committee, it is clear that if we start with this discussion about a single ID,
we are starting from a place of un-comfortableness in many ways. But I believe
if we start from the position of, these things are going to occur, what are the
options of which we obviously know one is some kind of a number. Then I think
we find ourselves in a position of helping, not in a position of having people
say you know there is not going to be one of those; get over it, move on and I
think that would be a service that we can offer and offer many discussions.
MR. BLAIR: Let me strongly agree with the way that Harry has worded the
statement. I indicated that there might not be any conflicts of interest. I do
not know whether this is a conflict of interest or not, but you know, I lead a
health information exchange. We have a master person index, we identify
patients and yes, for a long time I thought the most simple direct way was to
have an identifier, but I think HIE networks all over the country now have
methods to identify the records for a patient to provide treatment, payment and
operations in a manner that is arguable more accurate than a single patient
identifier without triggering concerns or anxieties and I think that is as far
as I will go.
DR. GREEN: Well I was just wondering if Jim was sitting over there thinking
he is glad he brought this up.
(Laughter)
MR. REYNOLDS: It was coming up whether Bill brought it up or not. He had the
courage to go ahead, because it was coming.
DR. GREEN: Just a few things here, I want to agree with your point and
endorse it, but then just observe that in the midst of talking about billions
of dollars, this is what got the Committee talking and there is a reason for
that and that is because the Committee’s charge is to look after the measures
that are necessary to assess and measure the health of the country and it is
our charge and responsibility to work of the value equation to see that people
get what they are paying for and that sort of stuff. And there is this nasty
little detail. This gets personal. And so it is not about a national ID, it is
about returning to the American people value for the dollars they are putting
into public health and personal health and that sort of stuff. We should
approach it like you said, that the work that the Committee is doing now, the
reason that this provoked this discussion is that all of us on the Committee
recognize how essential this is to us fulfilling the charge we have been asked
to fill.
DR. W. SCANLON: In the context of I think the sensitivity of this; I mean we
have seen over the last few months that the opportunity or the possibility of
having reasonable discussions is not always there. And this is potentially one
of those and so along the lines of what Jeff said, I think that in addition to
knowing about how important this is, thinking about sort of second best
alternatives to a national permanent identifier is something that we need to
consider. What is the fall back strategy that improves upon the status quo but
does not go to the third rail issues that have been identified before. And that
could be a very big Contribution of the Committee. Things get blown out of
proportion so often and then they get lost, totally, and we do not want that to
happen.
DR. HORNBROOK: For those of us that have been working with identifiers in
research you know that even when you have the number on everybody you do not
get one hundred percent match and you need to have multiple attributes to
assure yourself of a strong match because there is data entry errors, there are
people who are illegally using SSN’s and there are people who for some reason
you could not read their number and so they got transcribed into something
else. And even then, when you use date of birth, name, address, gender and
social security number, there is still a set of people who are matching one to
many and you cannot figure out which one it is. So that you may get to 93
percent, you may get to 95 percent but you never get to one hundred percent, so
just to remind us that identification is more than just a single number ever.
MR. BLAIR: That is exactly what I was alluding to when I was saying,
arguably more accurate than a single number.
MR. HOUSTON: I just think what Bill had proposed, I think, if maybe if we do
make recommendations in the context of the whole range of options which
includes a national identifier as well as other options, and maybe put together
a list of and do it almost in tabular form, a list of pros and cons and other
comments or criteria we think are important, I think that would work very well.
I think it would allow us to both raise the idea of national identifier, but in
the context of what are the range of options and if something needs to be
established, one of the options or some other option needs to be pick that
allows us to move this forward. So I would propose that maybe we could do that
through privacy or we could do that through whatever sub-group or a separate
working group and maybe try to put this together in fairly short order.
MR. REYNOLDS: I will address that in a minute. Bill. No, Jim.
(Laughter)
MR. REYNOLDS: I have been talking to Bill a lot lately, so if I am talking
to a Scanlon it has got to be Bill. So give any answer you want and it will be
attributed to him in a minute so you do not have to worry.
MR. SCANLON: What was I going to say? I think number one; I feel very
protective of the Committee. I really would not want you to proceed very much
in this direction until I can run it by because otherwise, believe me, this is
political. This is not the scientific realm or the logical realm. Bad things
will happen if it looks like you –-
MR. HOUSTON: Well Harry named us on every letter and so we are not too
concerned.
(Laughter)
MR. SCANLON: It is worse than that. I mean institutionally the Committee
–- well bad things can happen believe me. The Committee could be
abolished, all sorts of bad things could happen especially in a context of
health reform where it is going to get nasty.
DR. FRANCIS: Bad things happen.
MR. SCANLON: So I think we want to proceed cautiously and we want to have a
client. I do not think we want just an unstructured inquiry. On the other hand,
we have applications from every state for health information exchanges. We will
be awarding those grants. Some of them, as you know, are at a very –- well
it is more of a plan than an actuality and in other states there is actually
some capacity and infrastructure there. And again, this is a problem that would
have to be solved. I would not start out with the word national identifier. I
would start out with patient identification and see what people plan to do and
what has been successful and what has not been. But again I would, for the sake
of your own protection and mine and Marjorie’s as well, I would rather get a
request from a client along these lines. I think this would be very helpful.
And again the timing would be important here as well.
MR. REYNOLDS: Judy and then we will close the time.
DR. WARREN: I just wanted to remind the Committee we did have hearings about
national identifier a couple of years ago.
SPEAKER: Wasn’t it about matching a patient with his or her records? It was
not about national identifiers.
DR. WARREN: Which we came up with to have that. So if people want to take a
look we did a good job of pulling together the standards and stuff that would
need to be involved. So if you will go back and look at that, we also put in
some of the recommendations and the e-prescribing letter about matching
patients to their data. And that was pretty good strategy, although I will
remind the Committee we did stop that line of inquiry and so I am supporting
Jim. Be careful.
MR. REYNOLDS: Speaking for the Committee and for the minutes, the Committee
is not intending to go forward to have discussion on an individual identifier.
We are continually focused on things like we have the stewardship stuff in
front of us, we have got the matching patients to their records that we have
done before, we have got the HIE’s in motion that are going on which is a
landscape that we are living in. Figuring out how to make sure that we have the
right person with the right data is in our purview, it is something that we
have dealt with and that is entirely what we would focus on and use as the
basis of anything we do, period. But we will still let Jim make sure that going
forward under those auspices does not put us in anything. And then if we know
that is the case, then I am comfortable working with Tony and I am comfortable
talking to Paul and others, about what they are doing with ONC.
So I am very comfortable moving forward with that because obviously CMS is
paying incentives and obviously ONC is driving the actually implementation and
the structure of this and we are in a spot to go ahead and step away from it
and kind of look at it across the entire landscape that we have. So that is it.
That is where we are headed. We will wait to hear from Jim but no more
succinctly than that. Please as we move forward it is very critical to do that.
So any other questions anybody has for Jim? Mark.
DR. HORNBROOK: To what extent can we push the safety educational perspective
that is — arms to patients for missed linkage, missed identifications?
MR. REYNOLDS: I think once we start a hearing on any of this, anything we
bring up about what the issues are that it creates if it is not done
appropriately, I think fit nicely in the purview. We have done that in every
other thing we have tended to do. And so I have no problem with talking about
the ecosystem around the fact that you have information that may or may not be
coordinated is exactly in play in all these subjects. It is just that diving to
an answer is not — well we are just not going to start with the answer. We are
going to start with the ecosystem and start with the issues and start with, as
Jeff said, best practices and some of the other things and what they have
gotten us to as we do it. Are there any other questions?
DR. FRANCIS: I just wanted to -–
MR. REYNOLDS: Are you going back there again? Do not go back there again.
(Laughter)
MR. REYNOLDS: We have made our formal statement as the Committee. It is in
the minutes. Please do not go back there again. I mean this is not a hearing
and we are not trying to create a hearing, but that is the last thing I do want
to say on this. Think about as your Sub-Committees, how you want to do this;
who wants to do it and who wants to own it? So what do we do? Just think about
that and as you come back with your readouts tomorrow, let us know what you
thought about as you got into your Sub-Committee about what this may or may not
be and who may or may not do it and then let’s go from there. Okay? But other
than that, I really want to shut this one down. I think we have put it in a
nice place, we have got Jim moving forward with it and I do not want to have to
restate everything that we just stated. It put us back in a place where we need
to be as this process goes forward. Any other questions for Jim on anything he
said? Okay. We are waiting for Tony. Tony, are you on yet?
MR. TRENKLE: Yes, I am on Harry.
MR. REYNOLDS: Oh great, perfect timing. Have you been holding your ears so
you would not have to hear any of this?
MR. TRENKLE: Yes, well I have heard some of it. Actually I picked up the
phone because it is a little faint.
MR. REYNOLDS: Well we were trying to talk quieter towards the end there.
(Laughter)
MR. TRENKLE: That would be a good idea.
MR. REYNOLDS: Words and tone so I think we are in a safer place. So we have
got Tony Trenkle on. Tony, welcome and we would love to hear your update and we
appreciate you joining us and I know that you are in a flurry of activity right
now so thanks for being able to share time with us.
MR. TRENKLE: Okay thank you, Harry and thanks and hello to all the Committee
and as you know there is a lot going on and so what I am going to do is just
kind of walk through a few highlights and then open it up for questions. Of
course the major issues on our plate right now are HITECH and of course the
Healthcare Reform Activities particularly from my office’s standpoint and the
admin simplification area.
With HITECH we continue to move the incentives through clearance working
closely with our friends at the Office of the National Coordinator. It has
received a lot of interest from the top levels of the Administration. We have
had a number of people at the Office of Management and Budget who have been
weighing in so there is a lot of interest in seeing where this goes and what
actually is in the regulation. As you know, although CMS is the “author of
the reg” we have many authors who assist us in writing the regulation and
as I have said to others, as we publish the regulation if there are things in
there that you see that you want to comment on, please comment and comment with
a solution in mind. We certainly welcome any kind of thoughts because this is a
fairly complex regulation and it cuts across several big programs –-
Medicare, Medicaid, Medicare Advantage, a number of issues that people have
raised, so any type of feedback is certainly welcome.
We are still on target to publish by the end of the calendar year. We
certainly think that we are in good shape to do that. And of course, I want to
thank NCVHS once again for the early work that they did in helping get this off
the ground with the April 27-28th hearings. I think it was a key
early marker that allowed people to really start assessing Meaningful Use and
what it would mean in terms of getting us to implementation in the next couple
of years.
Internally at CMS we are also spending a lot of time working on a number of
operational issues related to the implementation and getting ready for 2011. As
you can imagine, setting up a program of this size and involving Medicare and
Medicaid creates a lot of operational and internal policy issues and we have
been kind of working an iterative process with the reg recognizing we have to
make a lot of early operational calls even though we do not know what the final
regulation will say. And so that has been a challenge to our systems folks and
their program policy folks.
We have been spending a lot of time with industry groups. A number of them
have come to us with concerns over the regulation or what they feel or fear
maybe will be in the regulation. Just some of the major issues just to let you
know, none of these probably should be a surprise. One is how high do we set
the bar on first year Meaningful Use? We have gotten a lot of comments from
groups on concerns they have with some criteria they have seen in the HIT
Policy Committee Recommendations and concerns about whether they can implement
in the first year.
The second major area is how broadly we define hospital-based eligible
professionals. This is an issue that has raised a lot of interest and concern
from hospital groups who are concerned that their outpatient clinics will be
put under the hospital umbrella and will not be eligible to receive separate
incentive payments. They have expressed a number of concerns and thoughts about
that issue.
The third major issue that has gotten a lot of input and activity is Quality
Measures and particularly from the specialty community whether we come out with
some measures that would disadvantage specialists. A number of discussions, and
of course NCVHS has had them in some discussions about the quality area and the
HIT Policy Committee also had a couple of days of hearings last month about
that. So it is certainly a big issue.
And then finally a fourth major issue is the Medicare versus Medicaid
criteria for Meaningful Use. A lot of interest from the states and others
working across both Medicare and Medicaid wondering how much we are going to
harmonize these criteria. So these and other issues will be discussed in the
regulation. I can only talk about them in a limited basis today, but we are
certainly aware of them and a number of other issues that will be particularly
interesting to various segments of the community. So expect a broad and a deep
response to the NPRM when it goes down and probably as I said, definitely
before the end of this year.
In addition to that, we have certainly been very active in the HIPAA arena.
We put out an RFI in October asking for feedback on our transaction code set
enforcement and asking for suggestions for improvement. We have received some
feedback and would welcome more. The response period was originally scheduled
to end on November 16th and has been extended to December
1st. So anyone who has an interest in this area and certainly this
will also add with the admin simplification. There is going to be more on an
enforcement effort in that, so any kind of feedback you can give us now would
certainly be helpful to us both today and depending what comes out of admin
simplification.
We have also continued to move forward both internally and externally in
activities related 50-11 and ICD 10, internally for 50/10 our systems staff is
on schedule making the changes that need to be made in order to start testing
by January, 2011. They have put a lot of effort into this and we think that
they are in pretty good shape. Internally with ICD-10, we have completed our
impact analysis.
Our contractor, Noblis finished that up recently and it is now available on
the CMS website and I welcome people to take a look at it. What it does is
identifies our high risk areas. It also identifies the what we call opportunity
targets, which is one thing that Harry and I and others have talked about with
ICD-10 and the idea that it is not just an implementation, but it is also a way
to change how you do business and the time to begin looking at that is before
you implement, not after you implement.
We are also in the process of establishing a Program Management Office and
have established a multi-component ICD-10 Steering Committee. We are working
very closely with different parts of CMS during application development
sessions to really sit down with the various business owners over periods of
times and really get into how ICD-10 will impact their business processes.
Externally, we have been engaging with industry leaders and a number of
conversations to begin driving the implementation. We are looking at the NCVHS
hearings next month as a good marker to tell us where industry is at and what
some of the challenges are and the awareness level and a number of other
things. So I personally am looking very much forward to hearing the feedback
that comes out of it. I have seen the draft agenda and it looks outstanding.
We have also engaged a contractor to help us develop a National Outreach
Strategy for both 50-10 and ICD-10. In fact today we are beginning focus groups
with various providers to get feedback on their current state of awareness and
preparation.
We are beginning to conduct and we had another contractor begin to conduct
an environmental scan of readiness that we are going to use as a baseline over
the next several years and in addition to that, we will be conducting regular
scans over the next several years of 50-10 and ICD-10 readiness to pinpoint
areas of concern and tie to outreach efforts. So the outreach and the scan will
work together. Obviously we are going to work together with a number of
industry groups to get feedback and to help us with the outreach.
In the personal health record areas, we are continuing our two pilots that
we have been running for the past year and a half or so in South Carolina and
about a year now in Arizona and Utah. We have decided to extend them for
another year, so they will be ending sometime next year. The South Carolina one
is now scheduled to end in October of next year and the Utah-Arizona one in
December of next year.
We have actually completed an initial evaluation with the ASPE of our South
Carolina pilot and we will be beginning an evaluation of the Arizona and Utah
PHR choice pilot. And we will be working with ASPE to make these evaluation
results available on the web once they are finalized. There have been a number
of interesting observations from that.
Any prescribing, of course we are continuing with our pilot to test codified
sig and RXNorm, but the pilot went in December and then we will make a
determination on the potential for adopting one or both of these standards.
RXNorm is probably in better shape than codified sig, as you would suspect.
With the DEA controlled substances reg is now in clearance. HHS has been
working very closely with the DEA and hopefully we have changed a lot of the
concerns that people had with the inferran(?). We worked with the DEA to modify
or to soften the language, or at least get them to get requests for additional
comments when they go out in a final rule which hopefully will be out sometime
in the next several months.
We also have the NCPDP 10.6 in clearance now. That is to upgrade the current
standard and hopefully that will also be out in the next several months.
I mentioned briefly, admin simplification. We have been working very closely
with a number of the Committees on the provisions. As you have all seen, the
House version came out and we will be seeing the Senate version –- well we
have already seen the Senate version, and certainly there are some differences
between them and we are working closely with the Committee staffs and they have
been very receptive and very interested in listening, learning and supportive
even when they have not adopted things that we wanted, I think they have been
very supportive and probably, I have dealt with a number of Committees over the
years on different areas, and this is probably been one of the better groups, I
thought, to deal with because they have been very interested in getting
feedback from a variety of the stakeholders on these issues.
And of course, we know if this passes, there are going to be many changes in
HIPAA. We are already looking internally at the work plan that we will be doing
to tie this in with the other activities.
Finally, another area that will be of interest to the Committee is CMS Data.
The Administration has a lot of interest in looking at ways to make it easier
to get CMS data while still protecting privacy and security. I think most of
you are aware of the data.gov website. We have also been working very closely
with the OMB CIO, Vivek Kundra and some of his staff, on a number of ideas they
have for making CMS Data more readily available and there will be more coming
out on that in the future but there is a big priority from the Administration
to make that one of our key areas of emphasis.
So those are the main things that I wanted to mention today. I can certainly
take your questions on these or anything else that you have. Thank you for
giving me the time to speak for a few moments.
MR. REYNOLDS: Tony, thank you. I commend you for all that you guys are doing
and the leadership that you are showing in a lot of these things. It makes it
exciting because you guys kind of get down to where the practical is and the
real money is and you are really showing great leadership. So thanks to you and
your whole staff and we really appreciate the way that you guys support our
Committee. So with that I will open it for questions.
MR. TRENKLE: Thank you Harry, before you mention that, I do want to mention
as well that we are very interested, as you know, in making sure that the three
Committees work together. We brought that message with you to David Blumenthal,
and you know that Jim and I and others will continue to work to make sure that
the three Committees are in sync and that we capitalize on the strengths of
each of them and do not end up in a situation where we are not fully utilizing
the capabilities that any one of them has.
MR. REYNOLDS: Right and I think that is going real well so far and so I
appreciate that. Questions? Leslie.
MR. TRENKLE: Good.
DR. FRANCIS: These are two questions about the last couple of items. First
of all about the CMS data; are their privacy, confidentiality and security
questions that you have open that you would appreciate any help from us in
addressing? And the second question is just the more specific one; we have been
following along on the question of personal health records and in particular
the question of how they are going in the two demonstration projects. Have any
privacy and confidentiality issues surfaced since we last got a report from how
that was going on? It is relevant either to our ongoing interest in PHRs or to
the more general question of PHR regs that HHS and the FTC are charged with
bringing out in the not so far distance future.
MR. TRENKLE: Right. Well thank you for your questions. To answer the first
one of course Amy Chapper from my offices works with the Security and Privacy
Sub-Committee and we certainly have some interest in the whole issue of
re-identification of data. That is something that is of very much interest to
us because, as you know, one of the big issues when you start giving out data
is the potential for abuse or reuse or other areas or secondary uses of it.
There are a number of issues, as we all know, so we are very much aware of the
work that the Committee has been doing and we are working closely with Amy to
see if there are some issues that we will need to get the Committee to look at.
One of the things at this point is some of this is still evolving and it is
politically sensitive from the Administration’s standpoint so I, at this point,
I will punt on it until we get further discussions going internally because
some of this, our Office of General Council has gotten involved in with the
Office of Management and Budget’s General Council and so there are a number of
things that need to be worked through. But yes, at some point, I would like to
sit down and discuss ways that the Committee can be helpful because I think
there are a number of issues that will be helpful to get further help on.
The second issue you raised on the PHR pilots, at this point they are still
very much pilots that we have not put a whole lot of outreach into and so they
are small pilots. The one in South Carolina is a contracted pilot, so we
actually went through FISMA testing before it went up and the contractor, the
health trio who was the HR vendor, had to go through a number of changes to be
eligible to launch it live. So we have continued to monitor that and continue
to look at ensuring that they are meeting the requirements that we set up
initially. Of course one of the issues, in fact, that came out of that is the
changing of the passwords that has to occur every 90 days which has caused a
little bit of concern. It came out in the evaluation study but, obviously, it
is something that we were told that we needed to do in order to meet the
security requirements.
The HR choice one is one where we do not have a specific contract with the
four vendors who are participating, so we had them all sign data use
agreements. And last year when we set this up, we had a number of discussions
with Google where we had a number of issues related to reuse or data and other
types of security issues that were raised in terms of our data use agreement.
So we did get those resolved and we continue to monitor those efforts as well
through our data use agreements. As I said, once again, the pilot evaluations
also ask the users about privacy and security and we are getting some results
from South Carolina and I am assuming the Utah-Arizona evaluation will be
asking similar questions and so we will be getting some feedback from those
participants.
DR. FRANCIS: Thank you.
MR. REYNOLDS: Okay, we have got Larry, Tony and Jeff.
DR. GREEN: Tony, this is Larry Green. Thank you very much for covering so
much so fast. When you went by the ICD-10 update, if I heard you correctly and
I am not sure I did, you made a comment about how it actually was changing
business. I am just wondering if you could elaborate just a little bit more
about what that means?
MR. TRENKLE: Okay, what we have asked our contract to do was to target
various opportunities that you can realize from ICD-10 because of the greater
granularity of coding information. For example, in the quality area, because of
the fact that you have greater opportunities to look at more detailed data, you
can achieve potentially greater information in terms of improving the quality
of care. And in the payment area there are certainly opportunities to look at
greater accuracy in payments because of the fact that you will have the greater
granularity and the ability to track payment data much more closely. So those
are two examples.
MR. REYNOLDS: Tony.
MR. RODGERS: Tony, this is Tony Rodgers. Is there anybody looking at or is
their consideration regarding the dual eligible population and data
harmonization between the state Medicaid and Medicare data? The reason why I
bring this up is that based on health reform, we may have an additional dual
eligible population which would be the SCHIP kids, who Congress has been
talking about allowing them to be wrapped around in terms of Medicaid and the
CHIP Benefit Package and so is anybody looking at the dual eligible population?
MR. TRENKLE: Yes they are. In fact, Cindy Mann who runs our Medicaid shop,
is very interested in that and as soon as I leave this call, we actually have a
new group forming within CMS to look at ways to improve the use and accuracy of
both Medicaid data and one of the areas, obviously, that they are looking at is
the whole area of dual eligible’s. And of course that has come up on the
providers side, eligible providers in the HITECH Program, so yes, we are
looking across the programs; the Medicare and Medicaid and of course the CHIP
work as well and so all these are being looked at from the data management
perspective and also from other perspectives. That is an area that is very high
on Cindy Mann’s priority list.
MR. REYNOLDS: Jeff.
MR. BLAIR: Hi Tony. This is Jeff Blair.
MR. TRENKLE: Hi Jeff.
MR. BLAIR: We are going through some really exciting changes with Meaningful
Use of health information technology and at the same time, we are also making
the transition from ICD-09 to ICD-10. Is CMS looking at how to ease the way,
not just individually, as there are a lot of organizations that are going to be
going through both transitions at the same time? Have you looked at either ways
to ease the burden of both at the same time or maybe even, and I do not have a
suggestion for this, to see if there is some way that the processes that are
put in place for one transition might be helpful to an organization as it goes
through the parallel transition, to see if there are synergies that could ease
their way?
MR. TRENKLE: Yes we are obviously aware of the fact that there is
significant areas of implementation that need to be done over the next several
years and that is even before we have healthcare reform and what they will
impact. We are looking at the various areas and how we can do some of the
synergy, both internally and we also have had some discussions with others
outside of CMS, but we recognize that this is an issue and we need to be
cognizant of what some of the ways are that we can build some synergy between
the two efforts. I mean there are several things that that involves. One is
obviously what are the systems impacts of doing these types of changes? The
second is from a business process standpoint, how is that impact the changes
being made and how they will impact the business processes? And then the third,
of course, is from our perspective, how do we work in terms of the outreach for
the HITECH and tie in the outreach efforts we are doing with ICD-10 and 50-10
and how we can work with the provider community and other areas to make sure
that this gets done as seamlessly as possible. But we recognize that it is
going to be a challenge as it is a challenge to us internally to be doing a
number of these implementations together.
MR. REYNOLDS: And I think another thing that they have done is that they
have produced an awful lot of collateral on ICD-10 while people are still early
in 50-10. So there is enough on the street. Usually you would think we would be
a little further along in 50-10 before we start hearing about ICD-10. I think
that the effort that they made and the actual collateral that they have
available to the public right now allows you, as you are doing 50-10, to think
of all the ICD-10 things you ought to consider at the same time. So I think
that is why I have commended them a number of times to be out front this far,
because you would think we would be a little further along in 50-10 and then we
would start hearing about ICD-10. But no, they are putting them both right out
there at the same time. Now there is still a lot of work, but I think you know
moving that up probably about a year, if you really looked at the normal flow,
moving that stuff up probably about 12 to 15 months earlier than you would
expect it with both of those being linearly implemented, I think has been very
helpful to the industry. But it is still a lot of work and all of us know that.
Are there any other questions or comments for Tony? You do not mind getting 15
minutes back to go work on that reg do you bud?
(Laughter)
Mr. TRENKLE: There is always plenty to do here.
MR. REYNOLDS: Yes, that is right. Hey, Tony thanks a lot pal and good luck
with everything that you are doing and we will be talking to you. Take care.
MR. TRENKLE: Alright, we will see you all.
MR. REYNOLDS: Okay it is 10:11. We will break until 10:30 and then we will
be ready to roll on the letter, which will get us started a little bit early on
our Quality Letter. So see you back at 10:30. Thank you.
(Break)
MR. REYNOLDS: Okay our next portion of the program is the discussion by the
Quality Subcommittee and it is chaired by Justine Carr and Paul Tang. I think
Justine is going to lead the discussion initially. This is a letter that we
would like to be able to finalize at the latest tomorrow. So, Justine let me
turn it over to you please.
Agenda Item: Quality Letter on “Meaningful
Measures” – ACTION, November 20th
DR. CARR: So what we wanted to do was start with the big picture of why this
is coming forward and what we heard at the hearings and how we got to the
letter and where it fits in the big picture. So I am going to just start with
the charge of the Quality Sub-committee which is included in the letter, but a
key focus is to monitor the nation’s health data needs and current approaches
to meeting those needs, identifying emerging health data issues, including new
methodologies and information system technologies, data bases and networking
that could improve those needs. And a second area of focus is to identify
strategies and opportunities for evolution from single purpose, narrowly
focused, categorical health data collection strategies to more multi-purpose
integrated, shared data collection strategies.
So what we did was following up on the fact that we are nationally focused
on Meaningful Use, and we wanted to now address meaningful measurement of
quality healthcare through electronic medical records. In convening the
hearing, with tremendous help from Matt Quinn I might add, we asked speakers to
address four questions.
One, how do we approach building meaningful measures? What is the current
process for developing measures and does it adequately address measure
development for key national priorities and sub-populations. Three, how do we
introduce new data sources, clinical data from EHRs, user-generated data, et
cetera into the measure development process? How do we exchange them for old
measures based on administrative data? And then fourth, how do we maintain and
update measures and what are the health IT system implications?
So, because you all know how much I love grids, I just wanted to let you
know that I have moved on from grids to circles and so I am going to just take
you through this. It is colorful and it has stages of development.
MR. REYNOLDS: For those of you who cannot see it you would not understand it
anyhow.
(Laughter)
DR. CARR: Please.
DR. TANG: No offense, Justine, but it still looks like a grid.
DR. CARR: Well the yellow hue is not part of what is up on the screen.
MR. HOUSTON: I think it is the influence of Paul Tang and his lumping.
(Laughter)
DR. CARR: I am here to learn. That is all there is.
MR. HOUSTON: Moving right along…
DR. CARR: So I think that I have color-coded this into three components of
quality measurement development.
MR. BLAIR: Thanks for the color-coding. It is now all clear to me.
(Laughter)
DR. CARR: Okay. I am going to now describe each color and its relationship
to the concept. So at the beginning of the circle is the topic that I call
Clinical Decision Support and Knowledge Management. And so in order to do the
right thing, we need to know what we are going to do. So what is the evidence?
And then, are there guidelines? And then, can those guidelines be embedded in
our electronic health records for care delivery? So that is one component.
The next component, as we think about EHRs, is how do you capture that data
element in the EHR with a uniform definition? How do you store it in the EHR in
a field that is queryable?
Next, how do you retrieve the data element from the EHR? And then, how does
the data element get aggregated with the other data elements to become a
measure? So those are the pieces that are within the EHR. Do we have the
information we need? Do we have the definitions? And can we get it out?
And from there, once aggregated, that data then becomes reports and this
becomes measure development. So you take the data elements and you create a
report and from the report you act on the report and then, hopefully, care is
improved leading back in to further evidence for greater improvement of
healthcare quality.
So I did put this together because this is how I have to think, but it is
helpful in many ways because when we had the hearing, and even in our own
discussions here, we have experts; we have, for example, Blackford who is an
expert on clinical decision support and Paul who is an expert on measure
development and others understand the EHRs and the standards and how data
elements get in there, SNOWMED, ICD, problem lists, lab data, and codes. So
sometimes when we have our quality Sub-Committee meetings, I come away confused
because we have been talking about all of these things and they really need to
be parceled into each topic.
So with where we are today, about to have a comment period on Meaningful
Use, we felt that it was most important to focus on the EHR and the capture of
data elements that would bring us to having building blocks for subsequent
measure development. So when we heard our speakers, we actually heard on all of
these topics and, in fact, each of these has a whole array of issues. So for
evidence, do we have it for specialty care? Do we have it for special
populations? What level of evidence is okay for guidelines? Are they
machine-readable? Are they EHR friendly? For evidence based care delivery, how
do you take that evidence and that guideline and have it in the moment, so as
you pop in the problem list, “heart failure” your “heart
failure” guidelines come up.
Moving on into the data elements, we have issues like, for data element
capture in an EHR with a uniform definition, who decides on the data elements,
the sources, the definitions, and the taxonomy. Again, putting data into
queryable fields, who educates the user into how to get that data in there in a
way that is easy for the clinician and easy to retrieve? On the retrieval, how
do you validate the completeness of the data? Are there missing fields? Was
something inappropriately, inaccurately captured? And how do you validate that
your query was even a valid query?
And then, with the data element aggregation, this was a key thing that came
out, because we think about the EHR and the physicians and nurses and physician
assistants and clinicians putting their data in, but when you want to tell a
story, you may have your data elements in the clinical registry but you may
need additional information. Their zip code, their address, other information
that is kept somewhere –- so who is the person who brings different data
sets together to tell a story. And then we heard a lot about field testing and
endorsing and I will not mention that word down there.
And then finally, after you have done all of these things, now on to the
measurement development. So when you generate a report, also who is doing risk
adjustment and who is setting the benchmarks? Are we striving for the highest,
for the middle or for the lowest? And then, we also heard, when the report
comes out and you receive a report on your practice, how do you drill down? We
heard interesting information about the American College of Surgeons doing
drill down on who are the people that are not compliant with their diabetes
control? And it was very different from what you would have assumed from the
rolled up data.
And then, importantly, as we make these public measures available to the
general public, how do we educate them so that they understand what it means?
And then, finally back to the thing, care improved? Who says? You know, what
are the criteria that care is improved? So I wanted to just set that stage to
say that this is the beginning of a journey, as Harry would say. Probably for
all of us and certainly for the quality work group, we will be talking today
about these data element storage issues. We will then be coming back to some of
these other issues. I would also say that this starting point is the EHR focus,
in many ways, and there are interfaces with privacy and security. We have
talked in the past about how you sequester certain data fields? As we get it
more structured, that conversation takes on new meaning.
Further we are going to talk about population health and we know that this
is not the universe of the information that you need, but we heard at the
Meaningful Use hearings about how valuable to public health it would be if
everyone had immunization status and how that could move things forward. So
this is probably one circle in a gigantic gear, perhaps moving forward, but I
hope it is helpful in setting the stage so you know what we are talking about
today and why.
So any questions?
DR. MIDDLETON: Just a suggestion, Justine and I love it because it makes it
very visual, very graphical and connected. I am a little worried about sort of
the tar pit or sand trap affect of the yellow background. I am just kidding.
But one thought would be on the right hand side, not only for symmetry but to
disambiguate clinical decision support and knowledge management, I might
suggest there to be two bubbles on the right.
DR. CARR: Okay.
DR. MIDDLETON: That is, the right upper quadrant and kind of the evidence
synthesis and guideline development and the right lower quadrant then would be
really CMS implementation and knowledge management.
DR. CARR: Okay, great. This would be iterative. It has been for me each time
I try to address this, we get more. And certainly these little bullets of
issues raised do not begin to encompass the depth of issues that need to be
discussed, but it is a starting point.
MR. REYNOLDS: One other thing I would like to point out, again the reason
this is important as Paul and Justine and everybody that has led this, is that
Tony talked about our synergy with what ONC is doing and so on and then you
heard Jim talk about paying incentives in 2011 for Meaningful Use. Well we all
know that Meaningful Use is going to be driven out of a lot of this EHR
adoption. And so that is why it is very important to be in that space quickly,
being able to put together the first installment of this journey, that then can
be moved on to Paul’s efforts and others in the ONC effort, as they are moving
forward to go the next level down as to exactly what is going to happen.
So that is the reason that diving into that chart, at that particular point
right now, is probably as important as anything, but yet drawing the whole
picture says this is what we have to do, you know, as an entire story and we
are not going to step away as NCVHS from the entire story, but we have to make
sure that our pieces of the journey are relevant at the times that those are
taken up to build a much bigger story. And so I think that is another way to
think about it. It is not that one is more important than the other, it is just
that the timing of some of this and the synergy of some of this, with as fast
as everything is moving, is extremely important. Paul.
DR. TANG: I might expand a little bit on that. Mark Smith gave a keynote at
AMED earlier this week, and one of the points he made is there probably is more
health reform in the Recovery Act than actually the Health Insurance Bill. And
one of the ways that get enacted is through the whole Meaningful Use really,
because it is laying an infrastructure for reforming the health system and how
I think we got it started in the quality Sub-Committee, is that the measures we
currently have are not the right tools to reform the health system. That is how
we got into the meaningful measures that Justine talked about. I think that
this really is our opportunity.
One, to work with the Policy Committee because the Policy Committee is not
in the business of developing measures and neither are we, but it seems like we
have been in the business of trying to influence measures that would be more
meaningful for health statistics, et cetera, but also to reform the health
system delivery of care.
So it seems like a really nice fit for NCVHS working with the Policy
Committee and for the context of the Recovery Act. So it is just a beautiful
context and also you had a beautiful circle.
MR. REYNOLDS: Are there any other questions on the circle? Are going to make
any other comments and then we will start the letter.
DR. CARR: I want to just say that it was our expectation that we would come
out of the hearing with some great examples and things to put forward, but we
were in short order, struck by the fact that there were many excellent
initiatives all going in different directions. And there was no anchor holding
them together and putting them back. So again, that is why we can back to if
EHRs are more prescriptive about what elements need to be in there, it helps
the vendors and it helps the measure developers, and it helps the clinicians.
To be super concrete, I just added this other slide. Again, many of us live
in the world of these quality measures and forget that not everybody does to
the extent that we do. So I am just giving you the example so that if we said,
what would be some data elements, a problem list, a visit date, a lab value, or
a medication would be examples of fields. So you might have diabetes and it
might be in SNOWMED code in the problem list and you have the date of the
visits or the dates or frequency of when the person came, the A1C value and
their insulin dose. And the next thing in matching that, you could tell very
many stories about how the person is doing or how long to achieve control and
so on. So other factors even, as I said, aggregating data then to zip codes or
other things might tell you about adherence to medications in different
settings. So shall I leave this back up or are we done?
MR. REYNOLDS: You can leave it up. Well you are going to go through the
letter. But one thing I would like to comment, since Marc Overhage is not here
and some of you may have seen the communication and others not, as Justine put
that particular piece of the chart, and understand I am not making his input
bigger or different than anyone else’s, he is just not here and so I would like
to go ahead and put it up front, so that some of you that saw it are at least
aware of what it was. So as Justine showed that chart a minute ago, and showed
values, one of the comments was, are these values that are readily captured? Or
are we talking about changing the work in every doctor’s office to capture new
information. So that was just a point that was made. As we move the letter
forward, I would not want Mark to show up tomorrow and not have had that on the
table, at lease for discussion, so that then we turn around and have to then
redo it again. I am not voting for it; I am just saying he is not here.
DR. W. SCANLON: I am not sure that you captured his whole —
MR. REYNOLDS: Well then please –-
DR. W. SCANLON: Since we had an interchange, let me talk about that
interchange. Mark did raise that point about this issue of are we implying that
physicians are now going to have to collect new information and have it in
their EHRs in order to satisfy these kinds of requirements? And my response was
no that that is not what our anticipation is. What we are thinking about is
that we want to, and to make it concrete I will talk from a program perspective
and let’s think about Medicare, from a Medicare perspective we want to be able
to take advantage of information that is already being collected as part of
normal medical encounters that we cannot take advantage of right now. We are
handicapped now by the limited amount of information that we have.
And so we are not proposing, and my example was assessment instruments, we
are not saying the EHR becomes an assessment instrument and every time you go
and have a visit with a physician that this assessment is filled out. No,
normal information is collected and normal information is collected in a
retrievable fashion. Those are the things that we are interested in and
hopefully there would actually be some economies on the part for the physician
from EHR in terms of their time. The information that is collected is collected
once. It is easily retrievable. We had Carolyn Clancy, at the hearing, talk
about physicians saying, let’s just review your whatever, and that that was the
euphemism for, I cannot find this in the records, so would you tell me again
what you told me last time?
(Laughter)
We are actually thinking about ways that we can increase efficiency as
opposed to sort of creating a burden.
MR. REYNOLDS: Tony and then Joan.
MR. RODGERS: Yes, one of the challenges is going to be that when you are
thinking of an EHR, you are thinking of a record that is developed from many
sources of information. And data harmonization is going to be a challenge. For
example, ethnicity – if you go into eligibility systems, ethnicity is defined
many different ways and if we want to look at health disparities, we are going
to have to figure out how to make those definitions standardized across
administrative systems, et cetera.
But I think that the important thing to understand is that the physician
EHR, that is the point of truth, the point where the physician is using that
information for the treatment of care. He has grabbed data from the hospital,
from previous laboratory information, previous medication history information,
even age and sex and everything else, that data now is really not something the
physician, has necessarily, validated the data from the outside.
Within this process, the question I have is at the point where we are
grabbing the data from the physician, are we asking the physician to validate
it? To say yes, I believe this information is correct, even if it has come from
EHR, come from a hospital source; come from another physician source, et
cetera. If not, then who or does it matter? I think it does matter, but that is
one of the questions physicians are going to ask. Are you telling me that the
data that I have pulled in, right to the EHR, that I now have got to validate
what I am reporting to you in terms of all the things that have happened during
the continuative care that have happened? Will the physician say that? That is
the question?
MR. REYNOLDS: Okay I have got John and then Judy and then Bill. Again,
Blackford, I brought this up just to make sure that this point was discussed
because as we start this letter, let’s don’t redo it three or four times.
Because that is the one issue that as I have watched the flow, is out there.
MR. HOUSTON: Recognizing Harry’s comments before about how this is a journey
and this is only part of the journey, this talks about the collection of data
elements and the letter does not go into discussion about who is ultimately
responsible for actually extracting those data elements and how they are going
to be used. I know from experience at my health system that in various areas
there is an enormous burden associated with providing data to various agencies
or organizations that are responsible for doing quality measures. And so there
is sort of this latent concern, I suspect people will have, that okay this is
great, we collect this, and then what is the next requirement that I have to
comply with associated with that?
And then a second point that I think is important here is if in fact the
intention here is to make this as straightforward and the least burdensome as
possible, at least to collect the data, you may want to make some type of
statement in here to indicate that you are not trying to include or create
additional burdens on the physician, but rather to try to develop a system
whereby the collection of data is seamless, that it does not create burdens for
the office, but at the end result, you get these things. It might be just good
to state that explicitly because I think that will probably remove objections
or concerns that people may have.
MR. REYNOLDS: Again, I have Judy, Bill, Blackford and Paul. I am trying not
to get us off here.
DR. TANG: So this is the beginning and not the end. And I would say that the
main motivation for the hearing and the recommendations in the letter are to
avoid the burden; because right now, the burden is too high and we get nothing
out of it.
MR. HOUSTON: Can we just state that?
DR. CARR: Yes well John, when we go through the letter in describing the
current situation, we say this situation not only impedes creation and adoption
of valid and comparable measures, it increases the burden on providers that
most report different quality measures to different agencies and health plans.
DR. TANG: So let me just acknowledge that that is the reason and we just
need to fix how we say it.
MR. REYNOLDS: Okay we have got Judy, Bill and Blackford. Then that is it.
Let’s read the letter. I just wanted to make sure this was on the table because
I know Marc will be joining us later and I do not want to have to go back
through it again. I want to be able to say to him clearly that he was not here,
it was said, game on, and move on.
DR. WARREN: So my question is, and this may be where Marc is coming from, is
as we saw the examples about data elements, its granularity of data elements,
because medication: insulin only tells me insulin. It does not tell me route,
it does not tell me dose, it does not tell me anything else about the insulin.
It does not tell me any conditions that it is taken under. When I am looking at
some of the others, we have the same things, so some of what he may be
concerned about is how we can build out these systems that will collect the
data that we can then aggregate accurately with that. And so, where does the
buck stop in doing that? Also in reporting hemoglobin A1C, in the system that I
work with, I can build that out about three or four different ways and I would
not be able to aggregate those four ways.
DR. CARR: Can I respond to that?
MR. REYNOLDS: Yes, you are standing up, you should answer.
(Laughter)
DR. CARR: So that point is well taken. I think that underscores exactly what
we are trying to do. We do not want to know was the A1C less than 9, greater
than 7, less than -– we do not want to know that. We just want to know
what it was. And then the second thing is to your point about the granularity
of the data, the point was made by several speakers that we have two audiences.
One is, and what we are focusing on right now is public reporting. Within each
practice or institution there is performance improvement going on all the time.
And that level of granularity may well be necessary to focus on a specific
thing. But by helping us focus on public reporting data set, quality data set,
building off the NQF, it is a starting point. It does not preclude adding
greater detail within a particular –-
DR. TANG: So I think we did not do a good job setting up the context well
enough. So instead of worrying about the details of what needs to be created
for the purpose of reducing burden and increasing effective use, we think that
is sticks out like a sore thumb that no one else is worrying about its use and
how to make the supply chain deliver on the goods. So this first letter is to
say that message, which is right now it is an enormous burden on a lot of
people and there are silos of activities going on, all well-intentioned, but
without any organizing context. So no one is worrying about the problem you
mentioned. And that is why our first approach is, we need to have a framework
and some way of providing oversight and coordination.
DR. WARREN: So we probably need to be sure it is said that way in the
letter.
DR. TANG: Exactly right.
DR. WARREN: To keep people like me from jumping down to what is -–
DR. TANG: So we got the message that we did not do a good job of saying that
is our main message.
MR. REYNOLDS: All right. We have got Bill and then Blackford and then,
again, there is going to be plenty of discussion on the letter. I wanted to
kick this off because now we have got a context. We will go through the letter.
Ask any of your questions during the letter and ask any of your issues. I just
wanted to make sure we had a good context.
DR. W. SCANLON: One of the things that I think that emerged in the hearing
was that there are internal uses and external uses for these data. And that, in
some ways, how one meets or satisfies those uses can be different. And this, I
think, goes to Tony’s point about, is the physician going to be responsible for
all the information? For external purposes, I think the most efficient approach
is that we do not rely on one source. Now we have information about this
individual coming from multiple sources and the external user is the one that
puts that together. Because, in some ways, the physician may be in an
impossible situation in terms of getting some essential information and we want
to use that when we create measures and when we do analyses.
And so in terms of reducing burden, we need to think about the data strategy
that satisfies our external needs as independent, to some degree, from the data
strategy for internal purposes, because there may be certain information that
is important for an internal purpose of quality and improvement that has to be
collected and maybe it is sort of different from the routine, but we still
think that it is going to be important for the future.
DR. MIDDLETON: I think a couple of thoughts have already been discussed
partially in the area that I think we are attempting to answer or address in
the letter. As Paul suggested and Justine said, it is a letter to start a
process. I think there are a couple of points that I think we might just want
to dwell on for a second and think about the language. It is clear that the EHR
is viewed as the data source, but the concept of EHR, of course, is different
everywhere you look.
Original CPRI and NCVHS language would suggest that the EHR is the system of
systems and all the rest of it, and to Bill’s point, therefore data will come
from a wide variety of places. But in terms of the data validation and the
process of these data acquisition to make it as much as possible, a by-product
of care with zero or negligible burden has been a goal since we all began
thinking about this.
On the validation front, I tell you what we all do anecdotally is if it
enters into the record from whatever source it is validated as part of the
clinical process of data review and authorization or attestation for its use in
the record. And that includes patient source data. We understand those data are
different and in a wide variety of ways, but if they are reviewed and
authenticated or attestation occurs and they can go into the record and be used
for decision support and quality reporting and all the rest of it.
The last thing, I think, and we actually did not talk about much, but it may
be useful to think about kind of the evolving context of EHR too. I think the
message of this letter is appropriate. I also think the landscape of EHR in
terms of its widening array of solutions which will be called the EHR, modules,
components and services, will make the gathering of these data different down
the road — so just a thought there about where is the technology landscape
going that we have to address.
MR. REYNOLDS: Our last comment and that is it. I am sorry. We have got to
get on to the letter.
MR. HORNBROOK: I just want to caution us about falling into an
unconscientious trap here that data are commodities. That once they are
created, they are fixed. You may have trapped yourself into a place you do not
want to be.
MR. REYNOLDS: It would be us.
(Laughter)
MR. HORNBROOK: I think you need to think about the question of data views
where what you are actually doing is setting up a program to look at the data
at one point in time, because the data that you are really looking at keeps
changing. And you want to use the information but you do not want to think of
having the data and then having it be in your command and control when actually
real information has now changed and it is different in the original source. So
I think if we fall into the sense of telling the secretary that we want to set
up a physical repository of commodity, we might not be doing ourselves a favor.
MR. REYNOLDS: Okay, let’s turn back over to Justine and Paul and Matt. And
Matt, I echo the thanks for everything that you did on working on this. And
let’s go through the letter and then everybody has kind of got a little
different context and we have heard some passions around the room so let’s go
at it.
DR. CARR: All right. Well I think what we will do is read through it, but I
am going to skip the first paragraph because I already just read it, which is,
what is the role of NVCHS and the Quality Sub-Committee? So paragraph two then
begins, the National Committee recognizes that the Health Information
Technology for Economic and Clinical Health Act, the HITECH Act, provides
substantial financial incentives intended to accelerate the adoption and
meaningful use of electronic health records. As EHRs become widely adopted,
there is an opportunity to measure healthcare processes and health outcomes in
new ways that can contribute to improved population health and higher quality
healthcare.
However, to capitalize on the EHR investment, the EHRs need to be designed
with quality reporting requirements in mind. For example, EHRs need to be
designed to capture relevant clinical data according to standardized data
definitions and quality measured definitions so that comparative quality
reports can be generated. Let’s stop there for any comments. Judy.
DR. WARREN: I just have one. Up on our tag line, can we change it to
“electronic health records?”
DR. CARR: Oh yes, right. The re: is “meaningful measurement of quality
healthcare through electronic health records” as opposed to medical
records.
MR. HOUSTON: Jeff said “through or based on” in the tag line.
DR. CARR: Either way. “Using”.
DR. TANG: Either “using” or “based on”.
DR. CARR: “Using”.
MR. REYNOLDS: John.
MR. HOUSTON: Yes, this is based on earlier discussion. The sentence that
starts, “However I would capitalize on the EHR investments, the EHRs need
to be designed” –- I think what we should probably say, “is to
seamlessly collect information in order to provide quality reporting.”
Something like that that speaks to the fact that really what we are trying to
do here is ensure that there is that collection. And if you use a word like
seamless then what you may end up with is to raise the point that this is
really not intended to be a burden.
DR. CARR: So do you want to make a recommendation here? What are the words
that you want like to see?
MR. HOUSTON: What I think I said was, “However to capitalize on the EHR
investments, the EHRs need to be designed to seamlessly collect data necessary
for quality reporting requirements” or something like that, or transparent
or efficiently. I am just trying to get –
DR. TANG: There is still body contact with the user and the system and at
least we think it is embedded in this concept that they have to design these
systems with that in mind.
MR. HOUSTON: I think going back to one of the things Harry has always said
to me and I think he said in the context of this meeting too, this is sort of a
step-wise process and the first step you are really advocating here is the
collection of data, right?
DR. WARREN: It is different. I mean I want to put my designer hat on. When I
am designing data collection forms and stuff in the EHR, I am designing with
the use in mind. And one of those uses is retrieval because if I do not design
with retrieval, you are never going to get there. What you are talking about is
the process of using the EHR, not the design of it. I think we need to focus on
the design for this letter.
MR. HOUSTON: I understand that.
DR. CARR: I think you are right because what you are raising actually is a
huge issue. I mean we could not say it in one sentence. It would be many
different dimensions. So my inclination would be to stay with things.
MR. HOUSTON: That is fine. I like it.
MR. REYNOLDS: All right, next paragraph.
DR. WARREN: Okay. And we thank Larry for this, to simply state clinical
quality is evaluated through measures which are composed of data elements.
Standardized data elements are the critical building blocks of current and
future measures. In order to understand the current activities related to
quality measure development and use, and how they relate to emerging use of
EHRs, NCVHS conducted a hearing on October 13 and 14 entitled Meaningful
Measurement. Testimony focused on four questions. Do I need to reread those
four questions? I read them earlier? I should reread them?
MR. REYNOLDS: Yes, again for purposes of people on the phone.
DR. CARR: Question one, “How do we approach building meaningful
measures?” Question two, “What is the current process for developing
measures and does it adequately address measure development for key national
priorities and sub-populations?” “How do we introduce new data
sources? Clinical data from EHRs use their generated data, et cetera into the
measure development process. How do we exchange them for old measures based on
administrative data?” And finally, “How do we maintain and update
measures and what are the health IT system implications?”
DR. WARREN: I just had a question on number the three. You said
“clinical data from EHRs.” Would this include data that is
contributed to them through various devices?
DR. CARR: It could mean that. I mean these are the questions that were
published for the hearings. We can talk about them but it is just a recap of
the questions that went to the testifiers.
MR. REYNOLDS: Leslie.
DR. FRANCIS: This may be a question that does not really work for you, but
one of the things that continually worries me about how EHRs are getting
designed is the question of whether they are missing things.
MR. HOUSTON: Whether what?
DR. FRANCIS: Whether they are missing things; serious things that might
matter for quality. I guess the question is did people talk about the question
of how we design meaningful measures in such a way that we do not lose
imaginative connections?
DR. CARR: I would say we heard some remarkably imaginative plans and in fact
that is what brought us back to our concerns that the complexity of what people
were envisioning, while elegant, would be difficult to implement. So I hear
what you are saying though. Are we missing things? And I think that there is
the sense that we are missing things in many ways where we are collecting it
inadequately. And to Judy’s point, we are collecting hemoglobin A1C with less
than this. Well we are missing the opportunity to say what it was.
Alternatively, we look at our national priorities and we realize we do not
collect anything about healthy people, initiatives, and BMIs and even a
standard data element in many EHRs. So you are absolutely right, we are missing
lots of things. But I think again today is to say for the work that is
underway, are we collecting it in a way that we can use it? They will be
further as we move into measures and put this against national priorities,
there will be lots of things that we want to collect and it will get back to
the evidence too, you know, what we need to do. So you are right, but I think
that is out of the scope for this letter.
MR. REYNOLDS: Blackford.
DR. MIDDLETON: I would just like to offer the observation, I guess, that it
is not subtle at all, how powerful measures are on EMR designers. If you look
back over the years when NCQA HEDIS came out, EMRs had to do HEDIS. When PQRI
has come out, EMRs have to do PQRI, and there are just many wrath examples of
that. I guess the challenge is going to be to find measures crisply and in a
way efficiently that do not add to this burden problem, because that is the
main crux of the issue. But a crisply defined measure that actually can draw
upon EMR data should be no extra effort.
MR. REYNOLDS: Tony.
MR. RODGERS: Yes. It sounds like with the EHR you are really dealing with a
patient quality measure versus a provider quality measure which is different
than some people look at quality measures. We are measuring what the provider
did in terms of quality they provided. With an EHR because you are getting data
and treatment goes beyond just that provider in many cases, and you are going
to be measuring that especially for the population, we have a patient-centered
quality measure that the physician is an actor in. And I think that changes the
nature of how the physicians will look on their role because it is multiple
physicians, multiple providers that have a role in that process. And so it is
not just on the provider in the office. It is on all of the providers who are
providing continuative care.
And by extension, there is a responsibility of the patient. And the reason I
bring that up, as we think about it, the patient is going to have to help
validate the information. The reason why that is important is because when we
exchange data, we find patients who tell us the data you are exchanging, even
though it is them, was incorrect. Now why it is incorrect, whether they shared
their card with somebody, whether somebody else is using their name, we do not
know but that has got to be part of how we think about this process of
validating patient-centered measures versus provider quality measures.
MR. REYNOLDS: Okay. Let’s continue. Did you have a comment, Paul? And then I
want to continue.
DR. TANG: I guess the comment is that I want to extend what Blackford said
because I think his point was very important which is why it is so important to
this Committee and this project, and that is you know how you say you can only
improve what you measure, well we are limited in our thinking about health
reform by what we measure. So there are a lot of studies coming out right now
saying oh, you know EHRs do not make a difference. Guess what, they are using
the administratively defined quality measures to measure. Of course it is not
going to change. We are not even measuring what we do. And so this is causing
the data and the evidence for the lack of change to perpetuate. Do you
understand what I am saying? This is so important to the reform effort so I
hope we helped to capture that notion in this letter.
DR. CARR: I just wanted to clarify with Tony. Were you recommending a change
or any language to be added to this?
MR. RODGERS: Is it assumed to be quality measures on the health delivery
system or is it assumed to be patient-centered quality measures? To me there is
a difference.
DR. CARR: Right. It is interesting because we did hear on both actually.
DR. W. SCANLON: And in some ways we are having a discussion about background
because the hearing was about measures but a letter has evolved. In the letter,
ultimately, the recommendations are going to be mostly about elements okay. And
the idea is that it allows either; that you can look at this at a patient
level, you can look at this as a provider level and you can look at this as
sort of a system level even for how a plan and the providers that work with
that plan perform. And it kind of deals also with Blackford’s point which is
that you do not change things when EQRA comes out because what you have got is
the data elements that do HEDIS, PQRI, PQRI2, 3, 4 and on and that is what we
are trying to anticipate in terms of an EHR that has efficiency built in for
over time.
MR. REYNOLDS: The reason I am letting some of this discussion happen, I know
some of it is background and it is discussion, but we have got to move. Not
everybody was there. Not everybody has read all the emails. I want to make sure
that the whole Committee is in context because as we fast track some of this,
if we do not have everybody walking along with us and getting the chance to
play, even if this particular letter does not answer everybody’s expectations,
we have got three or four more coming fast. The point becomes –- I want to
make sure we have got everybody going with us so that we do not all of a sudden
have somebody hit the brakes on us and we are going, oh well, we did not talk
about it.
So that is why I am dealing with this a little differently. We usually come
in, read the letter, the Sub-Committee did it, but the problem is we are about
to go through a number of these pretty quick and we just want to make sure that
we have got everybody in the game, playing the game, and asking their questions
and understanding their context. So that is why we are doing this one just a
little bit different, so next paragraph.
DR. CARR: “We also heard testimony about the work of the AHRQ funded
National Quality Forum Health Information Technology Expert Panel, HITECH,
which created a quality data set of NQF endorsed quality measures and their
component data elements. Through this initiative, NQF decomposed its current
portfolio of more than 500 quality measures into data elements that have been
standardized and can be made accessible to all the stakeholders in the quality
measures supply chain, including developers of clinical trials and clinical
guidelines, EHR clinical decision support rules and quality reports.” I
think you wrote that. That is a very long sentence. And then the next sentence
–- “A common data platform could help align the efforts and output of
the quality and performance measurement field.” With the exception of, and
we will fix that very long run-on sentence are there any additions?
MR. REYNOLDS: I am not sure I like the word platform. You are talking about
data set above, because as soon as you say platform, people would read it as
some kind of a system. We are really talking about data. Above you use data
element.
DR. CARR: Infrastructure?
MR. HOUSTON: How about framework?
DR. CARR: Framework, okay.
MR. HOUSTON: How about dictionary? No.
DR. CARR: I think that is too specific.
DR. HORNBROOK: Just a question here. The FDA sentinel network will involve a
lot of surveillance of huge data sets looking for adverse affects of drugs and
other things. I think we will also be looking, in the comparative effectiveness
world, for examples of surprisingly good outcomes that were not expected, so
you can do single detection for both bad outcomes and good outcomes. Is there
any reason to mention this whole sphere of work that is in the FDA that nobody
has talked about?
DR. CARR: Well I think it opens the door to an array of areas. That is
certainly a very key and important one and I think we could perhaps come back
to that in a framing sentence to say what the impact is. So can we hold that
thought and come back to it once we get through this?
MR. REYNOLDS: I have one other question about this. Since we do not tend to
mention a lot of the testifiers, which we heard a number, and we have focused
in this paragraph on a specific data set. Did we take everything into
consideration and are we endorsing this or are we just saying in this
paragraph, are we just using this as an example out of the hearing or are we
endorsing that and not mentioning the other things that we may or may not have
heard? And if we are doing that, fine. I just want to make sure that if that is
what we are doing that we all accept that that is what we are doing.
DR. TANG: I do not think that we are endorsing it per se in this letter. We
are using it as an example. We did hear about it. I am not sure there is
another counter example actually.
DR. CARR: Well could we begin it by saying an example of the focus on
–-
MR. REYNOLDS: There we go. I am good with that because again, short of doing
that, we are right early in the letter. We heard all these hearings and here is
who we are going with. And I think that might be good. Even if at the end we
recommend that, that is fine, but at least it puts it in a context as to why we
are using it as an example.
DR. CARR: And to your point, we might also want to maybe put a little
sentence in here about the excellent initiatives that we heard about.
DR. FRANCIS: This is just a process question. When you did the hearings, did
you get written testimony as well as the transcript? Is that written because my
experience with the PHR hearings was that the written testimony was incredible
rich? Is the written testimony in PDFs that people could access online?
DR. CARR: All of the presentations are on the website and Susan sent around
an excellent summary and there is a transcript verbatim of what was said. But
whatever written materials we have are in the form of power points and they are
on the website.
DR. FRANCIS: There was not any discursive written testimony?
DR. CARR: No.
MR. REYNOLDS: Okay, John.
MR. HOUSTON: I am just looking at the paragraph. When you make that an
example, you have a sentence at the end that does not seem to follow. I
understand the importance of that sentence, but is there another place for it
either in the beginning of the paragraph or otherwise?
DR. CARR: Yes. I think that we have found a couple of things in this
paragraph. One is that it is only three sentences, I think. The second is, I
agree, that this idea of saying an example and if it is okay, I hear what you
are saying and we will rework this so it flows better.
MR. REYNOLDS: Yes, we will leave that to the Committee. Good point. Bill.
DR. W. SCANLON: On that point though, I guess I saw the reason as why we
brought up NQF there was because we were going to switch from measures to
elements.
DR. CARR: Right.
DR. W. SCANLON: Maybe in the next paragraph we actually are more explicit
about that. And then if we start off with some of the material in the next
paragraph and then talk about the QDS as an example of your standardized
definition of elements, then it will come out less as if it is one testimony
out of the many that we heard.
DR. CARR: Okay. So the next paragraph then says, “NCVHS believes there
is an opportunity created by the HITECH Act to greatly facilitate the
development and utilization of more comprehensive and consistent measures of
healthcare quality and support the effective use of EHRs. We see an opportunity
to influence the incorporation of quality data elements and quality reporting
functions, NEHRs, through the meaningful use criteria associated with the HIT
Incentive Program. We believe this strategy needs to be developed quickly to
support the goals of healthy reform and to capitalize on the opportunity
presented by the HITECH HIT Incentives to influence the development of EHR
products.” And then it says, “in order to accomplish these goals in a
timely way, we recommend that HHS take the following actions.” Are you
still suggesting that we move that paragraph up?
DR. W. SCANLON: I think it involves a combination of moving some of that up
and then incorporating the NQF –-
DR. CARR: Okay. Yes, I agree.
MR. REYNOLDS: Maybe the first part of it and then the last part as part of
your recommendation.
DR. CARR: We can do that. John.
MR. HOUSTON: During this last paragraph on this last page, the only thought
in reading this and looking at it again was that you use opportunity like three
times. I am not sure whether the second “is an opportunity” could be
“we see an opportunity”.
DR. CARR: Got it. Okay we will fix that.
MR. REYNOLDS: Okay let’s move on to the recommendations.
DR. CARR: Okay, then we have four recommendations. “Recommendation one,
commission the Institute of Medicine to conduct a fast track study building on
its earlier Pathways to Quality Health Care Series, to produce a letter report
in six months recommending a national strategy for quality measurement
development, maintenance and effective use in performance reporting and as a
basis for payment reform.” Before the hands go up, Paul do you want to
elaborate on that a little bit?
DR. TANG: So there was a report, I forgot how long -– it was a few
years back –-
DR. CARR: It is Performance Measurement 2006. It followed on to Err Is
Human.
DR. TANG: And one of the biggest recommendations was the National Quality
Board. That is a new entity and it did not get very far. We wanted to sort of
decouple that recommendation for a whole new entity to say okay, this is also
what we said about the HITECH Act, this is a new opportunity to try to get
quality measures and performance measures particularly since we are in the era
of health reform and new payment criteria, we need to get this performance
measures down well.
So we thought that we would have a re-look on how we can organize this
supply chain in a way that produces meaningful quality and performance measures
in the context of putting in this new infrastructure, these EHRs into place. So
it is a bit more precise scope for what was previously covered in that report
but with a new target and a new sort of time.
MR. REYNOLDS: Okay, Tony, John and Leslie.
MR. RODGERS: Is the focus of such an organization the public payers and the
public sponsored beneficiaries or is this to drive HEDIS measures and
commercial quality measures as well?
DR. TANG: Well I would almost like to stay away from even using HEDIS as an
example because it is almost a counter example. So it is not the plans
measuring. We want the providers to be intimately involved in assessing and
improving their output. And it is not just about care, it is about health. So
there are a lot of concepts tied into this, but it is really to empower a
different group and for different use. And so the answer is that we are trying
to measure the process and output of our health system.
MR. RODGERS: Well the reason why I bring that up is CMS is in the same
process right now for Medicare and Medicaid. Are we going to recommend moving
that process into a broader process? That is the questions.
DR. TANG: I would think that CMS could take advantage of work like this and
I think that they would like to have that available. I mean I am just imagining
that they would like to have available information that more closely mirrors
the care and the output of the care.
DR. MIDDLETON: But Paul, I might suggest the IOM will make the
recommendations based on all the issues. We are suggesting launching the
process to analyze in toto systemically from the entire quality measurement
development life cycle perspective, what do we need? Public, private, provider,
patient –-
Mr. RODGERS: This is broader —
DR. TANG: It is absolutely broader than the public is really talking about.
DR. ROGERS: It needs to be harmonized with the public.
DR. MIDDLETON: That is why there might be an organizing entity thing that
will result out of it.
MR. REYNOLDS: All right we have got John, Leslie, Marjorie and Larry.
MR. HOUSTON: I am focused on one word and I think it is the first time that
the words are sort of phrased. At the end of this paragraph for the
recommendation, you use the phrase “payment reform” which you have
used the word “health reform” and otherwise in the letter, and this
is the first time I believe that you have raised the idea that there is
“payment reform” as being the basis for this. I react to the phrase
“payment reform”. Can we use –-
DR. TANG: We can go back to health reform. It has just got a new meaning. We
are not talking about QI projects anymore.
MR. HOUSTON: I know, but it is quality, quality, quality and all of a sudden
recommendation “payment reform” and I just look at this and –-
DR. TANG: So it was our attempt to say there is a new timeliness to it
because we are reforming the health system or we are attempting to –-
DR. FRANCIS: Why don’t you just end that sentence after quality measurement
development?
DR. CARR: I think I like the idea of health reform.
MR. REYNOLDS: Okay, Leslie.
DR. FRANCIS: So I sent this question around actually, and I did not get an
answer in the email. I did not see any discussion of whether the IOM is willing
to do this or whether it has the things on its plate, or even whether it has
the capacity to do it within the next six months? I know they have had a
long-standing interest. I have read To Err is Human and Crossing of
the Quality Chasm and so on, but whether they have sort of said their peace
or whether this is the kind of thing that they would be the best entity to do
it. Was there a lot of discussion of that at the hearing?
DR. TANG: Well the notion of commissioning an IOM report is our suggestion
for a recommendation from this Committee. It is not something we discussed with
the IOM. Most of their work comes at the request of Congress or the
Administration. So they do some that they decide to do as well, but they have
to find their own funding. So this is something that we are proposing to
recommend and the Secretary would decide if this meets with the
Administration’s agenda for how it sees healthcare going.
MR. REYNOLDS: One of the things we may want to do and I have listened to
Marjorie and Jim quite a bit on this is if we change that so say commission a
fast track study building on such things as earlier work by the IOM on such and
such it touts the IOM but it does not just give the Secretary one possibility.
Michael reminds me of this all the time, because if somebody said do not give
it to the IOM then do we just –
DR. TANG: I guess we actually inserted some of our judgment in this in
saying you know this contains political elements of it. And you would not want
a suggestion to just say such and such or commission its own. It seemed like
there was a role that the IOM does fit and I do not think the Secretary is
bound to listen to either recommendation –-
MR. REYNOLDS: No, but I think what would be good is if you added a couple
more words as to why you have that opinion. In other words, you just put down
an opinion and you just explained that with your other words that we do not
want this out there in this place. You want it to be based on that and I am
fine with that. But that then gives a background as to why we recommended that.
Jim.
MR. SCANLON: I would give the Secretary some wiggle room because you know it
may be that when you see the Meaningful Use reg this will be answered or there
will be another path you see or that health reform may bring something else.
And again, we have the greatest respect and spend a lot of money at the
Institute of Medicine but it may not be -– it is an narrow recommendation.
DR. CARR: I think the reason that we saw it as such a close fit was actually
as you look at the report that came out in 2006, these points were made with
astounding alignment of what we are saying now and so we thought of that as
building on work that is already done.
MR. REYNOLDS: No, that is fine. Just again, explain why that would be a good
thing. So Marjorie, and then, Larry.
MS. GREENBERG: I think my comments follow that. Where we were going with
this letter, recommendations two, three, and four, which we actually have not
discussed yet, but seemed to follow rather nicely. But frankly, recommendation
one just seems to be like where did it come from? It is almost like a
non-sequitur and more than that, since recommendations two, three and four are
pretty concrete and relate to the previous few paragraphs, are they supposed to
be on hold until the Institute of Medicine does that study? Or can they go
ahead and parallel? I mean there is just a real, to me, disconnecting here.
DR. TANG: Okay so let’s try to tease out whether it is because we did not
supply enough rationale, or whether it truly is a disconnected recommendation.
So we saw, to answer your last question first, we did think that they could go
on in parallel. In other words, you could work on two, three and four and in
fact, NQFs plan with HHS funding is to work on that kind of process because
that seemed to be one avoid but two, a lot of stakeholders have already bedded
and said this would really be helpful if somebody did this.
MS. GREENBERG: Well maybe you need to be clear about that or make those one,
two and three and then say at the same time we need a broader view.
DR. TANG: Well the way we thought of it, and so here is what we need the
Committee’s feedback on, is the recommendation sound or is the documentation of
the rationale lacking? We did, it just stuck out as a sore thumb, and the other
folks who were participating in the hearing please speak up, that there is a
lot of good work and even new work going on, but nobody really is talking to
each other let alone being in charge –- charge might be too strong, there
is no oversight, there is no framework to use Blackford’s word, to cause one to
predict that it would converge into something that could not only be reused in
the scientific community or the performance measurement community, but also the
policy community and that seemed like a crying need. And we did prioritize that
to be the primary thing that would move the good efforts that are going on into
something that could be productively used in both policy and performance.
DR. CARR: I am wondering if it would be helpful to briefly enumerate the
recommendations that came out of this performance measurement report, because I
agree with Marjorie, as we thought about it, we were sort of immersed in it and
we did not really give a lot of background. Would that be okay?
MR. REYNOLDS: Yes. Well I think I will make a comment on that. I challenged
the last letter that came out of Privacy because we have tended to have a
process. We have talked about observations and recommendations so that every
time we did a recommendation, right before it we kind of said okay so you have
read our background, now here is our context and here is our recommendation. So
all of a sudden, the last letter we have in Privacy. which I understand is why
we did it that way, but I think we need to go back to that because I think the
people reading it, and I was where Marjorie was and I actually went back
through the letter and said where did the IOM come from? You did not warm me up
or anything. So philosophically in the past, the observation and recommendation
has always been good because no matter what you read through the maze earlier,
you went ah-ha, there is an observation and ah-ha this is why they did what
they did. I have learned that from other Chairs and everybody around the table.
So Larry you have been waiting patiently and then I have got Mike and then
Bill.
DR. TANG: Can I say one statement about that? We had 18 months in Privacy;
we had 18 days for this one.
MR. REYNOLDS: I understand. The philosophy can still play. I am talking
about the philosophy and how we communicate regularly is to set somebody up
right before we pop them with a recommendation. That is all.
DR. FITZMAURICE: Harry it is on this point.
MR. REYNOLDS: Go ahead.
DR. FITZMAURICE: We do foreshadow this in the first paragraph on the second
page when we talk about furthering the initiatives discussed or independently
designed and described as siloed efforts without a consistent connection to the
national goals for healthcare improvement. We could take some of that phrasing
and put it back into the recommendation to remedy the situation.
MR. REYNOLDS: That is good, but for all Sub-Committees going forward let’s
make sure that we re-engage everybody right before we start popping them out no
matter how fast we do it. That could be one sentence or that could be a three
page observation which we have had in the 18 month Privacy letter. You know, we
had some of that stuff. So, Larry and then Bill.
DR. GREEN: Marjorie surfaced the issue that I wanted to raise. I do not want
to be accused of over warming Harry up.
(Laughter)
I thought we went by at least the hint of a suggestion that I heard from
Marjorie that I actually liked a lot. I think recommendation one is actually
recommendation four. And if we just lead it off a little bit, Harry with
something like to organize and sustain attention to these efforts and to Jim’s
part about leaving the Secretary some wiggle room, you know commission the
Institute of Medicine or another independent, credible entity to do that. I
think Harry, if we reordered it, that current recommendation two, three, and
four are the warm —
MR. REYNOLDS: Perfect. Again, let’s just make sure that as we communicate to
the outside which we do –-
DR. TANG: We just needed a 19th day that is all.
(Laughter)
MR. REYNOLDS: Well you got it and we are giving you tomorrow as 20 and so it
ought to be pretty good by the time we see it again. It ought to be flawless
and we should not have to chat about it anymore. Okay, any other comments?
Let’s read these last couple and then let’s get on to our next agenda item and
you guys will have time tomorrow to come back and tell us how you fixed it.
DR. CARR: So I am supposed to be two now -– recommendation two?
MR. REYNOLDS: Yes. Just read through them and we will not necessarily go
through them.
DR. CARR: “Initiate a process to identify essential data elements to
include the EHR systems. These elements would be the building blocks for
quality measures and risk adjusters. These data elements need to use precise
standardized definitions and terminology and must be assigned one codes to
ensure that they can be unambiguously identified. A set of data elements also
needs to be sufficiently robust to support the computation of current quality
measures and to support research on potential future measures. The NQF’s
quality data set may serve as the basis for this repository of elements. The
strategy to House data set maintained and expanded should be developed and
funded.” That is a long recommendation.
MR. REYNOLDS: Okay. Jeff and Judy and Bill.
MR. BLAIR: The sentence begins with “initiate a process”, it helps
my understanding a great deal, are you saying HHS or CMS or who should initiate
the process? I think in each case, in each recommendation we should be specific
about the agency or department that should be performing these actions.
MR. REYNOLDS: Okay Judy.
DR. WARREN: Mine was about assigning like codes. The way this is written, I
do not know what like codes are being assigned to, whether it is assigned to
the measure. If we are going to list them, we should have RXNorm, like and CT
on there.
MS. GREENBERG: You could say EG assigned standard codes, EG like or
something like that.
MR. REYNOLDS: Okay, Bill.
DR. W. SCANLON: We have had in our discussions some problems with what this
actually is about and I think the words repository and house, have the
potential for confusion. They kind of go together. We are not talking about
building a data set where there is going to be all these data someplace. What
we are talking about is, I mean I know that the word dictionary has special
meanings, but we are talking about specification -– compendium of
specification or something like that. I think potentially we are misleading
sort of in doing that.
The other thing going back to what you said earlier about singling out NQF,
maybe you should insert in front of “basis”, an “initial
basis” or this set of data elements because I think there are others that
are possibly plainer.
MR. REYNOLDS: Okay. Yes, Sallie.
MS. MILAM: Could you speak for a moment to the reference for the repository,
why the thought is that you would need some sort of centralized repository as
opposed to a federated structure?
DR. TANG: So by the way, just a reminder that this NQF work was a contract
from AHRQ, so it is sponsored by HHS and it came out of the AHIC quality work
group. So this has been a vetted discussion with a vetted need and then they
commissioned somebody to go do this. So this arose out of a process actually
similar to this and so the thought was if there was not a place to — well I do
not know if there that there is a difference between distributed versus
essential, but if there was not a place to repose these data element
specifications, then it would not be available to everybody and everybody would
not be looking at the same thing.
DR. CARR: I think there is confusion about the word repository.
DR. WARREN: A different way of putting it is you need to have your data
elements encoded in standard terminology; you need one version for that data
element, not a federated version that has multiples. You can multiply encode
these in different terminologies and then you cannot share data again.
DR. TANG: So the examples that Judy gave and other folks that talked about
how we can make sure, because that is what the EHR vendors are looking for,
give me one place and then the people who have to get information out of these
systems want to find one place.
MR. REYNOLDS: Okay but a library is non-confrontational, repository becomes,
I am putting my data somewhere else. Just for purposes for you to think about.
That is all we are saying.
DR. WARREN: To me a data repository means that there is data in there and we
are not really talking about the data, we are talking about the metadata around
this data element.
MR. REYNOLDS: Let’s go to recommendation three and four. Our next speaker is
here and I want to cognizant of that.
DR. CARR: “Recommendation three, EHR vendors should use standard data
elements as defined in the quality data set when reporting on meaningful use
measures. Recommendation four, require that a certified EHR have the capability
to add in a timely manner, additional, essential data elements that are
identified in the future.”
MR. REYNOLDS: Okay, John, I think I ignored you a minute ago; did you have
your hand up? Then Larry and Bill and then Marc and that is it.
MR. HOUSTON: I did have my hand up. On recommendation two, the beginning
says you “initiate a process to identify”. Are you really saying you
want to create a data repository? All you are really recommending here is to
initiate a process and it does not have any outcome to it, and I am reacting to
that a little bit.
MR. REYNOLDS: Okay, Larry.
DR. GREEN: Harry, I think for both of these we need to resurface issues that
just moments ago both Blackford and Tony brought up about the use of EHR vendor
being too restrictive. Tony brought up the issue of PHRs and Blackford, if I
heard you right, you brought up the notion of modules, and modularity. It seems
to me that both of these recommendations should adopt language that is broader
than EHR.
DR. TANG: Okay, we will take that into consideration.
MR. REYNOLDS: Bill.
DR. W. SCANLON: Maybe this is related, I feel like we have missed a
recommendation here which is that for three, that should be “certified
EHRs should include the capacity to store and report these standardized data
elements” — that kind of relates to what is current too because again,
sort of from this external perspective, I do not want to rely on the EHR vendor
to give me the meaningful measures. I want the EHR to give me the data that I
am actually combining with other data to get the measures that I want to use.
DR. TANG: Okay so I did not hear how that is different from recommendation
three.
DR. W. SCANLON: Recommendation three now implies that the EHR vendor is
going to be reporting meaningful use quality measures. I want the EHR to be
able to report data elements that can then be used in combination with
potentially other data to create measures.
MR. REYNOLDS: Okay Mark one and Marc two did you have your hand up? You are
going to have to introduce yourself before you are allowed to speak.
(Laughter)
DR. HORNBROOK: The whole issue through the last sentence up there on
recommendation three, “synergy to house the data set” — I mean most
people would hear that as the medical records data as opposed to the
information about measures. So we need to be very careful about using words
that seem to imply raw data as opposed to metadata, meta-variables.
MR. REYNOLDS: Okay, Marc Overhage.
DR. OVERHAGE: Hi, Marc Overhage, Regenstrief Institute and Indiana Health
Information Exchange. I have no conflicts. A couple of things, one and I need
to dig this out because we have been looking at this for a variety of reasons,
there is actually some language in the ARRA HITECH provisions that can be
interpreted to say that these measures must come from EHRs. So that is
something that we may want to clarify before we send this letter forward. I
think there is some controversy over it but there is some language that I think
we have got to at least address in our recommendations so that we do not
confuse. The other thing is that –-
DR. TANG: Well I think that he is right in the way it defines “clinical
quality measures” it does talk about data coming out of an EHR. I think
the way we interpret that is that there are sources of data that feed into an
EHR, but you are right, in that sense if you interpreted strictly, that would
have to come through an EHR then.
MR. REYNOLDS: Okay, thank you for your magical work in 19 days and on the
20th it will be amazing.
(Laughter)
Mike, one last comment.
DR. FITZMAURICE: I just wanted to elaborate on Judy’s comment about
including RXNorm and I would also add SNOWMED in there too. I noticed that you
did not have SNOWMED in there and the Department is heavily invested in
SNOWMED.
MR. REYNOLDS: Okay, thanks for everybody’s input. The Sub-Committee has
plenty to do. I would like to now, and I am real excited to introduce the next
topic. We are going to be hearing from Dr. Patrick Conway. He is the Chief
Medical Officer for ASPE as well as the Executive Director of the Federal
Coordinating Council on Comparative Effectiveness Research and some of us that
are close to that are excited to understand what that means. So we are really
excited about that. Dr. Conway, thanks a lot for taking your time and we
appreciate your comments today.
Agenda Item: Department’s Work on Comparative
Effectiveness Research, Briefing
DR. CONWAY: Thanks, I think we are pulling things up. Of note, on the last
conversation that was definitely interesting. I actually manage the National
Quality Forum Contract for the Department of Health and Human Services as well.
So the last conversation on the quality data set and the EHR retooling, which
we are asking them to continue that work that was started previously, in
addition we have identified 75 high priority measures in addition to the data
element work. So that is on their website; like many things, you have to search
the National Quality Forum website a bit to find it. We are trying to fix that
as well. But they are actually progressing on this EHR portion for the
Department link of quality and health IT. It is not the topic for today, but it
was very helpful to hear people’s comments from before.
MR. REYNOLDS: Good.
DR. CONWAY: So I will go through the slides relatively quickly so we can
have discussion. The Recovery Act allocated $1.1 billion dollars for
comparative effective research, $400 million to the Office of the Secretary,
$400 million to NIH, and $300 million to AHRQ to establish the Federal
Coordinating Council for Comparative Effectiveness Research. There was a report
from this Council on June 30th that included a definition criteria
and recommended priorities for OSC/CER funds. This report went to the President
and to Congress.
There was also a report from the Institute of Medicine, actually parallel
tracks on June 30th, on CER priorities and research questions. The
two reports actually ended up being quite complimentary, so the FCC report, I
would say with the common federal definition which was a first, priorities and
strategic framework set sort of a high level stage if you will, whereas the IOM
report named 100 specific research topics, a core of research topics to focus
on and prioritize those into four quartiles. So if you put the two together,
which HHS is now doing, in terms of funding priorities, I think they come
together quite well.
First the FCC definition; it is slightly on the long side. We broke it out
with bullet points to try to illustrate. The first two sentences are sort of
the core of the definition and then we did want to highlight some other points.
I will not read the whole thing, but to highlight a few aspects, so CER is the
conduct and census of research, comparing the benefits and harms of different
interventions and strategies to prevent, diagnose, treat and monitor health
conditions in real world settings. So a couple things to highlight there, one
— benefits and harms, the other — prevention, diagnosis, treatment and
monitoring. So a broader definition than if you just focused on treatment.
Secondly, that the purpose of this research is to improve health outcomes by
developing and disseminating this information to patients, clinicians, and
other decision-makers, about which interventions are most effective for which
patients under specific circumstances. So this is really calling out that the
purpose of CER is patient-focused, clinician-focused, informing decisions at
the front line of care and that we have to develop and disseminate this
information and figure out how to encourage adoption to improve health
outcomes.
I will call out a couple of the bullet points. The second bullet point, once
again, broad defined interventions, you can see everything from medications,
procedures, to diagnostic testing, behavioral change, interventions and
delivery system strategies. So if there is an intervention A and a defined
intervention B, rigorous research design, we would put that within the rubric
of comparative effective research. What is not in a, and I know a lot of you
know this, but just to be complete, randomized trial that compares to a
placebo, given that a placebo is not a defined other intervention, at least in
our definition, would not count as comparative effectiveness research.
And lastly, pertinent to this group, that this research necessitates the
development, expansion and use of a variety of data sources and methods to
assess comparative effectiveness and actively disseminate the results.
The priority station criteria that we wrote, number one was potential
impact, based on a number of factors. Number two the potential to evaluate
comparative effectiveness and diverse populations and patient sub-groups. This
was highlighted a couple of places in the report but that comparative
effectiveness should be very compatible with more individualized and
personalized medicine. As we are able to identify which patient sub-groups
specifically benefit from different interventions, it gets us closer to a world
where we have more personalized medicine.
Third, when there is uncertainty within the clinical and public health
communities regarding management decisions.
Fourth, that it addresses a neater gap unlikely to be addressed through
other organizations. This really gets to the point that we are primarily
focused on the Office of Secretary dollars.
Fifth, the potential for multiplicative effect. So if we think about a CER
data infrastructure investment, the ability to leverage that investment,
whether it is patient registry, or distributed data network, both to conduct
comparative effectiveness research, be a foundation for later comparative
effectiveness research, and potentially enable quality improvement reporting
the other things that we want to do in healthcare. Those are the type of
investments that have the highest potential.
So the strategic framework: Basically there are four major vertical
activities if you will and then several cross-cutting themes. So research;
self-explanatory. Human and scientific capital: so the idea that we have to
invest in training of researchers and methods and development. Third, CER data
infrastructure, in the report we call that everything from claims based
investments to clinical data networks, to patient registries, to practice
networks. And then fourth: the dissemination and translation of CER.
So how do we translate this evidence into practice? The cross-cutting
themes: Priority populations and we name several priority populations:
children, the elderly, persons with multiple chronic conditions, racial and
ethnic minorities, and persons with disabilities. It is not meant to say that
these are the only important priority populations, but we did want to highlight
these populations in the report. Priority condition: There is a number named in
the MMA Act currently and then types of interventions that are the full
spectrum of interventions.
This was the recommended priorities from the Office of Secretary. I would
note that the idea was that we have an ARHQ and an NIH funding mechanism. What
are the Office of Secretary investments that would complement those
investments? So you will see supporting investments and research in human and
scientific capital and conditions and the reason being those are primarily
areas where ARHQ and NIH make significant investments. Whereas data
infrastructure we really saw the Recovery Act funds as an opportunity to really
build that foundation of data infrastructure that could be leveraged long-term.
So we named that as a priority investment and also dissemination and
translation and priority populations.
I will say our spend plan has been approved by OMB and has gone to the Hill.
All the solicitations are obviously just coming out now from the Office of
Secretary Funds, and so they are not all out yet. So I obviously cannot, since
this is a public meeting, I cannot go into details about solicitations, but I
am able now to talk about very broad categories which are going up on the
recovery.gov website as well.
So with the Office of Secretary funds there will be investments in Medicare
data and Medicaid data, so this is building out the claims data investments.
And I want to highlight two points here. So one is thinking about how we link
that data to other sources, whether it be national death indices, more
longitudinal data, part D data, really thinking about how we can link the data
to be most useful. And then point two, is there is likely to be support for
researchers to access the data? So currently, there is Rezdac(?) and other
mechanisms, but really thinking about how we make this data accessible to
researchers. And potentially even when a grant NCR from NIH is received, the
ability to have discounted fees to obtain Medicare data or, and we do not know
this, as cheap as discounted free, although that has not been defined yet, but
really pushing on this access point.
I will go to data networks: So distributed electronic clinical data
networks, investing in distributed data networks. There are examples of these
in existence now such as HMO Research Network. Community health applied
research networks: So this is via HRSA, so really focusing on that population
and building some infrastructure there. Patient registries and cancer
registries: A registry of registries. I know a lot of you will have questions.
I may or may not be able to answer the questions on some of these. I know this
is somewhat vague, but I am being extra careful not to give away anything.
And then in terms of the FDA: Building the CER clinical data and standards,
infrastructures, tools and capacity. So thinking about how we leverage FDA
data. There are also investments focused on persons with multiple chronic
conditions both on the data side and research. And similarly with pediatric
care, investments on data involving pediatrics and research through some of
these networks.
Dissemination and translation: So investments on the dissemination of CER to
physicians and other providers, patients and consumers. Much of this work will
be operationalized out of AHRQ. Implementation strategies and AHRQ networks: So
to leverage some of those AHRQ networks to really disseminate and translate CER
and learn how to best disseminate and translate information. And then fourth,
accelerating dissemination and adoption of CER and delivery systems: So this
concept and actually it came out in the IOM yesterday as well, to really
leverage the delivery system arm in research. So to think about both on the
federal side where we have a delivery system, how do we leverage this
information through that delivery system, informing patients and clinicians, so
not mandating but informing. And then, second on the non-federal side, how do
we think about grants or contracts to encourage delivery systems to leverage
this information and to measure results.
Last is a research slide so here I highlight again and this is complementing
NIH and AHRQ investments, we do have a senior group in the department that is
literally, I keep saying bi-weekly but it ends up being weekly, meeting to
review solicitations and make sure everything is coordinated and look at the
whole portfolio of investment and are now actually going through with sort of
the last remaining funds to say where are the gaps? Is there anything that we
did not invest enough in, et cetera? And that is at the Director of NIH,
Director of AHRQ level, Office of Secretary.
So on research, behavioral and economics and change: Joint NIH, AHRQ
investment. This is great to see two research agencies working together.
Delivery system: Regionalize emergency care delivery. So a fairly select
investment, but something we actually heard in multiple listening sessions, the
importance of at least having some investment in this area. CER chronic disease
prevention: Potentially starting to think about how when we build data
resources like linked administrative claims, how do we start to do research on
these? Sixth, the centers for excellence for racial and ethnic minority focus
CER: So this will be a joint effort by the Office of Minority Health at NIH,
focused on these populations. And then the centers of excellence and CER for
persons with disabilities: So this would be a joint effort from the Office of
Disability and from the Disability Office within ASPE.
In terms of inventory and evaluation: So we do need to inventory what is
going on in CER. We currently still have in statute the Federal Coordinating
Council as an annual report do a statute which includes an inventory. So the
goal here is to collect both comparative effectiveness that is done through
public funding, but also try to collect CER conducted in the private sector as
much as possible. And also to make this a living inventory, if you will, where
there are contributions embedding over time.
Second, and this is difficult and we are working hard on this solicitation
currently. So not only can I not say what is in it, I do not know what is going
to end up being in it.
(Laughter)
It is an evaluation impact assessment: So trying to have all the projects
having an evaluation of metrics, but also trying to think globally. You know,
how do we try to build the story around, not build the story, how do we
evaluate what was effective and what was not effective? Why were certain things
more effective than others? It is very difficult to get at, given our complex
healthcare system, but trying to evaluate the impact of the portfolio.
And then lastly, we did fund out the IOM report that we mentioned and some
FCC support. I know I always talk extremely fast and part of that is that I do
have to leave around 12:30, so I apologize for that. And that is it.
MR. REYNOLDS: Well then we better question fast.
DR. CONWAY: That is true, that is the danger of that.
MR. REYNOLDS: All right. Tony.
MR. RODGERS: Just a quick question. Under clinical effectiveness, are you
including behavioral health on the social and not just the clinical side? —
Social impacts and determinants? I noticed that it was not up there.
DR. CONWAY: So let me try to answer that. So we, the definition would argue
any intervention and including and I always struggle as I think the whole
spectrum is actually intervention, so that will display by bias right there,
but including any social interventions. Any intervention that can be defined in
A versus B, I think is within our definition of comparative effectiveness
research. Now the next question is you know what is the distribution of
funding? There I think certainly was a struggle and is an ongoing struggle on
sort of the balance of what is funded or not. I think you know some of the
solicitations will probably have potential to fund that kind of work. So not
excluded, what is hard to know is that we cannot speak to the specific
solicitations and then we do not know exactly what is going to come back and be
funded.
MR. RODGERS: But it is within the scope and because it is really important
in Medicare, the core morbidity rate is very high, especially with mental
illness that impacts the clinical effectiveness of any particular type of
intervention.
DR. CONWAY: Yes, so it is within the scope of the definition. Like most
definitions as well, not surprisingly, there has been discussions about what
exactly the words within the definition mean, et cetera.
MR. RODGERS: The reason I bring it up is because the scope of what we look
at in terms of quality measures or data elements has to include behavioral
health information going forward.
DR. CONWAY: I agree.
MR. REYNOLDS: Okay, Don.
DR. STEINWACHS: I was a little concerned initially when prevention did not
show up anywhere in the definition, but I was happy to see then that you have
one of the initiatives there as looking at prevention. Did I miss it?
DR. CONWAY: It is in the first sentence. Prevent, diagnose and —
DR. STEINWACHS: I cannot read, thank you, Patrick.
(Laughter)
DR. CONWAY: And I thought we even had it in the defined interventions, but
we should anyway.
DR. STEINWACHS: Let me just mention one other thing. It seems to me many
times when we talk about what we have learned over time that the differences in
what people get out of different systems of care are probably much larger than
the difference between getting treatment A versus treatment B. And I was just
wanting to comment some about where you see that systems of care coming into
these initiatives? And I am sure it is planted in a variety of places.
DR. CONWAY: So in the interest of transparency, my research career, although
somewhat on hold at the moment, was largely focused on QI interventions and
systems of care including large networks or practices or hospitals. So I am
biased, but I agree with you that the systems of care and I agree with you that
it is incredibly important and I also, I am probably biased because I have
tried to design these trials and submit them, I also think it is incredibly
difficult to design well, and get funded, historically at least, et cetera. It
is definitely within the definition.
I also name broad categories of investment including things like delivery
system. I think this to be well, my opinion is that $1.1 billion sounds like a
lot of money, but in reality when you divvy it into three divisions with lots
of priorities and investments, et cetera, I think we will make some investments
in that area. And I hope that investment grows over time, because I do think
that those delivery system interventions have high potential to improve care as
you said.
MR. REYNOLDS: Okay, Marc.
DR. OVERHAGE: I follows on a little bit and I think I heard you say and I
have heard others say that placebo is not considered an appropriate control for
clinical effectiveness research. And the question is when you are doing
services or system interventions does that imply that normal care is an
inappropriate control for clinical effectiveness research?
(Laughter)
DR. CONWAY: How can I answer that? I think that gets to the heart of
different constituencies or organizations or people when they read that
definition will operationalize it different ways. So I think we tried to be as
specific as possible within the definition, hence all the words, yet still you
know you do not sort of hit on every nuance. I think some will say that
comparing it versus a usual care, which is a commonly accepted intervention,
and so that is intervention A, versus different care, which is intervention B,
is in fact comparative effectiveness. Some would argue that usual care should
not be in and we need to compare intervention mild, if you will, versus
intervention major and not have usual care count. I know this sounds like a
hedge to your question, but I cannot fully answer exactly how that plays out.
MR. REYNOLDS: Blackford, you are next and you did not need to set an alarm
to remind everybody that I have not called on you.
(Laughter)
DR. MIDDLETON: Boy that is a new most embarrassing moment. Patrick, no
matter what they are doing to me now, it is a pleasure to meet you and I really
commend you on lots of good work because I think coordinating these activities,
in a broad way with a big view, with a big vision as you are articulating and
displaying here, I think is just critically important. There are two things
though I would just like to highlight. One, I have been involved intimately
with the NQF work on the structural measures, the QDS effort and now the CDS
effort and hopefully in a month or two, the HIT Utilization Measure Development
type work. And I think all of that points towards really a need for frankly a
science around HIT which is lacking. We do not really know how to describe a
dose response curve for almost anything that we do with HIT. We implement CPOE
and do not know what response to expect. We know there are toward and untoward
responses and it is very difficult to predict the response because of the lack
of a scientific foundation for most of what we do, frankly. Yet we are spending
billions of dollars.
So I would like to suggest you include in the portfolio the sub-straight of
work that will help us describe taxonomy and measurement and performance and
assessment in a comprehensive way to get at these kinds of dose response curve
issues. And I want to do it, very personally.
(Laughter)
With all the right folks, of course. The second thing is pertaining to this
idea of a scientific foundation for HIT or its assessment; you know it is
difficult to think about the theory of evidence in the usual way. What I mean
by that, just briefly, is we are often times thinking about the theory of
evidence in a classical CER framework around randomized, controlled trials. And
all these even observational cross-sectional or other types of survey, work to
look for correlation or even better to look for causality. I think though we
are entering a world, as things become much more connected and much more
information intensive, where there will be whole new classes of information
perhaps and the need for softer inference mechanisms to allow for association
to be viewed in ways that historically have been viewed negatively.
Very simply I am thinking of the kind of evidence that will be coming from
“Patients Like Me”. When cohorts of patients self-aggregate, natural
experiments occur. We sort of have an epidemiologic framework in qualitative
research methods and whatnot to address some of that, but I think we need to
think this through carefully and critically from an HIT context point of view
as well. Thank you.
DR. CONWAY: Thanks. I agree with both statements. So I think you are
completely correct on sort of the science and health IT. Not surprisingly,
exactly how and where that is funded with HHS is an ongoing discussion, because
there are multiple entities that fund that kind of work within HHS.
So once again, I always feel that I have to disclose my biases because they
are true, I am not a randomized trialist, and everybody sort of knows the
background they come from. So I have a Masters in Clinical Epi, so I actually
believe that we can do things like instrumental variables, whatever
methodologies are.
I also think a lot of those methodologies are somewhat limiting and I know I
am not smart enough to think about the next methodology, but we need to fund
someone who is doing that kind of work that says when we say pragmatic trials,
when we say alternative designs, when we are talking about a new world where
there are patients cohorts on “Patients Like Me” instead of what may
happen, i.e. discounting it, how do we actually think about the methods that
would be reasonable in those types of studies and what it means in terms of
funding them, publication, et cetera sort of the whole putting them into
practice.
So I mean I think, I am also trying to think about what I can say, we called
out the methods and the methods work in the report and I think we are going
through a gap analysis right now and so I think that this methods type work is
one of the things under discussion. Do we need to find more work in the methods
area?
MR. REYNOLDS: Okay we have Leslie and Larry.
DR. FRANCIS: These are two big issues and I fully realize that they are big
and they are messy and I would just love your thoughts about them, okay? The
first is, I think I saw fleetingly the word “cost” but I never heard
it come off your tongue.
(Laughter)
So the question is, are any cost data contemplated and if so, how and
relevant? And that is one. The other is that I worry a lot about comparative
effectiveness research and how outcomes get defined. That is, what is the
endpoint that you are interested in and in particular, if the data that are
collected are largely quantitative rather than qualitative -– so for
example, pain, which is a qualitative question. Whether the kind of data that
is simplest to collect drives how you think about outcomes. Whereas what might
be very important to patients, say comparative levels of discomfort, and that
never gets captured.
DR. CONWAY: Right. So let me do the cost one first. Cost was not explicitly
included or excluded in the statute. It also, in the definition, is not
implicitly names in the definition. Where you probably saw the word was in
potential impact in parenthesis, cost was one of the factors in impact. So
certainly the FCC report highlights comparative clinical effectiveness, but is
also does not exclude cost effectiveness analysis. Nor was it excluded in the
statute. So cost is not excluded as something that one could look at.
DR. FRANCIS: Let me just follow up on that. You say it is not excluded, does
that mean that we are planning to look at it? And when you think about it in
terms of priority, the question of if we have got large cost differentials,
anyway –-
DR. CONWAY: So it is not excluded. So we could look at it. Solicitations may
include things like the word cost. I am trying to think of any of the ones that
already posted do. Some of the ones from AHRQ actually talk about value, so
quality over cost, so not excluded. I think the level of the focus and the
amount of funding is to be determined, so how much is more focused on value
equations or cost effectiveness, I think, remains to be determined. It is not
excluded in the statute.
Your second question was –-
DR. FRANCIS: Qualitative and whether the ease of collecting certain kinds of
quantitative data is going to be a driver in what outcomes get looked at?
DR. CONWAY: So here AHRQ has been the primary driver of this work. They do
have a solicitation, I think, that actually came out earlier this week talking
about building on patient-centered outcomes and specifically named that we need
to think about broader sets of patient-centered outcomes and that they are
soliciting research that thinks about collecting qualitative information as
patient-centered outcomes and expanding what we traditionally look at. I think
that this is amazing how all the things are related. I mean that goes back to
the other aspects of measurement and how we define measurement and also goes to
health IT in terms of how we collect the measures.
So I mean all these things are a sort of triad of health IT, quality
measurement and comparative effectiveness which are the three things that I
primarily work on at HHS. I mean, as you all know, they intersect
substantially. But I think we have a lot of work to be done in that area and I
think the good news is that it has been specifically named that we need to
think about outcomes in a different way than we traditionally have, i.e.
mortality.
MR. REYNOLDS: Okay. We have Larry and then Marc and then we will get you out
on time.
DR. CONWAY: Great, thanks.
DR. GREEN: Dr. Conway, I want to thank you for parking your research career
and doing this.
DR. CONWAY: Cincinnati does not thank you. I have been on leave for two and
a half years. Every six months, I have a great conversation with my Chairman.
(Laughter)
DR. GREEN: Congratulations.
(Laughter)
DR. GREEN: To go back to Blackford’s word, coordinate, in those very complex
slides you just showed us, the acronyms AHRQ, NIH, CDC, and HRSA all appeared
here and there and then I hear the word coordinate. And I wonder if you could
explain to us, really fairly explicitly, what the federal strategy is to
coordinate comparative effectiveness research across NIH, AHRQ, HRSA, and CDC.
DR. CONWAY: So first, we have pushed impressively hard on this issue in my
opinion and let me explain what we have done. So we set up for the federal
coordinating council we had representatives from not only across HHS but also
OMB, Department of Defense, VA, et cetera so there was coordination in sort of
the initial priority setting. Since then, as we have gotten to the level of we
need to write solicitations, we need to post them, we need to see what we
funded, and I think HHS is a champion in this. They have actually brought in
NIH, AHRQ, all these different departments within the rubric of the Office of
Secretary, and said you know, what do you want to fund? Let’s vet that through
a similar process. Let’s look at it as an entire portfolio. Let’s actually have
the leadership of NIH, AHRQ, and Office of the Secretary at a table looking at
solicitations. So I mean literally reading, like this is the next grant
announcement.
So coordinating to that level, approving those solicitations, then once they
are posted, we have a reporting back mechanism where we are going to report
back to that group. This is what we funded. You know, what are the gaps? — And
so the circular quality improvement, plan, do, study, act. So it is literally a
process where we are coordinating on setting up the solicitations and as we
start to get information back on what we funded and ultimately even evaluating
the impact and having that be a continuous circle of coordination and improving
the investment. Obviously, with health reform, I do not know exactly what CER
looks like in health reform, nor do I want to place any opinion on that. But I
think the coordination function, wherever CER lives, is going to be important
for the portfolio to be as impactful as possible. So I think in my former life
I was a management consultant and looked at private sector portfolios as one of
my jobs. I mean this idea is similar in public sector portfolios; we need to
continue to look at the portfolio investment, evaluate the results and realign
priorities as we see those results come back.
MR. REYNOLDS: Okay, Mark Hornbrook.
DR. HORNBROOK: One anecdote. One RFA from NCI on genomics did not have the
word cost effectiveness edited out of it and several members of Congress wrote
very strong letters to the NCI about what was going on at NCI. So we are in a
strange world these days and the whole debate about the homography guidelines,
I am sorry to see that we have not found a way to use this particular thing as
a way to create a place for rational discussion about some very complex issues
instead of having these strange he said/she said and large polarization.
MR. REYNOLDS: You might want to say you have to leave right now.
(Laughter)
You do have that option.
DR. CONWAY: Yes, duly noted. Thank you for the comment.
MR. REYNOLDS: Listen we really appreciate your time. You are doing great
work. It is part of making a big change. So thanks a lot for everything you did
and with that we are going to break for lunch. I am going to leave it until
1:45 since everybody has to kind of go out of the building and get something
and then come back. So we will make it 1:45. Please do not make that 2:00. It
is 1:45. It was 1:30 so I am giving you fifteen. I will find you.
(Whereupon, the meeting was adjourned for lunch at 12:28 p.m.)
Afternoon Session 1:45 p.m.
MR. REYNOLDS: Let’s go ahead and get started here. I appreciate everybody
getting back on time. So what we are going to do is we are going to now move
into the portion of the program with Dr. Steinwachs and Dr. Scanlon, and we are
going to build off of the excellent work of our consultants and look at how we
are going to move forward with enhancing health information capacity for the
21st century. And with that I will turn it over to whichever of you
appears to be leaning forward that you would want to take this.
Agenda Item: Enhancing Health Information Capacity
for the 21st Century
DR. W. SCANLON: One of the problems with health information is that it has
never been in real time. And that is why we will finish this presentation at
1:46, okay.
I will start off with sort of a bit of recap in terms of kind of how we got
to the point where we are today. As you mentioned, we had sort of this
excellent help over the last I guess about six months or so from Dan Friedman
and Gib Parrish in terms of revisiting the report from 2002 about a vision for
health statistics for the 21st century. And a portion of why we were
motivated to do that was the fact that since that time, the potential for
having information has changed by the increasing prevalence of health
information technology. And with that increase, we are hopefully now sort of in
a period where the rate of increase is accelerating dramatically due to ARRA.
And so we wanted to think about how we can take advantage of that. Now Dan and
Gib had a series of interviews with stakeholders and informants as well as
input from the Committee and produced an interim report in September that we
discussed.
In that report, there are a large number of recommendations in the original
2002 report. We concentrated, sort of in terms of our review, on a subset of
about eight and then in that interim report that Dan and Gib produced, they
were boiled down to three. And these are the three that we were talking about
then.
After presenting this and after the Populations Sub-Committee had a breakout
session to discuss the implications of this, one of the outcomes was to change
the focus somewhat and not just in terms of labeling, but to think about not
just health statistics, but health information and a vision for health
information for the 21st century. And coming out of that discussion
and thinking about where we are at this moment, which we discussed this
morning, where we have got this huge undertaking underway in terms of health IT
and meaningful use, so what is the best thing to do at this moment as opposed
to thinking only about the much bigger picture alone? What are our
opportunities for this moment?
So one of the discussion points was to think about how we can address those
recommendations that we had in September in the most expeditious way in the
short term and set up work for the longer term. And in doing that I think, and
these are discussion that came out of the breakout session as well as
discussions that the sub-committee has had through conference calls, is the
idea of let’s focus on the objectives first and think about the linkage between
those objectives and the technology opportunities that we have. And then let’s
think about the other elements that are in those recommendations as part of the
implementation plan, which might be a longer term sort of effort, because we
are a Committee with limited resources.
So from that first recommendation, it seems that there is an information set
of needs in at least three major areas. One is guiding policies and actions to
protect and improve public health and the healthcare system. The second one for
the private sector, we want to have data that are helpful and relevant to a
variety of different stakeholders. And then the public sector has a stake at
multiple levels. It is not just federal. It is both state and local as well,
both with respect to public health and the operation of public healthcare
programs.
This is again taking language out of those recommendations that we had in
September and in some respects they sort of reiterate some of the things that
we have talked about in terms of our objectives. At the quality hearings that
were in October, you could even add to this list of things that we want
information for. And it comes down to different levels. We want to know some
things at the individual level in terms of the care the people receive, the
communication about that care, and their ability to manage their own health. We
want to be able to serve at the provider level to give providers some internal
feedback as part of some kind of quality improvement. Payers need information
in terms of improving the efficiency of their policies.
One of the things for right now if you want to talk about the health reform
that we are debating, one of the handicaps in terms of thinking about
healthcare delivery innovations is the fact that you reach a point in terms of
using existing data where you say I have explained this amount, but there is a
whole large segment that I have not been able to explain and the question is in
terms of a reform of a payment policy, a coverage policy, or some kind of
administrative procedure, are we doing a good enough job in terms of being able
to explain things, to make this a safe and worthwhile kind of change to do? So
these are kinds of different information needs that we need to keep thinking
about pursuing.
This is just again, some more recap. Today when we talk about information
that we have, we have it from a variety of sources. We are in the room where
the Board of Scientific Counselors for NCHS meets and they are certainly very
much in the population survey business. As the advisor to NCHS, they are a
critical part of the information that we have. There is a Vital and Health
Statistics System and the disease registries that we have. We have claims data,
administrative data, and we have clinical records and the possibility of
merging data off of clinical records exists.
With all we have talked about, we have real problems with all of these. If
you look at any sort of individual data source, it has relatively limited
content. We do not really, and we have had hearings on this, that we do not
frequently take advantage of all the linkages that might be possible. And in
part, we need to expand the array of potential linkages and that is something
we need to be thinking about.
And then the last sort of limitation that we have is that if you have any
one of these sources, if you cannot link to some other existing data set, and
you have to think about getting supplementary data that becomes an incredibly
costly exercise. Kind of a classic example is the whole issue of administrative
data and clinical data. If you have got administrative data on a population and
you want to go and enrich that by getting clinical data, medical record
abstraction is hugely expensive and therefore, rarely done. So we need to be
able to move beyond that.
So where we were at the end of the Population Sub-Committee’s discussions
was thinking about for June 2010 to have an objective on a memorable date. We
wanted to think about a report related to this work that has been done and that
it would in great part focus on how electronic health records can contribute to
improving on the situation that we have now. How can we basically put together
new and existing data sources in ways that are going to be more effective in
meeting information needs that we have?
I think we have probably discussed this here this morning to death, so why
are we focusing on EHRs? I mean they are the opportunity of the moment that we
are making this investment with. Jim talked about $19 billion and it could be
up at $50. CBO’s estimate is that it is going to be about $36 billion. People
keep talking about the $19 billion as if that is all we are going to spend.
That is what we are going to spend if we manage to achieve the offset and get
some savings. My particular perspective is that if we do not work hard, we are
not going to get the savings. We are going to manage to spend the money, but we
are not necessarily going to get savings. And so that is something that we have
to be very cognizant of.
As an Advisory Committee to HHS, one of the things that make it relevant for
us to be thinking about this is that HHS has the ability to influence the
outcome here through the meaningful use regulations. Not just the ones that are
going to be coming out this next month, but over time. And also because, as
this Committee is responsible for the statutory responsibilities with respect
to HIPAA, there is a potential link with HIPAA, because I think we have to put
on the table that claims are a vehicle that can potentially be used to improve
information at various levels. And so the question is, what is the optimal
strategy in terms of trying to do that?
The other thing which Jim brought up this morning is that within the reform
legislation that is in the Congress right now, they have raised issues with
respect to HIPAA that will come back to this Committee. And so there is this
question of if we are thinking about HIPAA, and thinking about having it
realized, and its original promise of administrative simplification, should we
be doing that in the context of what is the greater good that we can get out of
this whole process? And if we are going to undertake the effort to look at
this, let’s maximize the return in terms of the things that we get out of this.
The other thing I think is that opportunities here with respect to augmented
claims and some kind of standardization here that it may be as a shorter term
return than some of what we envisioned for health information. That over the
longer haul, what you really would like is to have fully functioning health
information exchange so that anybody that is receiving care has access to
information about the care that they have received from others and it is all
seamless and there are no problems in getting information about this person to
optimize their care and then there is no problem about retrieving that
information at the right points to either measure the quality of care or make
appropriate payment, et cetera. But where we are now, my perception is that it
is going to take a longer period of time to get to that point than it might
take to get to a point where we can actually think about augmenting claims and
having that information database to be able to engage better and more efficient
efforts at reforming the delivery system.
We put down here that that may be possible within five to ten years. That
seems totally unacceptable in terms of the length of time, but it may be more
reflective of the reality that we need to think about dealing with. This also
seems like an effort that what we are really doing is we are building upon
NCVHS efforts that have occurred over the years. And I am more specific here
about areas in applications, in terms of quality measures, meaningful use,
secondary uses, and data linkages. The Standards Committee and the Privacy and
Security Sub-Committee can probably talk about specific hearings they have had
too that relate to this effort. I was not there and so I could not list them in
this power point and we had to finish at 1:46 and so we did not have any more
time to revise it.
(Laughter)
But the reality is that we have a history here. We are not starting from
ground zero in terms of trying to develop what we would say about this area.
This is building upon what we have been working on, at least since the time
that I have been on the Committee, and given when I came on and seeing some of
the work that had been done before; we are certainly building on a longer
history than that. And so this is not out of the blue and therefore and so
thinking about a report for June of 2010 seems feasible. As long as we have the
right expectations about the report for June, this is not going to be the
definitive piece to fully specify where we see the future to be.
What we did was we asked each of the sub-committees to make a contribution
to this because we see each of the subcommittees, and this is open for
discussion though, there was no vote to say this is how we will operate. Have
each of the subcommittees think about what they would like to contribute to a
June 2010 report with respect to this general theme of how do we better meet
the information needs for the 21st century by making more effective
use of electronic health records.
So Justine and Paul, we will let you comment and we will let each one of the
subcommittee chairs comment about their slide and then we can have a discussion
with this.
DR. TANG: Well actually, I think this is a repeat of this morning’s
conversation since we did the letter, but we see this as a real important
opportunity to focus not only on where data used to be, but where it is going
to be in the future. And it is going to be much more in the clinical side, much
more about real time feedback to clinicians as they deliver care and improve
health so that we can constantly measure and improve our practices.
DR. W. SCANLON: Right and I agree with you completely. It was almost as if
this morning’s letter was Chapter 1 of the report. It was so close it linked in
terms of what we are planning and thinking about doing. Don, do you want to
talk?
DR. STEINWACHS: So Bill reminded me when I got here today that I had not
filled out the slide which meant that I had not consulted with the
subcommittee. So these are Don’s ideas for discussion at the subcommittee
meeting. But it seems to me that one of the challenges that we have out there
is how do you construct denominators when you talk about using electronic
health records and you think of many times person reported data and not just in
systematic surveys, but there may be survey mechanisms like people are doing
things on the web and other mechanisms and you begin to think about making some
advances in this area.
I have hope, partly because if you look at how much we have spent over the
years trying to figure out how to deal with the non-respondents to surveys and
how do you deal with there are usually only about 50 percent maybe 60 percent
of physicians who respond to any survey and so we figured out what to do with
the non-responders.
Well in a sense, we have a similar kind of challenge here as a population
orientation. The other thing is that we have some real advantages. As you link
together electronic health records and administrative data for instance, most
administrative data systems that are tied to insurance have a denominator. They
know something about the denominator population. It is partially geographic,
but it is a denominator. And if you think about the other linkages that you
could create here, you could really build off of this.
So I think one of the things that we ought to lay out is how do we begin to
approach this issue? The second, which is to me sort of the great hope thinking
of electronic data linked to administrative data, is that we have lamented now
for all the time that I have been on this Committee and I guess all the time I
have been in the field of health services research, there is very little local
or state data. And so that we have, many people see as one of the best systems
in the world sometimes for national health data, coming out of the national
center and other sources. But when you say what does a community have or what
does a state have? So as you look at expansion of health insurance, expansion
of electronic health records, administrative data and link those together, you
are now talking about the possibility in a reasonably short period of time,
that indeed you could populate such things as information on access to care,
information on indicators of problem or disease status, if you could link in
patient report, which I keep hoping that electronic health records are going to
have, a comp on it which is functional status reported by the patient in that,
you could really get a lot of mileage. Then what you would be dealing with are
statistical issues about how to deal with the biases and the representation.
But it seems to me that those are doable issues because you know which
practices, which clinicians, who is contributing and in many way, which is not.
Special populations, we talked a little bit before, Tony did I think about
the issues of capturing race and ethnicity to our systems and other things. But
yet in some ways, that is not a new issue. We have been dealing with that for a
long time. Social security deals a little bit differently, maybe than Medicaid
does and health plans that many times do not have it at all. So this is an
opportunity it seems to me to get at special populations and with a level of
information that we have not had.
When you come back to our framework for looking at health information and
health statistics, we talked about a variety of data sources that we would like
to have linked in in order to be able to look at health. And so I have noticed
that President Obama’s Open Government Initiative, I heard a presentation by
the EPA the other day, you can now go to a website and you can pull down all
the environmental exposure monitoring data in this country. And so in concept,
depending on whom you are looking at and how that overlaps monitoring sites,
you could begin to link together exposures that populations have that may have
something to do. And a colleague of mine has been doing some work over the past
five years beginning to pull health data together with environmental monitoring
data and there are differences. It makes a difference whether you live in some
of these locations.
Resources: The National Center now produces the information on much of the
health resources in terms of the physician practices, hospital discharge, and
other things. You can begin to see an opportunity to expand the detail that we
have particularly in physician practices in this country which has been a very
tough thing to collect. Not necessarily to replace NPSIS, but to expand beyond
what NPSIS does now.
We are weak in occupational data. I have not decided to propose electronic
health records capture occupation, but Blackford told me they should. Didn’t
you Blackford? I do not know.
DR. MIDDLETON: They could.
DR. STEINWACHS: We are one of the few countries that do not have that and
therefore does not have access to the kinds of exposures and risks that go with
different occupations. And so we neglect the fact that most adult Americans are
in the work force and a good part of the exposure that they have in their life
that affects their health is in the occupational setting. And again this might
be something we could open a door to that we just do not have right now.
So these are some ideas at least for the Population Sub-Committee to
discuss.
DR. W. SCANLON: Right and I think what Don has said sort of underscores the
point that in shifting from this idea of health statistics to health
information needs, we are not really shifting away from having information
about population health and the population. In fact, what we are thinking about
is how we enrich those data in a feasible way. I guess because it is more than
just efficiency.
One of the problems that we have had in population health over time is the
limited amount of resources that are available to gather that information. To
be able to have a strategy that takes advantage of this investment in health IT
and combine that with what the traditional investment or even an expanded
investment in health statistics would open up all kinds of new opportunities.
And I think it is in that context that we thought it was not a big leap to go
from the vision for health statistics for the 21st century to this
information needs for the 21st century. So Judy and Jeff -–
Standards.
DR. WARREN: So what I listed on the slide was basically the core piece of
what our Committee looks at in looking at terminology standards or at the
values of data elements and looking at sending that data from one system to
another or messaging standards. And then this morning I was intrigued with one
of the comments that Justine made about standard queries. And so looking at do
we need to start even looking at create some of those standards because as we
start getting to the more sophisticated measures and metrics, we may want to
have ways of pulling that data together and that are traditional transaction
standards.
So the way that I thought that our Committee could do is that Jeff and I had
started planning a series of hearings on following up with lessons learned that
we learned when we did the standards for the NHIN and for e-prescribing of how
do we develop these standards, how are they adopted and then how are they
implemented in ways that make sense and that are the most effective and are the
most efficient? And one of the areas we were pulling together were people that
we thought had never really been pulled together before. So pulling in the
research community and start getting them introduced to these standards and how
to use them in their research. Pulling in the administrative community, which
already has quite a bit of experience, in using them and then pulling in the
clinical community, with EHRs. Jeff, do you want to add anything to that?
MR. BLAIR: Nope, well done.
DR. W. SCANLON: Okay. Good. Leslie.
DR. FRANCIS: So this is the first of what was meant to be two slides.
(Laughter)
And I also need to apologize, like the others; this is in certain respects,
just me and things I would hope our Committee would talk about.
So first of all, I took this as about getting new sources of data more
widely than just from EHRs. So there are a lot of questions about whether HIPAA
does or does not cover some of the kinds of data that we might be wanting to
get or that there might be interest in getting. We have a recommendation from a
few years ago about expanding the scope of HIPAA. We also have thought about
that in connection with PHRs. I do not know whether there is any interest in
getting data collected for research, but one of the very tough questions has to
do with informed consent and new uses of data in the research arena.
One of the things that has been brought to the attention of our
Sub-Committee that I think we are going to be continuing to think about, are
differences in state law and the impact on data sharing. And then Don mentioned
to me earlier that some of the kinds of limits that might actually apply to
federally collected data already so that is one set of what is just there right
now in the current landscape.
Now the next slide which is blank was going to say, so here are some of the
very live issues for our Committee. Security is an ongoing question. You know
if there are large data sets. That I think should go without saying. With
respect to confidentiality, something that actually our conversation at lunch
led me to think about is, is there any interconnection between the generation
of large data sets and recommendations about the possibility of sequestering
sensitive health information and would all information go how, and what? What
about questions about re-identification as data sets get combined? That is a
huge question that will need to be thought about.
The secondary uses report raised a number of very important kinds of
questions. Among them, particularly if we are interested in sub-group analysis,
questions about group stigmatization, whether there is any issue about whether
or not there might be unexpected inferences drawn from data of toxic
neighborhoods come to mind. You know, given what you were talking about. And
then, finally I am assuming there would be a separation of this, but I do not
know from any kind of feedback loops to individual providers, individual
patients, what would be the connection between any of this with disease
surveillance and so on. So that is my list. I am sure Committee members might
have more too.
DR. W. SCANLON: All right. I think what you have got is a sense of what our
discussion have been to date and it is really something that was meant to be
brought to the Full Committees for your input and your decision making in terms
of the directions that we go. I think Paul was totally accurate when he said
that this morning’s letter in some respects is kind of a precursor of what we
are thinking about here in the sense that we have objectives in terms of why we
want some information. And we have to think about what are the practical
methods in terms of how we are going to get it.
There are then an issue of both, here is a vehicle, but how are we going to
make that vehicle be created? How are we going to protect privacy and
securities in those kinds of things? So it is something that seems to cut
across the Full Committee. If we have a June 2010 date though, we have to keep
it manageable. Health information needs is a huge topic and we want to say
something meaningful, but we do not want to kill ourselves or fail by June
2010, so we have to find the boundary that is going to make this a feasible
task for this Committee. Those of you who have been on it for awhile know what
our resources are. Most of us have day jobs and they tend to interfere. So I
would like to just open it up then for discussion.
MR. REYNOLDS: Let me do this. You guys have been working on this. Let me let
you guys jump in if you have any comments on what you have seen and what you
have heard, and then I will just open it to everybody else for questions.
DR. FRIEDMAN: Thanks. I do not have a lot to say. It certainly is an
interesting approach. I think that certainly one of the things, and I am not
telling any of you anything you do not know, the Committee is going to need to
deal with what kind of data and reasonably the expected to be derived from
electronic health records and whether or not those data will contribute better
to our understanding of health care as opposed to sickness and as opposed to
health and the influences on health more generally.
Certainly, when I think about the sources of data that we look at on
population health, it is not confined to administrative data and survey data. I
mean, there is a whole world out there of data that are used and that need to
be used because of the limited influence of healthcare on health. When we think
of the possible contribution of electronic health records to better
understanding population health there are a number of sub questions that I
think we need to look at. One is can electronic health records replace any
current information sources on population health? The second would be, well if
they cannot replace them, could it supplement them? Could it complement that?
And then also, where won’t electronic health records be able to contribute at
all?
And in some ways, if you look at test cases, and one test case right now is
England, and obviously they have had electronic health records at the GP level
where they have been rolling it out for about 10 years now, they have GP to GP
sharing of electronic health records. They just rolled out a summary care
record for London which basically means an abstract of the electronic health
record that would be shared, available to clinicians and also put on the
National Spine, it is called the National Spine, and one of the things that I
think the Committee would do well to look at is talk to some of the people in
England and try to get a better understanding of the extent to which those data
and particularly the summary case record data, are going to be used for
population health purposes and the extent to which they are only going to be
used for clinical purposes. And the NHS’s secondary uses service, that is
something they have been grappling with for quite a few years now and their
answers on that are by no means clear even after looking at it for several
years.
So in summary I guess my comments would be just if you are going to focus on
the EHRs, you need to go into it with some sense of skepticism about how those
data and whether those data are going to be useful really for population health
as opposed to for just improving clinical care. So that is all.
DR. PARRISH: I agree, of course, with Dan’s comments and have just a couple
more. I think that there are opportunities available for population based data
which are becoming available now that really could address a number of the
issues that were raised by the various Sub-Committees. The issue of having
true, population based local and state data remains critical for many programs
that want to try to do something at those levels as opposed to just national
programs and certainly the Bureau of the Census is not with its American
Communities Survey, is actually now rather than having really complete data at
a local level every 10 years, is working on having those data available on a
sort of a three year cycle or even down to almost neighborhood level.
And I think there are some opportunities now both with that and the health
sector working with some of these other groups to have that kind of true
population based data available on a more regular basis at a much more granular
level and in the relatively near future if there is at least even a fraction of
the funds put into that, that are being talked about here in terms of
healthcare information technology infrastructure.
I also want to follow-up on a comment that Blackford made earlier which is
that I think now really is the time to start thinking carefully about other
kinds of data sources that are becoming available because of the Internet and
other sources out there that can be now looked at and hopefully compared with
actually gold standards sources of data to see in fact how reasonable they may
be to be used in the future as sources for really talking about population
health.
DR. O’GRADY: I would like to just take a step back for a second and think
about this to sort of to set a little context for this and I am very encouraged
by the way you guys are going about this. I am going to step a little bit out
of what a scientific counselor is and a little more into being the next policy
maker. When you think about why this organization was originally set up and
what really makes an effective FACA in terms of informing policy, development,
implementation, et cetera. It is that we work in a system, certainly the fact
that this organization represents HHS wide is very important as a counter
balance to the extreme problem that HHS has with silos. FDA looks at their
world, CDC with theirs, and CMS with theirs. And this organization has the
ability to take a broader view across all that and think about the
interactions.
The other part of this is when you think about both in terms of the way
legislation is developed and then the way it is implemented, this is a
representative democracy, people are extremely heavily lobbied, not always
getting the full range of views and there are very good ideas out there. You do
not have particularly good lobbyists. Sometimes they are not heard. And so a
really good FACA has that ability to help the policy maker kind of look up from
the, I have got a bill too. I have ARRA to implement and I have a certain short
time period and half my day is filled in with people coming in to talk to me
about it. It is a great counter balance to be able to help them look up, see
the broader picture, and see these interactions.
So you have talked about a number of things here today. And I can see you
have talked about data linkages, there are all these other policies that are
evolving somewhat uncoordinated within the department. FDA is doing all kinds
of trying to think about post-market, tracking of different things once their
drug is already carried. CDC has strong responsibilities both in sort of
Homeland Security as well as more general disease tracking. Clearly CMS has
their own responsibility and has received a fair chunk of the money in terms of
thinking about doing this.
What I think that you can really contribute here is to avoid some of the
flaws of the past. On the financial side where we had CMS basically set up
their own system for paying providers and then the private sector had a second.
And that was just a waste of everybody’s time and money. Whereas, and I am
taking the broader picture, not just make this a Medicare focus. Is this really
a national, is it really HHS wide? You have the ability to weigh in with a
perspective and with a standing that I am not seeing a lot of other actors that
can do it.
So I would think about, and I think the point was brought up by Bill in
terms of your resources. I would also bring it up in terms of you will have a
limited opportunity to your target audience here. I am assuming as policy
makers, for your target audience, pick your four or five main things that you
really think are essential for them to consider in things like Meaningful Use
and how this rolls out and what this system will really look like in five or
ten years. And do not allow the kind of the lobbying and other things to limit
their options before they even test their options. And that is just I think a
great contribution that you have the potential to make.
MR. REYNOLDS: Okay, we have got Paul.
DR. TANG: I think this is an opportunity to breathe life into health
statistics.
(Laughter)
I meant that in a positive way. I mean statistics being what it is, the way
we gather data to produce statistics is always tethered to the past and
backward looking mainly because of the limitations of getting the data. So with
the opportunity of having data in near real time, I think we can transform
statistics into a living thing and we have heard this before from testimonies
about the bi-directional nature and providing real time feedback to influence
today’s care using the track record of what has gone in the past. That is one
point but it readily gives it a new flavor I think to write a new book.
The second piece is to reemphasize the data sources that we have talked
about here. It is not just the EHR. I know that was one of the things that I
think Dan mentioned that we did consider other sources and it is not only the
patient home-monitoring as an example, and home-monitoring does not have to be
like blood pressure and glucose, it can be just the fact that they got out of
bed. Those are really important pieces to outcomes and to the independent
living and that is what we are going to have to deal with for us baby boomers.
The other thing to tie together is like and H1N1 comes in as just a really
nice nexus of all these requirements and potential uses. So right now, in fact,
the county that I live in, we had, for whatever reason, had a real
mal-distribution so we got virtually none. So we could not immunize the county
of Santa Clara. Yet with an EHR presumably and this NHIN, couldn’t we know
where the high risk people are and actually we would use the maps that the CDC
produced to figure out what is the diagnosis of the person sitting in front of
you. Because when there was no seasonal flu, then the diagnosis of the person
sitting in front of you was by definition, the H1N1 flu. And that is how that
feedback in virtually real time affected the care of that individual patient
that day. So you could both affect the care of individuals, and then of course
spread the information that would contribute to the map, et cetera. But that
could also be used then on the distribution side for the vaccine instead of
distributing it by zip code. So I mean that is sort of just a live example of
how we could be using this in a real time way and with bi-directional flow of
information.
MR. REYNOLDS: All right you have got Mark, Ed, Tony and Blackford.
DR. HORNBROOK: I think the basic principle in intercepting EMR data is that
once it is entered, why doyou need to enter it again with another chance of
data entry error. So if you want something about a person’s utilization of
services or something about their clinical characteristics, something about
their diagnosis, you should be able to extract it out of the flow that goes
into the EHRs which as we have watched now is actually a system of systems and
inside each installation, you have to customize to that cluster of systems
because you are going to have a hospital system or two or three or five or ten
and several outpatient systems. But hopefully, inside an organized system they
are all linked so that they feed into each other.
So given that context and given the assumption that we think there is going
to be with the incentives, a growth of these electronic systems, you get down
to the question of whether you could think of a patient as a unit of sampling
and then develop the methodology to go discover all that patient’s electronic
systems and have that patient marked inside those systems to all of a sudden,
pipeline data to NCHS so then NCHS gets to accumulate all of those data feeds
and to put together a full record of that patients medical care and clinical
care experiences.
The other way to look at it, of course, is to look at clusters of patients
that sit around inside single systems that you could then negotiate with the
system to send data on some or all of their patients to fit in and then you get
to Paul’s perspective of if you have people who have elected to be in that
system, how can you put together a population based equivalent sample out of
sampling of systems when those systems were not randomly selected. They are
purposely selected.
But that has certain efficiencies because when you have a thousand people
sitting inside Kaiser Northwest to be in your sample, you have got one set of
negotiations, not one thousand negotiations. And I think that finally you also
have to think about and this is maybe getting too micro for you, but there is
the EHR that is the operating system probably built in months that optimizes
the exchange between patient and doctor as you are going through transactions
and then there is an extracted data set, the data warehouse, which is optimized
for group analysis on lots of patients. Not just one patient at a time. And the
nice thing about data warehouses is that there is usually some kind of quality
checking and some kind of consistency in defining the variables and some kind
of documentation around it and they are simpler than going into the 1600 tables
that make up the electronic medical record on an outpatient system.
So once you get into this notion of electronic medical records system on a
defined population, whether it be a single system like Kaiser or a sample of
systems such as the VA and Partners and other organized systems, you begin to
think about what is possible, not only for statistics, but also comparative
effectiveness research. If you can start defining ways to get at the nature of
the inputs and what is coming in for the way of care, what the patient’s
morbidities are, what is going on after that pattern of treatment, and
eventually of course some sort of outcomes and some sort of patient reported
outcomes.
So clearly when you are talking about cancer, you want to see some measure
of performance, even if it is just a performance index given by the oncologist
to get at some of your outcomes for the patient when they have gotten their
cancer treatment. I think that we need to decide whether we are looking at the
next five years or the next twenty years. And how far you want to push the
informatics that are possible today in the systems that are well developed
right now. And as far as I know the ones that have at least 10 years of
experience include Kaiser and VA, but there are of course many other companies
out there and there are some that probably have 10 years of experience with
in-patient systems. Or we want to look at the future where you are going to be
able to get at patient’s personal health records.
I heard the other day the first story of where PHRs are relevant outside of
international travel. And that was a story on MPR about patients in health
vault, putting their medical records or their claims data whatever they can
get, on to health vault and then sharing them with other patients with the same
conditions to check on what their doctors were doing for each other and
comparing patterns of practice. So it was essentially a patient self-help
mechanism through health vault. That is the first time I have heard of an
equivalent kind of social model of why you could create a PHR. It is a
systematic view of “Patients Like Me.”
MR. REYNOLDS: Okay, Ed.
DR. SONDIK: I have the feeling that what I am going to say is going to
overlap and so I apologize. I will not go on at length. I think that and I did
not hear the preface to this emphasizing electronic health records, so I guess
I think of this as not necessarily focused on electronic health records but
perhaps sources more broadly. I think it is important, and I am asking how this
differs from the shaping of vision, and what I would hope this would do is
focus on what the policy makers or decision makers need to know for the next
five to twenty years. What do they need to know to enhance health information?
Secondly, I would suggest, and I do not know exactly how to do this, but
that if this could be done in terms of some set of decisions as opposed to the
general need for information, maybe build it around some different types of
decisions. We have already heard some but it could be the at the moment
treatment decision. So how does this information link, for example, to what is
on NLM and what NLM cold look like in five years? To the physicians own
evaluation of his or her own practice or the group practice but focus it not so
much on information sources but on what decisions we want to make.
I think another area that did not come out, at least I did not get it in the
Committees was research and how this would support research. So what needs to
be done to enhance the capacity to support research? Could the, and I do not
know that there is an answer to this, but could the current situation with
breast cancer screening, for example, be helped by enhancing the health
information capacity? Could we know more? And I do not know whether we could. A
lot of sources went into that. What kind of infrastructure would be needed to
enhance this? Are we going to need new kinds of personnel or new kinds of
training? — This sort of thing.
To me, focusing on the decision side, the use side, even if it is not a very
broad range. You know, how much can you really deal with? You cannot get
everything. But I think that is the breathing life into this and the idea that
people are really using this. I love the other document. I like the big
diagram. I still like the big diagram with the context and in fact, I was at a
meeting this past week and I used that, but it is very static. It just sort of
sits there and says here are all the variables. And then what I did, was I put
up the Evans and Stoddard is that the people from Canada? And I said, what we
need is we need a dynamic model of this. This stuff sits here. We do not have a
production function for health. We do not know how to do that. We do not know
how to answer those questions.
So somehow if we can enhance our health information capacity for the
21st century so we can figure out and get a better handle on how all
of this fits together to produce health or how it fits together to enhance this
doctor’s decision when there is a patient in front or evaluate the patients
over time or a better handle on functioning. I think then we are really going
to have something. I think it is a very tall order but on the other hand, maybe
if it is focused on some small set of decisions, this can be done by June.
MR. REYNOLDS: Tony and Blackford.
MR. RODGERS: One of the opportunities we have if the timing is right, is the
fact that each state has now been given money to start statewide planning for
health information exchange infrastructure. It would be nice if this document
would focus on how to help states really create the requirements so that
population health information will be collected at the tap point. And I think
the tap point in the future, is the health information exchange organization.
That is where the investment of dollars is going. That is where the
accountability will be for moving data. That is where the data becomes fluid.
And that is really where we want, I would think if you want to avoid having so
many sources of data that you have to figure out how to communicate and
harmonize.
The top point is going to be the health information organization and the
exchange process that states establish. If we could inform that process with
the population health requirements early enough in the planning process, then
we could help those states and assure us that we do not have 50 different
solutions moving into 50 different directions. And I think that is a really
important thing for us to consider.
MR. REYNOLDS: Okay, Blackford.
DR. MIDDLETON: I guess the comment that I wanted to offer to Dan and Gib,
the team putting this together, is to draw the connection between, I think what
is going to be emerging as more and more clearly, the connection between kind
of our population assessment methods and techniques and data requirements and
the individual assessment techniques, but particularly to draw the connection
between quality measurement and decision support.
In a way, of course, decision support is moving from the population to the
individual, or really already is there and reports on the individual care will
increasingly be N of 1 reports as we get to very tailored and personalized
medicine. Your genome will dictate your care at a level, and thus your quality
report may be the quality of your care uniquely. Thus it is not a population
report, it is an individual report. I just wanted to draw out that connection
and think about in the 21st century vision how we actually measure,
therefore not only the health of the population but how well current knowledge
is being applied in decision support at either population or individual levels.
I think that is a somewhat different thing than what we have done historically
with statistics. But in terms of assessing health, it is perhaps the most
important proximal measure of how well we are advancing health, if you well. So
I just wanted to throw out that idea for a future vision.
MR. REYNOLDS: Paul.
DR. TANG: I would use another expression, but unfortunately, Ed did not take
the first one wrong in terms of breathing life because every time I hear the
work population, it just seems so last century. And I think this is the
opportune time to talk about how we are starting to slowly go towards, and
Blackford’s comments, of affecting care today. And I think that one of the side
benefits is then the data becomes more accurate, because we are going to reuse
the data.
And part of the legacy of it measuring the past is nobody really cares and
hence the data, there is no incentive to feed these reports accurate data. But
if we reuse it, like Ed was talking about and Blackford was talking about, and
then the other thing that Larry and I were talking about over lunch is this
whole maintenance and certification, if in one fell swoop you are talking about
how you measure your care and how you measure your professional capabilities
and performance and then how do you get rewarded for it so actually what Larry
was talking about is how they submitted data from the practice in order to gain
certification and then that organization, in turn, sent it to CMS to get the
PQRI reimbursement. So that is sort of the full cycle of life from
professionalism to delivering accurate care to getting compensated for
delivering care and not the volume thing. So I think that goes back to the
quality letter that we talked about which is something or somebody needs to
make that circle come together and then essentially you get all of it at once
instead of measuring the past, trying to affect the future and trying to get
rewards. It just seems like that sort of lends additional weight to the report
recommendation.
DR. W. SCANLON: I think in listening to everybody that we probably have a
multiple set of objectives here. And priorities that we probably would not all
rank everything that we want sort of out of this same order. At the same time I
think there is an opportunity here for a win-win for everybody and that is the
issue. How do you turn this into win-win for everybody? And a part of that
strategy in terms of doing that I think goes back to sort of what Mike was
saying about whom are we talking to?
If you live here in Washington, there is a 24/7 activity which is called
healthcare reform. They are making this investment because they are trying to
change the healthcare system. And so if you can ride that horse, you can
benefit. And so this morning’s letter is a perfect example of something that is
aimed at healthcare reform. It is we need measures or quality of care that is
being delivered because we cannot do reform without bundling, we cannot do
reform on paper performance. We cannot do sort of capitation methods unless we
have a much better fix on the quality of care. And it is in that context, if we
adopt a strategy that we want to do something that the people we are talking to
that have the decision making power to spend the money are going to listen to
because it is serving their needs and some of our interests are going to be
simultaneously served as well.
Jim in prior sessions has emphasized this point at different times. We had
hearings about data for health reform because he was concerned that this is an
issue within the department. How do we maximize data for health reform? He was
the one that talked about when we were going over the report in September; we
have to make sure we include health reform. So it was one of the objectives. So
I think it is that. We should not lose sight of that.
The second thing is, I think, we do need to think about the short term and
focus more on the short term than the longer term. And the electronic health
record is the most immediate short term sort of possibility. Personal health
records may be coming, but it is going to take a long time for significant
penetration and in terms of what the public sector care about, we are talking
about a public sector investment here, this managed populations are either
going to be the last or never going to be holders of personal health records or
they are not going to do a great job maintaining them. So we have to think
about where the public sector is in terms of this. And the public sector needs
to operate quickly and it needs to serve the bulk of the population quickly.
And so therefore, I think, what are we going to be able to do in the next five
years or ten at the max, on a broad scale that is what is going to appeal to
the people who have got the power to write the checks?
DR. FRANCIS: I think I am going to sound very last century.
(Laughter)
Because I really do think that there are reasons to have denominator based
data that have to do with overall population health, independently of the
quality of care measures that we have because it might just turn out. I do not
know but we have massive increases in the quality of care that are not
correlated very well with increases in overall health.
MS. GREENBERG: Amen.
DR. FRANCIS: We just might have the people who get care getting better care,
but not have that be correlated with overall improvements in population health.
Now that is an open question. I am not saying it comes out one way or another.
And of course I am here because of privacy and confidentiality not because of
data driven expertise but I would also say that I think the kinds of concerns
that you raise about privacy and confidentiality are going to look different if
what you are interested in is improving the target person’s healthcare. Then if
what you are interested in is knowing a whole lot more about how the population
is doing too.
MR. BLAIR: Leslie, could you give an example of what you are talking about.
Some people resonated to that and I felt like I –-
DR. FRANCIS: Yes. So one of the issues is, we might be getting better
healthcare of people who get into the system. Okay. That is one thing. Another
thing is we might be putting a lot of band aids that are actually very nice
band aids on people whose overall health is deteriorating. I mean, we could
have much better management of lead exposure and a whole lot more lead would be
an example of that. And I want to know, not only are we managing our lead
exposures well, but also are we in a situation in which our population is being
exposed to a whole lot more lead with deleterious consequences? So that is just
a separate question. It has been the traditional population or public health
question rather than the medical treatment question and I think that has been
part of the traditional mission of this Committee and what I think is that it
is wonderful to get better information about quality of care. I am not saying
that that is a bad thing. I think it is also wonderful to think about whether
we can use the quality of care information and all the other wonderful sources
of information that we have now on the population health question. But I just
think those are two different questions and the kinds of constraints on data
use might look different depending on which question you are asking too.
Anyway, that is all.
DR. TANG: Can I just respond to her comment?
MR. REYNOLDS: That will make six people mad.
DR. TANG: Well I just wanted to clarify that the whole last century comment
was not on what is measured but how it is measured. And so I am not tethering
it to just care for sure.
MR. REYNOLDS: All right. We have Don, Larry Marjorie, Judy and Paul who
jumped himself ahead.
(Laughter)
DR. STEINWACHS: A couple of thoughts on priorities for what they are worth.
It seemed to me if part of the metrics that we were talking about would be part
of every electronic health record and maybe part of meaningful use included
measuring weight and height so I guess not we have to measure what is the
circumference at the tummy and circumference at the hips and some ratio of that
and also smoking and some measure of physical activity. I had a meeting
yesterday and I suggested that maybe we ought to quit measuring blood pressure
every visit by every provider and instead try and use that kind of thinking to
introduce some other things that are now more critical maybe because we have
got high blood pressure recognition up very high. We do not have the treatment
and the control as high as we would like but that goes differently than do you
have it right now.
It seemed to me that there are some avenues here that deal with both
prevention in our society, that maybe electronic health records would help open
up data at the community level, at the state and local level. The other, which
is certainly I think you keep hearing from policy makers the concern and it is
not just the Medicare program it is Medicaid and many others that those of us
that have multiple, chronic disorders are swallowing up lots of the resources
in the system and so again, you say, well what do we put together for
administrative data from our surveys and other things we know that it is true,
but we do not really know a lot about what would you do to intervene? And what
are the problems that are local variations in that?
And again at the meeting that I was at yesterday at the IOM, a person got up
and put together, for a person with three chronic conditions I think it was,
took a guideline care for each of those conditions and put it into a what would
that person have to do every part of the day in different domains and it is
more than a full time job. And some of those were contraindicated. And so you
were not sure what would be done. And so we are doing things that are
essentially damaging probably out in the system because we still have very
little knowledge about the management of multiple conditions.
And so just to finish with one point, I saw this as very complementary to
what we are doing now in national surveys and in Hanes and so on it was not to
replace those, it was to try and step beyond that either in adding some
richness in some areas but also adding things at a more local level. The health
information exchanges to me seem to be the way to open up for the community so
that you could have some data that comes closer, but then you need the tools to
be able to understand the biases and how that relates to standards or
expectations we have.
MR. REYNOLDS: Okay we have Larry.
DR. GREEN: So I was thinking of our agenda. So we are in that 1:30 to 3:30
part of the agenda, right?
(Laughter)
MR. REYNOLDS: Yes, you are in the right place.
DR. GREEN: Enhancing health information capacity for the 21st
century strategic planning, and I have three points that I want to make. One is
this conversation for the last hour and 9 minutes has relieved me of some
skepticism I brought to the meeting today about whether or not the discussion
that we had in the Population Sub-Committee was on track, pertinent, would gain
traction with the Committee. What I want to consolidate here is the notion that
it seems like virtually every person at the table has something they want to
say about this. And it appears to me to be very cross-cutting, engaging and as
I listen to everyone talk, they seem to me to be talking about things that
matter that are pretty important and I heard Mike say that at the beginning of
the conversation that it looked to him like NCVHS was positioned properly to
take on some of the stuff we were talking about. I can repair that if I heard
him wrong.
So I want to weigh in and say that it sounds to me like this is really
important. It is in scope and perhaps probably timely with both short term,
like Bill says, but also some longer term implications. So from a strategic
planning point of view to some extent what we have been doing is a bit of an
environmental assessment. And I am going to quit being skeptical about it. I
think the Committee ought to move in this direction, together. I think there is
work for all the subcommittees to do and I suspect it is timely enough and so
that is the main thing I want to say. This is just another personal opinion is
if we do this, I would like to solicit everyone’s agreement that we are not
talking about an adjustment here or a tweaking of a healthcare delivery system
or a healthcare research system or a healthcare statistical system that is
performing optimally. And no one would logically expect a complete remake of
the way healthcare is rendered to not lead to adjustments and everything else.
So it is really a claim about the size of the things that we are talking
about undertaking over some decades of the NCVHS’s work. It has had the
opportunity to sculpt things, craft them, improve them, add to them a little
bit every now and then you sort of see one of these things that it is a jolt
that hits. This is a jolt. This is a very big we are talking about. Every time
I have heard Dan and Gib talk, they make me think when I go think about it, my
pulse rate goes up a little bit.
(Laughter)
And getting a little anxious because the material that we have in our book
here. This is pretty cool stuff and it is pretty challenging and it is not
small potatoes. So it is a big deal and one other thing that Dan said earlier
in the meeting that I want to underscore is I would also suggest if we go in
this direction we recognize that we are not leading the planet on this set of
problems. In my opinion, we are, in many instances fairly far behind others and
that a lot of the information and knowledge we seek, has already been developed
at least in part somewhere else and we would do well to open our eyes and ears
to things that are going on outside the United States.
MR. REYNOLDS: Marjorie.
MS. GREENBERG: I have so many things that I want to say that I am going to
try to be brief.
MR. REYNOLDS: Should I put a time on you?
MS. GREENBERG: No. It is always the last person who spoke. So just in
regards to that, we do have a report that Dan prepared several years ago about
international experience or thinking on use of electronic health records and,
et cetera for population health, so it is on the NCHS website and I can send
you the reference if you do not have it. I think we sent it to the Co-chairs of
the Population Sub-Committee. It is interesting the comment that population is
last century. The fact is we have never actually talked much in this country
about population health. We have talked about public health. Population health
is really global, it is Europe or it is Australia but it is not so much the
U.S. so maybe we are finally bringing it in to the U.S. vision in the
21st century, as opposed to falling back into old century thinking.
Regarding short term versus long term it is true that we have windows of
opportunity as Bill has eloquently described and we should see our
opportunities and we should take them the way they did in Tammany Hall days.
(Laughter)
MR. REYNOLDS: Marjorie, were you there?
MS. GREENBERG: Probably, probably. It is like I said; I have seen my
opportunities and I took them, and carpe diem and the whole bit. And although I
know I have been making you think about June 2010 until you are sick of it and
cannot wait probably until the celebration is over. Hopefully not, but the fact
is just like the quality letter, it is beginning. It is not an end.
And so we should think in terms of on the short term, what we need to do to
seize the opportunities that are out there and also to make a splash in June in
the sense of getting people’s attention. Yes and maybe influencing things that
really are going to be decided in the next few years, the next few months maybe
like with the quality letter. But I am confident that NCVHS is here for the
long term and so I think because we are still talking about the 21st
century we are not talking about only the first decade of the 21st
century.
So you can have a short and long term agenda and I think I would recommend
it. And I think you know you could put some teasers out there for the 20 years
in June, but do not feel like everything has to be done by June 2010, because
it does not. Now I did not pray, I did not even ask Leslie to make her
comments. I do not normally pray during these meetings but you did see that I
said Amen. But we met and you will hear a little bit more about this tomorrow,
but Harry and Debbie and Katherine and Larry and I, met with the five or six
former Chairs in Charlottesville right after the September NCVHS meeting.
It seems like a long time ago now, but it was really a very magical few
days. It was very exciting and confirming in a lot of ways and they all send
their best. But we did these individual interviews with them, Susan Kanaan
interviewed them, and so we were all sort of like in this conference room
snacking and talking while each person would go out for his or her interview
and then we had lunch together and then after lunch we had a round table
discussion and there was so much rich stuff that came out of all of this. Just
as there is out of this afternoon; you are a great group with a long history,
but the main thing that I happened to be sort of with them all, because I was
not being interviewed at the time, in the holding room and the main thing they
were talking about, whether it be John Lumpkin, Don Detner, Ron Blankenbaker,
Judy Jones what have you. We could do health reform and not improve the health
of the population. For every one of them that was there main concern and it
came out again in the afternoon.
So I think at least your former Chairs are very much on the same page as
Leslie. And I think that I also, of course, we want to improve quality of care
each one of us who gets care and each one of us who has a family member who
gets care wants their quality to be the best care possible. And it is so
confusing. I mean this whole thing that has come out this week about
mammograms, it is just a microcosm of first of all how much success we may or
may not have with comparative effectiveness research, but just how difficult it
is people saying well yes, it is find to say we should talk with our doctor,
but do you have seven minutes?
So there are just so many issues here but if we take our eye off population
health, and I am not saying keep it only on population health, but if we do, I
think it will be a mistake because I am hearing people talk about well we want
to supplement vitals and we want to supplement Hanes and I am completely for
that. But what you maybe do not understand is less than a year ago we were at
risk of not being able to capture 100 percent of the vital records in this
country. You know, it is embarrassing. I am going to meetings about civil
registration in the Pacific Islands and I am saying the U.S. is at risk of that
very same thing or of diminishing the data that we get out of vital statistics.
We are still at risk of not being able to continue the Hanes survey. We do not
know what is going to come out of the 2010 budget.
So I am just trying to give you a reality check here, that is you think
these population based data systems have some merit, do not take them for
granted. All the money is going into EHR right now. It is not going into
population health data systems.
And one final thing that I want to again welcome our new member and say that
he is right on and it is really the way public health people, if not population
health people are thinking on all those areas that do have to do with disease
and healthcare. How can we work with these health information exchanges so that
we are not going to all these people, all these physicians individually, all
these organizations, this registry, that registry whatever, so that there is a
hub that will not only serve clinicians and health planners and everything but
we will serve public health and population help. So that is it.
DR. WARREN: Leslie and Marjorie have already talked about a lot that I was
going to say, but I responded to one comment that Ed Sondik mentioned about
what the health of our population would be like going forward and then the
whole thing with quality care. So I could think that we had really good quality
care, but we have poor population health, and so the same comments that have
been made before came up.
And so I started thinking for the baby boomers, what would health be? It
would be the capacity to manage your fatigue, to manage your pain, manage any
kind of nausea problems you have, manage your weight, manage not taking in
quite so many substances you should not, being able to walk, being to engage in
exercise, being able to not as you age have confusion that is caused by other
problems and being able to cope with whatever health problems you have. I am
not sure we are measuring any of them. And so maybe we need to start looking at
what measures of health do we need to start measuring not indicators of
treatment of disease as we start looking at population.
MR. REYNOLDS: I would like to build off Judy with some of the things that I
was thinking. I am not totally familiar with the population health environment.
I just learn what I listen here. But as you think about it, maybe there are
some new populations that we are having to deal with as we talk about health
reform or any of this.
So for example, affordability, because if you look at affordability around
the country right now a lot of people look at the amount that you pay for a
premium and if you have got the wrong deductible and the wrong co-insurance you
just eliminated affordability. As you look at things like access.
I was in Des Moines, Iowa not too long ago giving a speech, and if you are
in Des Moines, Iowa after 6:00 walking around downtown and anybody you know
whether you live there or not gets sick, there are no urgent care centers,
there is no doc in a box, there is no general practice open. You are going to
the emergency room whether you do or do not have insurance, whether you live
there or do not live there.
And then you go through the quality we talked about earlier which really the
quality we are talking about is lifting all boats up so that every EHR, every
doctor, everybody captures some of the similar things and then when you think
of protocols and so on. Personal accountability that was brought up, that is a
population issue right now. Personal accountability, you know how many people
are really taking account.
You have all seen what I went through with my parents. Co-morbidities are a
population. Co-morbidities is absolutely a population in my opinion because if
you have got people that are running around with co-morbidity trying to figure
that out and trying to figure out what good health is, what good care is, and
what quality is and how you can make a difference is almost impossible unless
you are in a closed system like Kaiser, but you cannot do that.
So when you think of short term and long term, to me, I also want to make
sure there are new subjects coming up. So if we say we are tracking along with
this, so back to the thing, if populations are what they were X periods of time
ago, then I have a problem with that. If it is all the EHR, then I have a
problem with that. But we have got to make sure we consider whether or not
there are some new subjects that create populations that would not normally be
the kind of populations that people might want to think about. The baby
boomers, Paul said it. That is a population. It is going to be coming like the
red tide here soon.
And so I guess that is something that I hope we take into consideration
because some of these things are short term enough that when we think we have
fixed it let’s say everybody has insurance but if they cannot afford it in the
ways that I just mentioned, then everybody does not and we have kidded
ourselves and that population just has the same thing. So just a layman’s
comments in the middle of all of this where we make sure that we do not defend
old definitions. We do not run to brand new. We make sure we listen to both of
them and talk about what we really need to do and I think then when we create a
service that I am not hearing anywhere, and I think this table is probably full
of enough disparities in day jobs and actions and thinking that we might be
able to bring that forward as an asset that are not things that are being
discussed and I do not hear them in many other places. So Mark and then Sallie.
DR. HORNBROOK: AHRQ has the medical care expenditure survey and they spend
the time to find the episodes of care for a relatively small sample. Most of
the NCHS data systems around utilization are not linkable across people. So
when you look at a system closest to like Kaiser Permanente you can see
electronic communications before you go to the doctor. You can see the doctor
visits, the nurse visits, you can see the prescriptions, the drug orders, you
can see the lab orders, you can see the ER visits, you can see the home health
visits, you can see the hospice, you can see every lab test, and you can also
see the breast cancer patients who are not taking their Tamoxifin, which it
turns out to be a tremendously important issue about cancer prevention in this
country that has just gone by the wayside. And you can see the fact that people
get reminders for their pap smears, reminders for their mammography, and
reminders for their dental visits. I have integrated dental and medical
insurance so the two talk to each other.
So what is out there in the world is something that really does allow us to
start thinking about measuring health in terms of transitions. Transitions from
being perfectly healthy to somewhat higher risk, to being really high risk, to
being frail and to being at the end of life. Transitions between places of care
are also extremely high risk. We have done data to show that when you
transition more than three times a year your hospital use is about seven times
the average for an elderly person. So the transitions themselves are risk
factors.
And so there is something to be said for the fact that already in this
country is a major health system that meets a lot of the needs that we need to
do for this Committee. And somehow we are trying to patch up what the rest of
the system was doing wrong. That is independent fee for service with all the
gaps and everybody doctor shopping and none of the doctors talking to each
other. And I do not know what you are going to do. You go around to six
different pharmacies you can drug shop and nobody knows that you are drug
shopping. Whereas in my system, all the drugs are available to everybody at any
time unless you happen to go out to Walmart.
MR. REYNOLDS: Is that a question, comment or commercial?
(Laughter)
DR. HORNBROOK: All of the above.
MR. REYNOLDS: Sallie.
MS. MILAM: I would like to sort of build on Marjorie and Tony’s comments. We
are in a position now where we have all 50 states that are going to have health
information exchange. So the theory is that we will have clinical information
about the patient at the point of care. That alone will be transformative. We
should see changes across the whole system.
What we have not talked a lot about are some of the privacy constructs that
are necessary to make that happen. One of the things that we have not changed
is the whole interstate issues. I wear two hats. I am the Chief Privacy Officer
and I am the Director of our Health Information Exchange and I am looking at
issuing an RFP for a systems vendor and trying to go live by July and when I
think about the conversations I need to have between now and July with
surrounding states about the data elements necessary to include in an
authorization so that I can send information across state lines is
mind-boggling. That in addition to everything else we should be doing to go
live.
So the interstate issue remains a big challenge to really pushing
information around the country like it needs to happen to be patient centered.
So we help the patient get the best care. So we have those improvements in
quality and all of those reductions in cost. At the same time I am reminded of
the Center for Democracy and Technologies Report. Devon McGraw’s report and
Justine, I think you were a member of the group. What we learned from their
research is that HIPAA really does lots of research well. And research is one
of those things with all of this information now becoming electronic and
moving, research has to be accommodated. It has to be facilitated.
And we heard that HIPAA gets in the way. The privacy advocates do not like
it because there is a huge ability to re-identify all of the information. On
the other hand, folks also feel, the researchers feel like it strips out
everything relevant. Nobody is happy with HIPAA in the research environment. I
think these are a couple of issues that are really important to helping our
country get to its vision and really being able to impact care for the patient.
DR. CARR: I guess I have more of a question than an answer. I am wondering
if we can restate what it is that we want to achieve. I think we want to have a
statement of it in June and have it as a roadmap for where we are going. I mean
everything that has been said today is very compelling, but I am struggling to
focus a statement of what we are doing and what it will mean for us.
MR. REYNOLDS: Jeffrey.
MR. BLAIR: Okay, thank you. You know I am sorry. I could not see who was
across from me that made the comment.
MS. MILAM: Jeff, it is Sallie.
MR. BLAIR: Hi Sallie. Obviously we are struggling with the same thing and
from our understanding New York and Minnesota are too and others that we do not
even know about. The thing about it in my mind is I feel like we need some
additional construct as we do our work on NCVHS and as other Policy Committees
and FACAs are doing their work. I am not sure this is the right word for it,
but I will put this word out and as it falls short, maybe it will get
corrected.
We look in terms of criteria for setting our goals; better patient care,
better quality of care, better health. We look in terms of the possibilities
for health research and sometimes the technology gets out a little bit too far
in advance of policy. Sometimes the policy gets out too far in advance of
technology and there is conflicting efforts and they are getting to the point
where we are beginning to have gridlock. And I think what Sallie just indicated
is an example of that because we have all of the wonderful things that have
happened in the last several years to move healthcare information technology
forward, to move public policy forward, to get the funding and the political
elements in line where there is funding for health information exchange and
electronic health records for primary care providers and the ability to reach
underserved population, we have all of these things in line, and we have the
NHIN developed and we even have thought out the legal issues and with the data
use and reciprocal support agreements and we go down all of these paths at once
and there is a lot of coordination and yet there is like one thing which is
like a stone wall right now. How do I want to phrase this?
The reasons privacy issues have come to the front is because people have
gotten hurt when the privacy of their records have been disclosed. We need to
have privacy of healthcare. We need to have transparency, accountability, and
public trust. But one of the elements here is that each of the states has come
up with their own privacy laws. So here you have health information exchanges
that have the technology, the DRSA, everything in place to share information
with each other and our state, we require that we have written consent from the
patient before we can share information with anyone. Well within our state we
can do that but how do we do that with the VA which is already within our
state? How do we do that with other states?
This is just an example, and I do not want us necessarily to focus on this
particular issue but I am wondering if there is some way we could bring into
our discussions a more complete way of trying to coordinate how we move forward
so we do not have these disconnects that appear to be stone walls after an
awful lot of time and effort has been spent. So I do not know whether the word
is coordination or pragmatism or something but that is my call is that somehow
we find a way to do a little bit better job of getting these things in sync.
MR. REYNOLDS: Okay Justine has another comment and then I am going to turn
it back over to Bill and Don to say that they are going to fix this between
3:30 and 5:30.
DR. CARR: So I wanted to say that I found this reference the main
determinates of health and longevity are 40 percent behavior, tobacco, alcohol,
obesity, 30 percent genetics, 20 percent environment and public health, and 10
percent healthcare delivery in hospitals and clinics. That is from Brent James
talk. And I think that actually this is a good framing thing of how do we bring
all of these things together and find the intersection of prevention, genetics,
environment and healthcare delivery. And I think the electronic health record
may be, even though healthcare delivery is a small part of it, that that can be
a vehicle that contains information that informs these other areas that are
actionable. And I just wanted to mention one other thing that we heard at our
hearing. We heard from Healthy People 20, who also have a FACA who is dealing
with basically behaviors and healthy living and so they have an interest in
interfacing with us. So that is my thing.
MR. REYNOLDS: Well you guys put some slides up and it has turned into, I
think, a good strategic discussion so with that I will turn it back over to you
and make your final comments.
DR. W. SCANLON: Well I go back to what I said earlier that I think that
there is a variety of perspectives around here in terms of priorities and that
absolutely none of them have to be sort of off the table. In terms of where we
are going, again, Paul’s comment earlier about this morning’s letter, I kind of
see it as a bit of a model where you set out a set of objectives and all the
things we have been talking about today are some of our objectives. We are not
saying that any of them get discarded.
But then there is this question of how do we make the maximum contribution
to achieving those objectives. And that is in the context of what other
resources we have available at the moment and this morning’s letter meaningful
use is a resource that we have available. If we can influence it in a positive
way to affect some of the things that we care about, then we are better off.
And I think that that is what we need to do.
Some of this discussion here will, I think, raise this dichotomy between
sort of population health and healthcare and information for each. But there is
also a dichotomy, I think, between having information that tells you about the
nature of a problem and having information that guides action to deal with the
problem. And we need both. And some of what we are talking about with EHRs is
information to be able to guide action with very, very specific information and
very much real time information. You know Paul’s example of H1N1, six months
from now knowing that there was an outbreak in some county is useless. It has
got to be yesterday we knew about this.
And so the question is how we get there in terms of an intervention. And
that I think we do not want to set up among ourselves false asquites because I
think we are on the same page objective-wise. The question is how do we move
the ball the furthest along in the short term and again I will go back, the
reality is that they have got this somewhere between $19 and 50 billion dollars
that they are going to spend. We do not want them to spend it in a way that is
not helpful to the kinds of things that we are talking about and so we think
broadly and give them guidance along those long dimensions that will make that
investment better from our perspective. I think we are all better off in the
end.
So I think it is not just for us between 3:30 and 5:30. I think it is for
some of the other Sub-Committees that may have other things to talk about too,
but it is also to think about where you want to do and how you want to
contribute to this and how you want to influence this? Because it does, as
Larry said, it cuts across every Sub-Committee of this Committee. It really is
and that is kind of I think what emerged in our thinking and discussion in
September was this is a Full Committee activity. Population in some ways is
focused on population and statistics, but the Committee is not just focused on
that. The Committee is focused more broadly.
DR. STEINWACHS: I want to just add my two cents. I thought the idea of
thinking about those things that we might have an opportunity more immediately
have an influence or impact on and that, to me, was sort of this idea of what
we might put into a June report with some things that we could get together the
information that would be good guidance and useful that it might have an
impact. And so that is identifying some opportunities. And Meaningful Use puts
into that and some other areas that we can improve the information we have that
it needed to improve health as well as to evaluate and monitor health reform.
It seems to me those are, at least to me, the two priorities and they are
interrelated.
And then we have talked about a series of opportunities and barriers that
are longer term that need to be addressed and the structure of states and
health laws and health care that is state bound. So I would think if each of
the Committees could think about the immediate term. Are there some things that
are opportunities that we could bring information to bear in the very short
term to go into the June report that we think could make some differences? And
the information we have and capacity both improving health as well as
monitoring and evaluating health reform. We ought to try and go for those.
But we are also making an agenda it seemed to me when we talked about this
initially. This is a beginning of a century and this is part of trying to build
on this agenda that started back in 1999 I guess into 2002 in terms of vision.
So at least I thought on the population’s side that part of what we were laying
out was a longer term work plan that certainly will continue to evolve and
change.
MR. REYNOLDS: Marjorie and I want to make a comment and Paul wants to make
one and then we will get out of here.
MS. GREENBERG: Okay well I think this has been a really rich conversation,
of course, what we are going to do with it between now and the end of tomorrow,
not to mention June, is a work in progress. But I just want to go back to
something that Ed said because it resonated so much with that old talk about
old century, whatever, but that old camera vision that Lisa Izonni and Cathy
Colton developed, but that you start with the questions, what are the major
questions you have? And I think if you are talking about each Subcommittee
thinking how they would want to contribute to this, I mean I think that could
be very compelling and very sort of alive as to even six major questions, but
not so big that they are unanswerable.
Then you get into one of Justine’s matrixes. What can contribute? What do we
need to contribute to that? And then as Bill said, there is something for
everyone because some of the questions could be very well answered by
electronic healthcare data or some of them could only be answered by vital
statistics. But you do not want to start with the data sources. Then you say,
okay, these are the questions, then you look at the data sources and then you
look at the barriers and the challenges which include the things that you,
Sallie and Jeff were talking about. And then, what are some short term gains
here that we could get? That would be a way that you would have the big
questions out there even in June to build on, but you would also have something
underneath. And Cathy and Lisa, their work would still live amongst us.
MR. REYNOLDS: The thing I would add is that we talk a lot about the stimulus
money and Bill you are right, there is a lot of money that is there right now
but I think I know in North Carolina and I am sure all the states now are
thinking about sustainability. So a lot of this work that can come out of this
and others, everybody is still struggling with sustainability. So when that
money goes away, what does it look like from a sustainable standpoint? And so
any work that anybody can do between now and when they have to figure that out
and the federal money may or may not be there after this thing takes off, then
that is going to be really important as we build whatever we build that it can
be used because basically, you have got 50 sets of novices in many cases out
there taking this thing that got this money coming along with it and then
trying to figure it out.
So just grouping up in those states is a struggle let alone figuring out
exactly how you are going to make something out of it that makes it jump. So I
think that can be another part of our discussion because that sustainability is
where others other than what the government money is going to have to start
stepping up in and writing checks and doing other things and there are going to
have to be some stuff behind that or you are not going to see that happen.
Paul, last comment and then we are out of here.
DR. TANG: This is just for Marjorie and Leslie’s benefit. You may not know
that we are the quality of health Sub-Committee, not the quality of Care
Sub-committee. So we have been working on health for a long time.
DR. W. SCANLON: I agree with you completely and it is certainly an issue and
I sort of resisted saying this to you about why I am driven in certain
directions. The financial side of this is so dire that is you make hopelessly
optimistic assumptions, in 75 years half the economy is healthcare. If you make
a realistic assumption, in 75 years the entire economy would be healthcare. So
what we are talking about if we can make some kind of incremental difference in
terms of efficiency, huge savings that would pay for all kinds of IT, but we
have not succeeded in the past and that is why this investment now is so
important to try and direct it in a positive direction.
MR. REYNOLDS: I agree. Okay with that, thank you. Great input. Everybody
played. Thank you very much. Now privacy is in here and population is in the
other room across the hall which is 1404. Okay with that we will see everybody
in the morning.
MS. GREENBERG: Oh, those who are going to the dinner, there are several of
us who can drive I think, so back at the hotel. So if you are staying at the
hotel, join us. I have to get home to give insulin to my newly diabetic dog so
I cannot stay the whole evening either. But I am going to come and have a drink
and a salad or something. So if you are staying at the hotel, do stop by.
MR. REYNOLDS: The Subcommittees in the morning start at 8:00. Thank you
everybody.
(Whereupon, the full committee meeting was adjourned at 3:36 p.m.)