[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL HEALTH AND STATISTICS
March 1, 2012
Doubletree Hilton Hotel
8727 Colesville Road
Silver Spring, MD 20910
CASET Associates, Ltd.
Fairfax, Virginia 22030
TABLE OF CONTENTS
- Call to Order, Welcome, Review of Agenda – Dr. Justine Carr
- Updates from the Department
- James Scanlon, ASPE
- Judy Murphy, ONC
- Lorraine Doo, CMS
- Standards Letters – 1) Claims Attachment, 2) ACA, 3)
Standards/Operating Rule Maintenance Process — ACTION, Walter Suarez and Judith Warren
- Population Health – Socioeconomic Status 139 (SES)
Workshop, March 8-9, 2011 — Update – Dr. Larry Green, Ms. Sallie Milam, and Dr. Vickie Mays
- Preparation for use of Data After Transition to ICD-10
Code Sets, Dr. Carr
- Impact on Quality Measures, Floyd Eisenberg
- Mapping between ICD (9-CM and 10-CM) and SNOMED CT, Ms. Vivian Auld
- Adjourn Plenary
P R O C E E D I N G S (9:04 a.m.)
DR. CARR: Good morning everyone, and welcome to NCVHS day one. I am Justine
Carr, Chair of the Committee, Steward Health Care System, no conflicts. And to
MR. J. SCANLON: Good morning, everyone, Jim Scanlon, Deputy Assistant
Secretary for Planning and Evaluation at HHS, and I am the Executive Staff
Director for the full committee.
DR. SUAREZ: Good morning, everyone. I am Walter Suarez with Kaiser
Permanente. I am a member of the Committee, Co-Chair of the Standards
Subcommittee, and I don’t have any conflicts.
DR. WARREN: Judy Warren, University of Kansas School of Nursing, a member of
the Committee, Co-Chair of the Standards Subcommittee and no conflicts.
DR. MAYS: Vickie Mays, University of California Los Angeles, member of the
full committee, member of the Subcommittee on Populations and Privacy, and no
DR. W. SCANLON: Bill Scanlon, National Health Policy Forum, a member of the
committee, and no conflicts.
DR. WALKER: Jim Walker, Geisinger Health System, a member of the committee
and a member of the Quality Subcommittee, no conflicts.
DR. HORNBROOK: Mark Hornbrook, Kaiser Permanente, a member of the committee,
MR. BURKE: Jack Burke, Harvard Pilgrim Health Care in Boston, a member of
the Privacy and Populations Committee, no conflicts.
MS. DOO: Lorraine Doo, with Centers for Medicare and Medicaid Services, lead
staff to the Standards Subcommittee, no conflicts.
DR. COHEN: Bruce Cohen, Massachusetts Department of Public Health. I am a
member of the full committee and a member of the Populations and Quality
Subcommittees. No conflicts.
MR. SOONTHORNSIMA: Ob Soonthornsima, Blue Cross and Blue Shield of
Louisiana, a member of the full committee and the Standards Subcommittee, no
DR. FITZMAURICE: Michael Fitzmaurice, Agency for Healthcare Research and
Quality, AHRQ’s liaison to the full committee and staff to the Subcommittees on
Quality and Standards.
MS. MILAM: Sally Milam, West Virginia Health Care Authority, member of the
full committee, Co-Chair of Populations and member of Privacy, no conflict.
DR. GREEN: Larry Green, University of Colorado, partner with Sally on
Populations Subcommittee and members of Privacy. No conflicts.
DR. FRANCIS: Leslie Francis, University of Utah and currently at Oxford
University in the UK, a member of the full committee and Co-Chair of the
Privacy Subcommittee and member of the Populations committee.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, member of the full
committee and a co-chair of the Quality Committee and a member of the Privacy
Subcommittee, no conflict.
MS. GREENBERG: Good morning, Marjorie Greenberg, National Center for Health
Statistics, CDC, and Executive Secretary to the Committee.
MS. MURPHY: Judy Murphy, I think I am a liaison to the full committee from
the Office of the National Coordinator for Health IT, and no conflicts.
MS. SQUIRES: Marietta Squires, staff to the committee.
MS. FORRESTER: Rachel Forester, Boundary Information Group.
MR. LAZARUS: Steve Lazarus, Boundary Information Group.
MR. ALFANO: Bill Alfano, Blue Cross, Blue Shield Association.
DR. CORNELIUS: Llewellyn Cornelius, Chair of the NCHS Board of Scientific
Counselors and faculty member, University of Maryland School of Social Work, no
MS. COOPER: Nicole Cooper, staff to the committee.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics, CDC
MR. QUINN: Matt Quinn, National Institute of Standards and Technology, staff
to the Quality Subcommittee.
MS. KANAAN: Susan Kanaan, writer for the committee.
MS. QUEEN: Susan Queen from ASPE, staff to the Populations Subcommittee.
MR. BIZZARO: Tom Bizzaro, from First DataBank.
DR. CARR: Usually at the end of the committee meetings, I say thank you to
all of the staff and the liaisons, writers and other support. I want to start
at the beginning because they do just such a wonderful job of getting us
organized and seamlessly placed, with a lot of work behind the scenes. So
thanks to all of you.
MR. J. SCANLON: Good morning everyone. I have a lot of updates for the
committee today and I wanted to get right into the areas. I am going to update
the committee on the policy areas and plans and initiatives, updating from what
we talked about at the last meeting. Secondly, I am going to mention briefly
the initiative on data standards, data collection standards, largely
demographic standards, that the Secretary adopted a couple of months ago, as a
result of the Affordable Care Act, and where we are next.
I am going to talk a little bit about a data strategy review and White Paper
that the data counsel was asked to prepare based on an extensive review of the
HHS portfolio. And then fourth, I want to talk a little bit about a request to
the committee from the Department to create a subcommittee or working group on
HHS data, access and use. And I will talk a little bit about that in a minute.
So let me start first with the — I think I talked at the previous meeting
about the HHS strategic plan. We revised it, and it is now available on the HHS
website. But in addition, there are a number of plans and initiatives in
specific areas, and I will briefly describe the areas. If the committee would
like to get a briefing in depth on any of those areas, we can certainly arrange
The first is on tobacco control, tobacco use, prevention and control. We
have another initiative on health IT, prevention and services, on hospital
acquired infections, on race and ethnic disparities. We have a large report
many of you have seen on eliminating disparities in health status and health
care in those areas. You have seen the national quality strategy, which arose
out of the Affordable Care Act, as well as the national prevention strategy,
also arising out of the Affordable Care Act.
We have a couple of newer groups. One is looking at early childhood health
and development, so we will look across HHS at what the status is, and
coordination and gaps. We just published the Alzheimer’s National Plan, and we
got some funding for that as well, which was announced previously. And we have
the latest group on disabilities. I was looking at disability programs, issues,
gaps, including data, across HHS generally.
And what we have tried to do is have — these initiatives are usually
carried out through a working group across HHS and other agencies, and we
usually try to have a data expert or an evaluation expert from our group,
network, within HHS, to work with the group so that, number one, they don’t
create a lot of new, unnecessary things.
And number two, build in the concept of evaluation and moving along, don’t
just create a document that goes on the shelf, but try to measure what is it
that you want to accomplish, and how would you measure that. So that model
seems to have been established. These must be working. We have already had a
complaint that there are too many initiatives and people are suffering from
initiative fatigue. So maybe we will slow down a little bit.
Quickly, on the Affordable Care Act, Section 4302, Data Standards, you will
remember that the Act required a secretary to adopt data collection standards
relating to race, ethnicity, language, disability, and gender as well. So under
the auspices of our Data Council, we formed a workgroup.
We looked at what were the current practices and the surveys and
administrative data and so on, relating to these items. It turns out that we
had some standards existing already, and again, these are data collection
standards. The focus here is really on a situation where the individual self
reports whatever the question is.
We actually recommended a set of standards. We took the OMB Standards on
Race and Ethnicity and we blew them up a bit, so that we have additional detail
within the Asian category and within the Hispanic category. That is really what
folks have been coming to HHS for over the years, and we finally got people to
agree to this.
We also looked at, for language, we were asked to develop a standard for
language, and again, these standards, at least this first wave, were to apply
to our population surveys. So basically these are the surveys where you are
calling someone on the phone, you are sending interviewers out to people’s
homes, mail surveys as well, but you are asking the individual to respond in
terms of how these categories would apply to them. So for language, at least
for this initial phase, we developed a standard question on English
proficiency, which seemed to be really what the issue was for the most part in
And then we have an option, if folks want additional granularity, they can
add specific language questions. It turns out that Hispanic is probably 75
percent of the — if you ask what language, Hispanic, generally, nationally, it
turns out to be that. And we have tables that show in every state what the
other languages are.
The disability was a very tough one, as you could imagine, because it is
hard to find one question or two that would be suitable. And let me emphasize
that these are meant to be minimum standards. This is the minimum, much like
the minimum datasets that the committee was involved in previously.
An agency could connect as much additional granularity and as much
additional information about that subject as they would like. But you have to
include the standard, and you have to in some cases be able to collapse back to
So for disability what we ended up doing was questions that the census
bureau asks. It is a six-part question on do you have any one of six visual or
hearing or other kind of mobility kinds of limitations. And so that actually
seems to work well in some of the surveys, so we have got that as well.
Those are now adopted, and now we have asked all the agencies in HHS to look
at their surveys and come forward with a plan to include those in their
surveys. So you can’t stop a survey and do this, but what you can do is, at the
next revision or certainly for any new survey, you would have to include these
It seems to be going reasonably well. The result here would almost be a
minimum demographic dataset. I wouldn’t say this is all the demographic data we
would want to have. I could think of others — education, and some other items
that we would want to have. But this is the start of a kind of minimum dataset.
The Secretary also has the authority to look at additional standards that
she might want to adopt. And what we were asked to do for this first round was
to look at socioeconomic status. Are there standards or measures, data items
and so on, relating to socioeconomic status, that could be considered as a data
What we have done is asked the committee, particularly the Population
Subcommittee, to take a look at the state of the art here. I think we have a
hearing on the public meeting on the ninth, and so we will see what the best
practices are, and then are there ways to go forward in terms of
standardization, or at least best practices there.
And then further down the road, I think we are expected to look at, are
there — now again, these are standards that apply to a situation where you can
actually ask the individuals, so you have a fair amount of flexibility, to what
extent can we apply standards to administrative data, which would be largely
claims or enrollment.
Well your degrees of freedom are much more limited there. You can’t ask a
20-item question on race/ethnicity. Or maybe you could. So we said once we got
the agencies underway with this survey, we would get back to the administrative
standards as well and see what we might look there for.
And then I think the third leg here is, really to what extent could
standards along these areas be considered for electronic health records. And
again, there I think you would have much more flexibility. You could literally
— and that is largely a clinical setting, of course, where individual language
and individual disability status actually makes a difference.
Surveys, the interest is not so much in decisions about the individual
population. But electronic health records, literally you are interested in the
clinical setting and what could be done. So there, I think we are not there
yet. But I think ONC, Judy’s group, has some interesting in there, not
necessarily right away but down the road. And I think we might want to see what
And I think there are some standards, some of the standards organizations
have already looked at the demographic for administrative and for clinical. But
that is where we will be heading along those lines as well. We have also, just
to finish out that chapter, we have also been, as Ricky said, initially we were
asked to look at gender or sex as a standard.
And the conclusion was, that what was intended was biological sex,
male-female. And so obviously we had a standard there. It works reasonably
well, although there are some issues. So we went ahead there. But clearly the
department is being asked now to look at collect data on other populations that
sexual minority populations, I guess would be the word.
So lesbian, bi-sexual, transgender, as well. And so we would be willing to
brief the committee on this at any time. We had actually added questions to
test to the National Health Interview Survey on sexual orientation basically.
So we are putting those through some field tests. And if all works out well,
next January they will be included in the Health Interview Survey.
Transgender was much more complicated. I think all of you know, this is much
more complicated, much less frequent in the population. And so we didn’t have a
standard, and all we could really do was to, along a path of listening and
getting some experts in to see what kinds of data what might be appropriate.
We have had two listening sessions, and we are looking at where people have
tried to collect this before, and we will have to see where we go from there.
But that is kind of that set of ACA standards.
Let me turn quickly, because I want to save enough time for the Data Access
Workgroup. The leadership of HHS, including my boss and the head of CMS and CDC
and ONC and AHRQ and a number others asked our HHS Data Council to sort of step
back and take a look at our entire portfolio of data collection activities,
analysis activities, both survey and research data, surveillance data, as well
as administrative data.
And to see, number one, is everything needed, number two, are we doing this
the best way, number three, are we addressing data gaps that are high priority
in a coordinated fashion, are we employing the latest technology. Things tend
to get stale if they are not looked at. Are we turning the data around as fast
as people need it?
And then in addition, how can we help assure that these various capacities,
including surveys and surveillance and research data, administrative data and
electronic health record data, as we move forward, that we move forward in a
way that reinforces each other, that we do this in a planning coordinated way
where in essence the sum, we are getting more than the individual parts because
we have thought about it in advance?
And what are the kinds of pilot projects and other activities that we would
undertake to help assure that, particularly from within HHS. So the data
counsel — and I have given you at your table the executive summary of what
came out of that, and we can give a full presentation at another meeting.
Basically we looked at, number one, what were the highest priority data gaps
that people thought we had? And obviously there are a bunch of them, but the
first one really was the ability to monitor the implementation and the outcomes
of health reform. So we did a very deep dive there, and we looked at virtually
every one of our surveys and added questions and modifications to almost every
one of our surveys.
So now we are collecting data in the National Health Interview Survey, the
provider surveys, the behavioral risk factor surveillance system, the MEPS
survey, the insurance surveys, the employer surveys — what we can get from a
survey, at any rate, in terms of variables and data that we would want to see
to be able to measure health reform.
It wasn’t just insurance coverage, which clearly we need, but it is
questions about affordability. Even with insurance, are people getting greater
access to care? Are they having difficulty seeing doctors, specialists, making
appointments? Are emergency rooms now being utilized greater than they had
We also looked at the supply of health professionals and so on. So that
again, the biggest parts of health reform, as you know, really come into effect
in 2014, Medicaid expansion and some of the bigger exchanges on reforms. So we
are able to at least monitor now where we are, as health insurance coverage and
quality and affordability and access. Are these things improving or are they
stalling? What is exactly happening before 2014 and then afterwards? We will be
able to monitor that.
On the administrative data side, obviously the surveys are helpful and
certainly the provider surveys are helpful. For example, we have a survey to
ask doctors’ offices are you taking new patients, are you accepting Medicaid
patients and Medicare patients and so on, some other related basically access
Then in addition we looked at administrative data. So we have a number which
is largely enrollment, Medicare enrollment, Medicaid data, and as you know
HEDIS type data, and Health and Seniors Survey, data like that. But it is
largely from our own programs.
In addition we have a couple of projects that are looking at claims data
from multi-payer claims databases. So it is not just the federal program, and
it is not just Medicare or Medicaid. In my own office, we are trying to put
together a multi-payer claims database that would be available for research. It
is very difficult, as those of you who have tried this know.
And then we are looking at what will the Affordable Care Act bring, in terms
of administrative data. So again, there is the Medicaid expansion. There will
be the operation of the state health exchanges, or the combined federal-state
partnership exchanges. So again, we are looking at what kind of data will we
have from there.
And then there is this concept of a federal data services hub in which
states could access this federal data services hub to do some verification of
the Internal Revenue sort of tax information. It would relate to information to
provide tax subsidies and so on. Again, we are at conceptual stages there, but
we are trying to see what that will bring as well. So at any rate, there are a
number of projects that have been identified that we can work on. It is, if not
a blueprint, at least a pathway forward.
And in terms of electronic health records, administrative data and
population public health data, rather than just think of, conceptually, how
these fit we have actually started some projects that we would try now to see
to what extent electronic health records can help us with public health
surveys, with public health surveillance, with research and with some special
populations research as well.
These would actually be demos, where we would actually look at, try to see
to what extent can the provider surveys at NCHS, doctor surveys, hospital
surveys, emergency department surveys, with some others as well, to what extent
could we use EHR data. And could we partner with some of the — and some of you
around the table are working with us on this — from your electronic health
records, can we see what kinds of data could be used for those purposes.
And again, I think you know this better than I, electronic health records
alone without — no one is going to get the entire electronic health record
reported to them obviously. So thought really has to be given to what exactly
is the nature of the function or report for public health or research, or
whatever, that would be useful.
And is that data already, is that content already in electronic health
records? And in what form, and in what manner? And then how would you actually
demonstrate how it would work? So we have some very nice projects going along
those lines as well.
DR. HORNBROOK: Jim, are you talking to the FDA about the Mini-Sentinel
MR. J. SCANLON: Yes. They are one. And they come into play with the
multi-payer claims database as well. So this is an FDA project where they try
to — and our office actually co-funds that. We have access to a lot of plan
data to look at signals of drug safety, potential drug safety issues.
And so, number one, we have access to all of the Medicare claims data, the
linked longitudinal claims data. That is available for drug safety research.
And number two, the Mini-Sentinel project, which I think takes a couple of
health plans plus some other large databases, and what they can do is, it is a
query based system so that if it is hypothesized that certain drugs may be
having certain kinds of adverse effects, there are ways to look at health plan
data and other claims data to see to what extent it is being shown there.
Now again, it is usually just a signal and not necessarily definitive. You
have to be very careful. All drugs have some adverse impacts, obviously, and
not all are discovered in the approval process. So it is very tricky in terms
of what is a real signal and then what do we do exactly.
That is the idea of using big data for these kinds of purposes. So let me,
if I can stop there, I want to spend a little time on what we are asking the
committee to do in the months ahead. This is a theme that I think would
resonate with all of you.
All of these data that HHS has and that all of you have, researchers know
how to get at it, and public health folks know how to get at it, and we kind of
push a lot of this data out to make it available to others as well.
But I think there is certainly in HHS, renewed interest in organizing this a
big better, making it available in forms that can be applied to patients, to
serve patients, communities, health care community, really at local levels, to
So things like Hospital Compare, things that are helpful at a local level,
besides regional, and market area and national, I think there is an interest on
getting as much of this out to folks who can apply this at a community level,
at a decision-maker level. And we talked previously about some of the
initiatives we created — Healthdata.gov, data.gov, and the health data
initiative warehouse within HHS.
The leadership has asked if we could now, to give a little more serious and
continuing attention to these areas where — and again, for research we can
always make improvements. But I think generally researchers know the data we
have. We clearly can make it available a little easier. But to some extent that
is probably not the main worry.
The main worry is getting it out to a decision-maker level, and local public
health level as well. So what we have been asked to do, the first thought was
to create a whole new FACA that would look at data access and use of HHS data.
And so that obviously didn’t make much sense. We indicated that we had a
committee to which this was a fairly, already recognized theme.
But what we could do, we could create a workgroup or subcommittee. And here
the model is going to depart a bit from the previous subcommittees. When we
renewed the charter just recently, we have the authority to create workgroups,
and subcommittees, and not every member has to be a member of the parent
committee. So this is more the model of Judy’s and the Health IT Policy and
Standards Committee, where we have the parent committee, and actually several
of you are on some of those as members or as experts.
So the idea would be, certainly any member of the full committee that wanted
to be on that, we would certainly welcome. We would need at least one member to
relay this back to the full committee. But what we would try to do, because
this is an area where within HHS we are not particularly expert at this.
How do you get this data, how do you get these communities of the folks who
actually build patient-related data, for patient choice and patient
decision-making, consumers, local health planning and decision-making, how do
you get those groups together with our data resources, in hopes of a developing
tools and applications that would serve those as well?
So in a way it is kind of unleashing the data, data liberation again, as my
good friend Todd Park says. He says that with the French, liberation, it always
sounds more exciting. So that would be the idea. What we would probably do,
subject to details, would be probably five to seven folks — we could change
We would want at least one committee member to chair or to be associated.
And we would probably then find other folks. From this community that is a
little harder to find. And we would basically create a subcommittee or a
working group, quite like the Health IT Policy Committee and others.
Their focus would be on taking our data, taking other federal data, other
available data, and really thinking of how to move this out, to the developers,
to the applications folks, to the folks who could use it. So you would probably
want literally folks who do this in the patient world.
It is a little different. We haven’t had, I am remembering, a consumer
member or a patient member, for the committee, for awhile.
MS. GREENBERG: We heard from several earlier this week at the Quality —
MR. J. SCANLON: This might be a way.
MS. GREENBERG: Very timely.
MR. J. SCANLON: So what we could do — and we can do this now. I finally got
the funding to proceed with this, because it would be another subcommittee, we
would have to arrange logistics and all of that. We are thinking within HHS who
are the folks who would be suitable for this sort of thing. If you would think
of folks you know in the community, and certainly if you yourself would want to
serve, let us know.
The privilege of being a full committee member is, you can serve on any one
of these. But I think we thought we had a deficit. I don’t want to compete with
any other subcommittees, but Quality, for example, you were talking about —
again, we are all aficionados in all of this.
But the focus here is really on getting meaningful information. Not silly
information, meaningful information, to the folks to whom it could actually
make a difference — patients, consumers, communities, and other folks who are
And it is not easy. Todd Park would say that you want to create a system
where the developers, the folks who actually take this data and apply it to wet
applications and other applications, are a different set of people than the
usual public health research community, clinical research community, that we
deal with. Or even the Quality.
So that is where we are heading. If you think of tools like Hospital
Compare, hospital referral area kinds of dimensions, I think we published to
Medicare use and expenditure for hospital referral areas and local areas. But I
would almost say it is sort of the first chapter of how this goes. And where it
will lead us, we are not there yet. It will be a bit of experimentation.
DR. CARR: So Jim, it is a perfect segue to our discussion yesterday. I
wasn’t here Tuesday but we had a great presentation on meaningful measures for
beneficiaries. And out of that we talked yesterday about the long-term goal is
to create the perfect measures.
But in many cases there is simple data distance to the hospital gender of
the physician, things like that. We talked about just taking what is available.
And as you say, having people that are not even from health care, but really
who understand how people digest data. I don’t know, Paul, if you wanted to say
anything more about that.
DR. TANG: I think I will give report summary tomorrow morning. But it is
definitely very connected and related to what we just had. Our interest is
very, very deep in this.
MR. J. SCANLON: We can build on all the work we have done previously,
obviously. But I do think we, in HHS as a rule, we don’t really have good
connections to that whole community of folks who take the data and mash it,
mush it, and make it available. And we are, as I say, we are in the first
chapter of this.
We put information available on our health data indicators warehouse. And
there are limits to what we can put on the web, particularly in terms of this
kind of data, claims data and so on. We can put indicators there, but we can’t
really put micro-records there. They could be made available for these kinds of
That would be the general idea. The rest we would just work through. I
finally got, earlier this week, the funding to proceed, checked that through.
So that is hopefully where we will be setting.
It would be roughly five to seven members. There is no magic number, and we
would sort of put the group together and then we would start with the office of
the folks —
DR. CARR: That is great. So we can recommend some members? I think we have a
project ready to go.
MS. GREENBERG: Thanks Jim. I think I agree, I really think this is very
timely. I just wanted to make a few points. First of all, like all of our
subcommittees, our workgroups, this one would report to the full committee and
all of its recommendations would be made through the full committee.
So in that sense, although we might only have one full committee member
actually on the workgroup — and Jim is shaking his head, because I know you
have worked this way with a number of workgroups on the HIT committees, as has
Generally this is something that we did very extensively as a full committee
back in — I like to be the historian, back in the seventies. Then we were told
we couldn’t do it any more because it violated the Federal Advisory Committee
Act. And then by expanding the number of the committee, and of course our size
is actually in legislation.
But in the last several years, the Federal Advisory Committee Act was
modified and so it does allow this. But it also does require resources, of
course, if you start engaging beyond the membership. So that is why it has
taken a little while to work out. But as Jim said, it is actually very
efficient compared to standing up a whole new advisory committee.
And I would say that the Friday afternoon session that we are having
tomorrow afternoon with CMS on their new line of information services, is
actually kind of the pilot. This is just being done, of course, within the
committee, but we are bringing in a panel of experts to comment that we worked
with Todd Parks and CMS on organizing.
So I think I just wanted to make that connection for you as well. I am
wondering, Jim and Justine, do you think it would be appropriate to have a few
members of the committee — I wouldn’t want them to have to commit on the spot
necessarily that they would be the chair of this or the member — but to work
with us on standing this up?
MR. J. SCANLON: Very much so.
MS. GREENBERG: So we did discuss this with the Executive Subcommittee during
a call, because as Jim said, it was part of our revised charter for this year.
And also, we wanted to make sure of course that the chair and the whole
Executive Subcommittee was on board with it. But we might want a few people.
DR. CARR: So why don’t we suggest at lunch today two things. One is, at
lunch time if you have an interest, let’s caucus here in this room. And as we
are speaking about lunch, I notice we are being asked to fill out the lunch
MR. SOONTHORNSIMA: Jim, thank you for the update. For this particular
subcommittee that you are talking about, you used the word local or community,
getting the information back to the local level, the community level. That is a
very broad term, and I know the intention is good. And I wonder, is it the
focus of the committee, this new committee, to define that more clearly, more
specifically, more succinctly? Because local level means different things to
different people. It could be at the state level. It could be at a case parish
Secondarily, at the local level there are so many different activities and
focuses that they all have — for example, disease management.
Then when you start going down that point of care type surveillance type
actionable items, then you begin to get to the varieties of use, uses for the
localized level information, dated, disease and management programs. So is it
the intention of the committee to cull this down a little bit, to scope it
MR. J. SCANLON: I think it remains to be defined. I think what folks at HHS
have is a direction, aspirational direction, the details. And there are some
god examples of using the federal data or other data to help with this. But I
think it would be within the purview of the committee to give us some
You can’t do everything at once. So maybe there are areas of opportunity
that would be worth pursuing, and then later return to some others as well. You
would be advising HHS, through the full committee, but just in terms of the
thinking, you would probably have public meetings where you would hear from
consumers and others about what would be helpful to them. And maybe find out a
little bit, as we have, sort of what seems to be working in some communities.
And then, it would be up to HHS to get the data, the kind of data available,
if we have it, make it available. Who would it be made available to, what kinds
of products, what kinds of quality assurance would you employ? But that really
would be grist for the mill.
MR. SOONTHORNSIMA: It sounds more like a framework that you are trying to
establish. Because you can’t apply it to all —
MR. J. SCANLON: I think we go as far as we can. You are exactly right. A
framework, in terms of an area to work in. But to the extent that we could
identify individual projects or initiatives, I think we should —
DR. CARR: It sounds like two parts. One is a proof of concept. And I think
that is a trump, right now. And then we actually, with what we talked about in
quality yesterday, and actually following on our community data initiative, I
think we have two areas where we already have in depth experience. We have held
hearings, and we could apply that immediately.
And then I think that would then inform what is the potential of this, and
stimulate other ideas from there. But I am really excited about it, because I
think we are just poised to now take what we want and use it.
MR. J. SCANLON: I think it is a logical extension of where we have been. I
think many folks think the technology is there to assist us. We still have
privacy issues, obviously, confidentiality issues. And we clearly would want
privacy represented on the group as well.
MR. SOONTHORNSIMA: To follow on that framework, I think this is an exciting
opportunity. As long as we are not trying to push something down and expect the
local communities to accept and embrace, it has a lot of capabilities across
the country. For example, some of these all-payer databases that exist at the
local levels, at the state level or even community level, that we might be able
to leverage and learn from.
MR. J. SCANLON: We don’t have to reinvent the wheel. If things are working
well, the committee can be actually sophisticated about this in terms of we had
done this before, and not pursuing avenues that are really, no one would find
useful anyway. But yes, I think we want some indication that it was being done
or it could be done. But otherwise, framework wise, we would be open.
It would be nice to actually have some specific, at least some highlights or
some areas that we could say this seems to work pretty well.
DR. SUAREZ: I have a few comments about the first part of what you said,
then a couple of comments on the second part, the requests. So on the first
part, it is amazing and so exciting to see it. My head is almost still spinning
on the many initiatives, a lot of strategy out there and a lot of effort,
moving in several directions. Certainly a lot of them now better coordinated, I
suppose, through the work of the counsel and the work of HHS leadership.
One important point about some of the section 4302 elements. You mentioned
briefly about the role of the standards organizations. We have to really look
at the work that the standards organizations have done in defining for
standards — this is not just nationally, international standards purposes —
how to codify how to capture and use, at least codify and capture, the type of
data that we are talking about, gender and race and ethnicity and language and
Because I hope that really, at the end of the day we are going to have a
consistent, harmonized way of capturing and using that data for not just
population survey purposes, but administrative data purposes and clinical
So that is one important point, is really in looking at this, what are the
standards that the standard development organizations have adopted and are
currently incorporated into the standard, are being called for to be used for
in all the work that we do here, whether it is administrative transactions
under HIPAA, or whether it is meaningful use metrics under the military use
So that is one. The second comment is really a larger comment about the
overall data and information strategy within HHS. The IOM recently issued a
report to CMS about the transformation of the health information system, and so
they are modernizing and transforming their health information system. I
participated in that advisory group that provided that report.
And one of the recommendations was, the development of a health information
model. I think now, this is sort of the new mantra across health care
organizations, is within our organization do we have a health information
model? Because we do all so many data collection efforts, and we collect data
in many different ways and for many different purposes.
And at the end of the day when internally even organizations are trying to
do cross analyses between different data, there is always a lot of data
manipulation that has to be done, and data harmonization, I guess one could
call it. And so my suggestion would be, is the data counseling considering
looking at a health information model that applies across the board through HHS
and its agencies?
Since CMS is going to have to, and will be looking as well if they want, the
DA has a health information model in place. And then others are going to have
to look at the CDC and others. So is the Data Council looking or considering
such an initiative, developing a federal health information model for health
MR. J. SCANLON: What we have — the Data Council is part of it, but there is
a sort of an umbrella group called the Health and Human Services Domain IT
Council. And they are looking at — well, I don’t know if I would say a model,
but I would say a framework, probably, would start. But there is a group.
They are just getting it on its feet, it just met a couple of times. CMS and
others were all represented on the group. And it is meant, exactly, Walter, to
bring the IT infrastructure and assets in line with the business and public
health, and the human services goals and models as well.
They have had a couple specific issues they have had to deal with. But the
idea would be to approach it, as you say. They may not be ready for a model
yet, but they may be ready at least to start along the idea of a framework. In
other words, who is part of this, and what is the nature of the data and the
systems that they employ.
I think you raise exactly the right question. We sort of think of CMS as so
different from FDA, so different from NIH. And they really are. But on the
other hand there are a fair amount of, if you could have a framework where you
show they are all public health or human services, and they often have the same
IT issues, as you could well imagine.
And I would bring that up to the counsel in terms of, is there some sort of
an information model or at least a framework we could work towards.
DR. CARR: I want to make sure we have time for ONC and also CMS.
DR. SUAREZ: Just a very quick one. I think the key word that you said about
this new workgroup, just to use the right term, is it is making meaningful
information available to local level. So the two parts are the meaningful
information part is a key element, and then the local level is the other one. I
will mention it.
In my mind all this brings kind of a new way of saying the devil. The devil
is in the granularity here. So the devil is in what does it mean to think about
local level? Is local level local public health? Or is it also the local health
care providers that are able to access that data and use it for population
health management and other things?
The term, the definition of what is local is going to be very critical. And
so the suggestion I had, I guess, in terms of organizing the workgroup — and
this would be our first workgroup, I guess, NCHS, and we have that experience
with the standards of the Policy Committee. We are actually running out of
DR. CARR: So the recommendation? Do you have a recommendation?
DR. SUAREZ: Yes, I do.
DR. CARR: What is it?
DR. SUAREZ: The recommendation is, I think each of our subcommittees should
have a person in that workgroup, if possible. Because I think this workgroup is
going to be cutting across all the different —
DR. CARR: Absolutely right. So at noon, however is interested. Vickie, did
you want to hold your question?
DR. MAYS: I was going to say, it is in the recommendations so I can actually
just make it during that time.
DR. CARR: Thank you. I would like to then welcome Judy Murphy as our liaison
with ONC. Thank you. We are delighted that you are here. And we will turn the
agenda over to you.
MS. MURPHY: I think you have a copy of the slides. They are missing a little
something because they don’t have the color. What I did, because I think Chuck
was your liaison before and he hadn’t been here for awhile, because he didn’t
work at the ONC for awhile, and then I am not sure what update he last gave.
I tried to make this a pretty comprehensive update, but I will go quickly.
You have the resource in the slides and I can certainly at the end answer any
questions that you have got on specific topics. Basically, as we think about
2012, we have had a big IT meeting last week for Health Information Technology
called HIMSS, or the Health Information Management System Society.
Let’s just say there were 37,000 people there, and they were all very
interested in Stage II, which came out on Thursday. So it was kind of an
interesting time. But one of the things that I want to preface this whole thing
with was, there is this thing in the air now that seems very different from
previous years. It was exemplified in the conference, but it has been
exemplified in a lot of the different meetings and things that I have been at
It is not that meaningful use and the adoption of electronic health records
is behind us, by any stretch of the imagination. But there is this sense of
inevitability of it now. It is like it’s happening. It’s not like there is this
whole thing about, oh, doctors aren’t going to want to do it. But it is more
like, okay, it is happening and it is a matter of time. So that sense of
inevitability was in the air.
And that has been important to us because from the ONC standpoint we have to
start paying attention to some of the other really important things beyond EHR
adoption. And again, I am not saying we are letting up on EHR adoption. In
fact, we have got a big meaningful use acceleration going this year.
But I do want to say that we really do need to turn our attention to a
couple of things that you already started talking about this morning. And the
first thing is really establishing the link between the use of health
information technology and quality safety, and truthfully, the whole triple aim
and health care reform. And so really starting to look at the research, and
look at the ticks and ties, if you will, between the use of the technology and
what the impact on patient outcomes actually is.
The second thing is consumer engagement. Now that we have got at least the
beginnings of the doctors engaged, the next frontier is really to get patients
engaged. And the purpose of the patient engagement is not to have them be using
the technology, except as that is a hook to getting them to be engaged in their
So the whole concept of patient as partner, and empowered to actually be at
the table with their providers, making decisions about their care and things.
So again, technology really just being the hook that is going to do that.
Because information is power, and once we get them the information that they
need, we think as a community that they are probably going to feel more
And the last thing and not the last, because it is probably one of the
greatest pieces of work, and that is true interoperability and health
information exchange. When I use that term I want to differentiate very clearly
from the program report I am going to give. When I use the term health
information exchange here, I am using it as a verb. Meaning, we are moving the
health data around the way that we need it to be moved around, and sticking to
the patient as the patient moves around.
So whether they are in a pharmacy or in an ambulatory care setting or in a
hospital, that we in fact, have the data that is needed to make decisions at
that point of care, longitudinally and in all our venues. Sometimes we get
caught up in worrying about the individual program that the non-health
information exchange. I want to say it is bigger than that.
And really being able to establish the standards and move that data around,
of course, is paramount. And that ties in again with the patient, because part
of the interoperability is being able to get the information to the patient.
With that, I will start my slides. What I was hoping to run through — and
again, a lot of the data is on the slides so I will just try to point out some
highlights. I am going to briefly talk about the websites, our campaign for
putting the I in Health IT, give an update on meaningful use, data is power as
we have already talked about this morning.
And there are some surveys that we have been using internally to look at the
adoption of health IT, and one of those is the American Hospital Association
Survey Data. So I will talk briefly about that. The Health IT Resource Center,
Health Information Exchange, Beacon, Workforce, Innovation, SHARP and eHealth.
And again, hopefully just hitting the highlights of all of those.
If we go to the next slide, some of you might be real familiar with this
particular website. It is the one that we have been publishing a lot of our
information on related to the programs that ONC and in some cases we paired up
with CMS, particularly around the incentive programs, to give all the
information off those programs. Where did the grant money go, who had them,
direct links to, for example, our workforce training and those kinds of things.
Instead of, as we go forward, linking everything to this website, because
what happened was, this was really the ONC website, and as I kind of alluded to
some things with other departments, and offices within HHS. And it makes more
sense from the end user standpoint to create a website that harmonizes that
So especially as it relates to meaningful use, we are, as we go forward
around stage two meaningful use criteria, going to be use the healthIT.gov
website more. And if you haven’t been there, again, the differentiation is
healthIT.hhs.gov is the previous one. This one is healthIT.gov. It has two
tabs, providers and professionals, and patients and families.
Under providers and professionals, the target there is really toward
practicing folks that want to understand more about health information
technology. So the target is not for health information technology
professionals, but for practicing clinicians.
And then the next one over is for patients and families so that they can
understand how they can become more engaged. And what you will see in the very
near term is a third tab. And that third tab will be for the health information
technology professionals. A lot of the information related to the notice of
proposed rulemakings that went out last week related to our stage two criteria
will be posted there.
And the intent then is to try to harmonize all the information between the
two programs — the actual EHR incentive program that CMS published their NPRM
related to, and the ONC NPRM, which relates to the CMS one, and describes the
implementation specification and the standards and the certification criteria.
We got a lot of complaints over the last two years related to the hopping
around that you had to do, the CMS website, the ONC website, and the FAQs
correspondingly. So there was minimally four places that you had to go. And so
we are really trying to focus on getting that all in one place. So watch for
that in the not too distant term.
Then a little advertisement, totally shamelessly, if you want to go to the
next slide, and that is the putting the I in Health IT campaign. And this is a
campaign that we have been focusing on again, the practitioners themselves or
the clinicians, nurses and docs, as well as patients.
Here is a couple of examples of advertisements that you might have seen in
magazines related to the practitioner side of the house. And if you go to the
next slide, I have got, I think, two examples of the patient ones. These are
patients who have agreed. They feel they are health IT advocates based on their
involvement in the use of health IT.
There are vignettes of them both in the written format and in the video
format on the healthIT.gov website, along with many other ones. And so, again,
the idea is to start to show the general public that there are people that are
empowering themselves through the use of Health IT in that patient space. And
so the second example is Donna Dryer, another one of the patients.
Then if we go to the next slide, I am going to hopefully remind you that the
programs related to the meaningful use of electronic health records are
multifaceted. We tend to focus a lot on the meaningful use category. But just
to remind you that there are lots of other categories.
This particular framework is a diagram that was published by Dr. David
Blumenthal when he was the National Coordinator for Health IT. In December of
2009, when the first stage one Notice of Proposed Rule Making first came out,
and it was his attempt, and I think it stands well still today, to show the
harmonization of the different programs that are being administered through ONC
If you look down the left hand side in pink and yellow and blue, you see the
different programs. So we have regional extension centers, we have got the
workforce training, we have the actual incentive program itself as well as
grants for health information exchange, standards, certification work, and
privacy and security work, as well as research.
All of those actually feed into the adoption of electronic health records
and the exchange for the purposes of meaningful use of electronic health
records. And it doesn’t stop there. Some of you I know are involved in this
implementation like I had been and still am, in the ONC.
And by the way, I have only been in the ONC for three months, so it’s not
like I am this long-term person. Prior to that, I had actually been doing
implementation of electronic health records for over 25 years. And in 2011 I
had actually shepherded seven hospitals and about 350 physicians to stage one
meaningful use. And they had received their payments before I left, so I was
happy about that.
The work is hard. And so sometimes what happens is, we end up focusing on
the technology. And so that bolded yellow box is a really important box. We
actually do have to get folks to the use of electronic health records in a
meaningful way. But just to really, really, really remember that it doesn’t
That again, the importance of the electronic health record is as the enabler
for the bigger box, the big, yellow box on the right-hand side. And that is
actually impact patient outcomes, and look at things like rural health and
disparities. And really trying to enhance the quality and the outcomes of the
care that we have been providing really in the nation.
I am going to give a quick update on meaningful use. In the meaningful use
category we have really seen some great strides in the last couple of years. So
some really good statistics. Fifty-two percent of office based physicians had
said that they intend to take advantage of EHR incentives. We are trying to
push that up, but we are proud of at least that much. Again, that sense of
The NAMC survey — and I don’t remember what that stands for, but that is a
survey of ambulatory providers — actually showed that self reported physicians
said 40 percent of them are today, in 2011, using electronic health records in
their practices. And that was up from the survey in 2009 of only 20 percent
saying that. So a doubling in two years is very, very, very significant.
And if you saw this in a graphic format, it kind of goes like this and then
it goes whoop, so it is really that hockey stick kind of idea. So 85 percent of
hospitals based on the American Hospital Association Survey, or AHA Survey,
said that they intend to attest to Meaningful Use by 2015. Hospital adoption
actually has also doubled in the last two years, 16 percent in 2009, to 35
percent in 2011.
And 2,000 hospitals and more than 41,000 physicians have received over $3.1
billion in incentive payments as of the end of 2011. And so of course that has
gone up already. But I kept with the 2011 statistics, just to keep it all tied
to on here.
So if we go to the next slide — I wish this one looked a little bit better
in terms of the Medicare and Medicaid eligible provider EHR incentives.
Although we have gotten out a lot of money, there is a lot more money that
could be given out. And that, especially because this is part of the stimulus
bill and we are trying to make sure we are stimulating the economy, the
denominator when we talk about eligible providers is pretty big. So it is over
So what happens is, even though we have paid 29,335 physicians, you can see
how that ranks out on the thermometer. It is not all that good. Now, that being
the case, there is a good chunk that have registered, so there is an additional
chunk of folks that have registered, and if you kind of eyeball that, it looks
like we have at least got one-third of the physicians who have either
registered or have been paid.
In 2012, CMS and ONC have banded together and set a goal of getting to
100,000 by the end of 2012. Now we intend to kind of blow that goal away, but
that is the official goal that we have agreed on and passed on to the White
If we go to the next graphic, I like this one a lot more, on the next slide.
This is the hospital one. If you look at this, you can see that one-third of
the hospitals have already been paid, another third are registered with intent
to be paid in the not too distant future. And again, we are looking at a
denominator here of about 5,000. So this really looks a lot better, and the
journey is moving a lot farther and quicker.
If we go to the next slide, this graphic kind of shows that 16 percent to 35
percent that I was talking about earlier, with that hockey stick. There is a
little bit of a differentiation between what hospitals call when they implement
a basic EHR, versus when they implement one that has clinical decision support
and computerized physician order entry, and maybe a bar coded medication
So that is reflected in the blue, which is the comprehensive, and is still
quite small. But again, even that number has doubled. I mentioned the
conference that we were all at last week, and it seemed like everybody was
there since it was such a big conference.
That particular organization does have a staging, not just basic and
comprehensive, but they use a seven stage criteria. And again, they were using
numbers like three percent, four percent, having this more comprehensive for
their top stage. And that is breaking loose now, and there are a lot more folks
at that level.
Their percentage, and I think they are a little bit behind our data, was at
7.5 percent of these more comprehensive or full electronic health records. So
again, I think we will see a lot of movement in that hospital space in the next
year as well.
So no conversation about meaningful use would be complete without talking
about our excitement last week. If you go to the next slide, the two stage two
NPRMs were published last week. The CMS one was published on Thursday. And the
ONC was published on Friday. We had hoped to get them out together, but there
were some actually formatting issues on the ONC one so it took a little bit
And so those are both available on that healthIT.gov website. The CMS rule,
again, talks about the EHR incentive program and lays out the stage two
criteria, along with a lot of other information, and is like 465 pages long.
The ONC rule is a little bit shorter than that. I think we are at 189 pages.
And those are not the official Federal Register versions. The Federal
Register versions will be posted on March 7th, and of course that is when the
60-day clock starts ticking. So it will be 60 days. I actually haven’t done
that math, but somewhere around May 7th will be when the comment period will
There are three ways to comment — the traditional mail snail mail, which is
not recommended but is available, the electronic, through regulations.gov.,
that is the preferred method, and actually we have comment templates out there
which helps organize the comments.
And it not only helps the sender of the comments organize their thoughts in
the categories to answer the questions that are asked throughout the NPRMs, but
it also helps us on the receiving end, since we do of course read all of the
comments it is important to have them organized in a good way. And then of
course they can be hand delivered to the Department of Health and Human
If we go to the next slide, one of the things I wanted to address, because
you are going to start seeing a little change in vocabulary, similar to some of
our Health IT vendors, what they have done instead of naming things in one way,
they change their naming convention a little bit. In the standards and
certification criteria, we decided to make things a little bit simpler.
So instead of using the multiple numbers that we did in our 2011 edition,
which was actually our stage one edition, we are going to just 170.314 will be
all of the comments, if you will, or the standards related to that. And you can
see the numbers there. We had to increase a little bit in our individual
criteria, but you can see one was 41 and the other one was 50, so it is not
But we are going to start using the term 2014. And one of the reasons for
that, you may have been familiar with the fact that we did have, and took, the
opportunity to update some of the stage one criteria through the NPRM that was
published last week, both on the side of CMS and on the side of the ONC.
And so rather than link that, and just call it stage two, we thought it was
better to link it to the year that it is going to be put into place. And so we
are trying to remember to refer to that as the 2014 edition instead of
consistently referring to it as stage two. I am confident that the industry is
still going to call it stage two, though, so that is more of a mechanical thing
that I am sharing with you, so that you knew what it was.
We will go to the next slide, and this is the only other thing I am really
going to talk about, about the NPRM, because I think it is a really important
concept. One of the fairly large criticisms we got about stage one,
particularly related to the certification and the ownership of the software,
was that you recall maybe that were core menu set criteria. And of the menu set
you didn’t have to pick them all. You only had to pick a subset.
But the guidelines and certification criteria basically said that you had to
own the software for all of the menu set items, even if you were only
demonstrating a subset of the menu set items. That wasn’t real well-liked, and
it didn’t make a lot of sense. And quite frankly, it caused some consternation
with people having to actually buy modules of vendor product that they had no
intention of using at that time.
Now you also might recall that many of the menu set criteria have become
core this year. However, with the certification criteria we are now saying that
an individual site that is buying software only needs to buy those pieces of
the software that they are actually going to demonstrate their meaningful use
So they do not need to own all of the software to run all of the menu set
criteria. They only need to have the software to run the menu set criteria that
they are selecting and demonstrating meaningful use on. So that is a pretty big
And along with that change, we have coined a new diagram. We have this
concept now, and this is no relationship to the blue button, even though it is
health information technology and that is what we are talking about. Blue just
happens to be the schema that we are using, based on the slide deck.
So the base EHR, and you can see the definition there, is that core
technology with the capability certified to meet the definition in our NPRM
proposed rule, of the base EHR. And the idea here is that many people are
amalgamating different vendor modules to put together their base and their
core. And we wanted to call that out and say that is perfectly acceptable. You
do not have to buy this all from one vendor. And that, in fact, competition is
So that base EHR is called out with criteria. We are actually working on
some slides where in one graphic you will be able to actually see the criteria
that relate to the base EHR. But for now, just know that there is a subset of
the overall meaningful use criteria that are in this base category.
And then there are some that are in this core category. And core, again, are
ones that everybody has to do. So it is the same definition as with stage one,
that everybody does them. And again, this is the EHR technology with the
capable certified for those core menu sets. And the ones that anybody would
select to achieve.
And then last but not least is the third ring, if you will, and that is the
menu set criteria. And here is where you would just be selecting a subset, and
only probably in some cases owning the software for that subset. So this is a
graphic that is actually in the rule, and we are using it in our slides. So I
think it will — well, now I am trying to remember if it made the rule, because
this was one of our formatting problems.
You will see it in a lot of our slides, at any rate. I will entertain
question on that at the end. I will just kind of keep going and I will try to
speed it up a little bit here. So the next slide references the Health
Information Technology Resource Center. And this was actually the resources
that were used by those regional extension centers, which you may recall were
And there were 72 of them. Some states had more than one, some shared with
the state. But there are quite a number of them around the country where they
got money to help physicians who were in small practices or in rural areas and
didn’t have the resources to select and implement health information technology
So the idea was patterned after the agricultural and regional extension
centers, and the idea was to get some resources out there for those folks. Well
this is the resource center that amalgamates, if you will, all of the concepts
and ideas that have been proliferated as best practice through the regional
I draw your attention to it today because on the right-hand side of this is
the National Learning Consortium. And what I mean by that is these Health
Information Technology Resource Centers resources have only been available for
the last two years to the regional extension centers themselves. So they were
buried on a website that were used only by the regional extension centers.
And in 2012 that is going to get opened up to everybody. Everybody and
anybody who thinks that they would like to see those resources, identify those
best practices, et cetera. And so this is real active work this year, and you
will be seeing and hearing a lot more about this as 2012 progresses.
Turning ourselves on the next slide, to Health Information Exchange — and
again, this is the curve — we have a really, really good story to tell in
terms of e-prescribing. And I think you probably know that there has been
groundwork laid long before the American Recovery and Reinvestment Act related
So this journey has been going on a lot longer. But when you look at this,
it is a really positive, positive story, and we do see a lot of adoption in
2010 and 2011, and again, that hockey stick kind of thing.
The biggest issue that we have got right now is that we want to and need to
see this kind of a curve for some of the other exchanges that we are trying to
proliferate. The two that we are really trying to spend a good chunk of time on
are lab exchange and the transition of care summaries.
The transition of care summaries is that that goes between venues. So it
could be hospital to rehab facility, it could be hospital to long-term care, it
could be hospital to ambulatory provider, it could be ambulatory provider to
consultant provider. We’re really trying to see these two areas proliferate in
2012. So again, you will be seeing a lot more activity related to that.
On the next slide we have just got some ideas. These are early ideas in
terms of the focus and then the approach. This is not a completely vetted out
plan. But as I mention in my introductory remarks, about half the people at ONC
wake up every day and say we have to make this happen this year.
The last two years, a lot of work on developing the standards, a lot of work
on making sure that the standards could be implemented, a lot of prototyping,
all the kinds of things they have to do to set the infrastructure.
But we have a new rule in ONC this year — if it is not implemented, it
doesn’t count. So this is the year of implementation, and we really have to
take this and really get it pushed out, so that it is ubiquitous in the
So if we go to the next slide, the Beacon Communities. These were our
shining lights. These were the communities that were looking at doing two
different things. The first was, showing the country how a community can band
together across disparate organizations that many times were actually in
competition with each other, how they could band together with public health
and provider health in the area that the beacon was in, as well as cross vendor
kind of population sharing.
And so there has been a whole mess of work done on really looking at how do
they pull themselves together, how do they govern themselves, how do they make
decisions. So that is the one area that they have been kind of shining the
light on and showing the way.
The second is to actually then take health information technology and show
the world, particularly us in the United States, how this can make a difference
to patient outcome. They were only stood up about 18 months ago, and so there
was a lot of time and energy spent on infrastructure. So similar to the health
information exchange, it was a time of get everything going, get everything
down. Now we are just starting to see outcomes.
So if you go to the next slide, some very preliminary outcomes. Each of
these communities, in some cases they have got more than one focus but
typically they are focusing on a disease state. So cardiovascular and diabetes
are recognized here. It is two of the things that they have kind of bit off.
How can they make a difference to chronic disease management in their
community when they band together and actually share information between and
across all of the different kinds of facilities? We are going to be hearing a
lot more about the beacons in 2012 as well.
If we go to the next slide, we have two different workforce programs. One is
the community college consortium, and this is over 80 colleges in the country
that were organized around five different consortiums. They were supported
through a curriculum development that was done in the country as well, except
that they had to stand their programs up and accept their first students before
the curriculum was done.
And so of course there was kind of a lot of dual threading, like always
happens when we have to get things going. The numbers here are staggering
though. You can see over 9,000 people have already completed the community
college program, which is a six month certificate program. And we have another
23,000 folks that are in flight.
Our biggest issue with this is linking them to the marketplace. In 2012 we
are really focusing on looking at internships and more vendor specific training
so that when they hit the road at graduation and they get their certificates,
that the job market is ready to accept them.
When we look at the university based training programs on the next slide,
the numbers are smaller. These are also in some cases certificate programs, but
they are typically for credit programs, and many of these students do stay and
complete a Master’s. So again, smaller numbers. We have 825 enrolled, and 567
who have already graduated. And you can see a little bit of capacity across the
So if you know anybody who is interested, we are accepting students into
this program and they can get up to $10,000 in a year subsidy. So it wasn’t
just subsidizing the creation of the programs on the part of the college, it
was actually subsidizing the students. The last program enrollment will be this
fall, so we are really looking to make sure that we fill those capacity slots.
Winding down, we have some innovation things going. I know when we say there
is a lot going on at HHS and at ONC and at CMS, it seems like there is a never
ending set of challenges and challenge grants and mini-challenges. Even our
surgeon general has gotten into that and issued a challenge for an app, if you
recall, that would help her get moving and exercise.
And so there is a never ending set of things going through. And the good
news is that people just rise to this challenge. A lot of these grants are
smaller. They are $5,000, $10,000, those kinds of things. But it really gets
attention to what we are trying to do, and that is get the marketplace moving
and get proliferation. And of course, some of these really do tie in with that
consumer empowerment that I was talking about earlier.
So again, over 100 new applications have been developed and winners in these
innovation challenges have received over $195,000 in prices. And you will be
seeing more challenges this year.
So our last program that I am really going to talk at length about, a funded
program, if you will, were our SHARP grants. They were represented in that
diagram across the bottom in the research area. There were four SHARP grants
given. And then have been working for about two years, and they have about two
years left on their grants as well, to wrap up in the fall of next year.
They are looking at what I consider some of these sticky wicket issues that
we have got. The first one is working on security. The second one is looking at
patient-centered cognitive support. And if you go to the next slide, we have
got health care applications and the secondary use.
So they have not been publishing a lot of their work yet. They are typically
doing some presentations and things and talking about their work. But we are
actually going to see the results of some of their research and things later
this year and early next year, so stay tuned on that.
Not actually a funded program, but the Consumer eHealth. I couldn’t do a
talk without talking about the Consumer Pledge Program that was kicked off in
September of last year. The consumer program is about organizations, whether
they are insurance organizations, nonprofit professional organizations, health
care delivery organizations, pledging that they believe it is important to get
consumers involved in their own health. And specifically identifying what they
can do to move that agenda.
And I will just give you one example that I am involved in personally. I
didn’t mention, I guess, in my introduction I am a nurse. And one of the things
that the American Nurses Association has pledged is that they are going to give
education to nurses.
So that nurses can understand what this is all about, and turn around and
teach patients. So providing patient teaching information on the American
Nurses Association website, for example, that folks can download and use in
their own settings, is a good example.
Other folks have said we are going to get our own constituents to start
using. So employers, for example, of health care organizations have said we are
going to get our own employees using personal health records, because what
better way if they use it themselves and they see the power of it themselves,
they will be turning around and advertising it to their patients when they work
with their patients.
Some of the logos — 250 organizations have already pledged. And if your
organization or you, even individually, haven’t pledged, please consider going
to that website. It is on the healthIT.gov website, and you can look for the
A lot of ways that we communicate, still not good enough, I think,
dissemination is one of the things that we have to continue working on,
communication and dissemination. But again, here are some call outs to the
different things that we do in terms of Facebook and Twitter and YouTube, et
With that I will close. I know that was kind of a whirlwind tour of
DR. CARR: That was great. I am going to make some edits to our timeline. We
were initially going to take a break at 10:30. We are moving that to 11:00. We
will take 10 minutes of focused, specific questions, and at the break there
will be time to have more conversation with Judy.
But I want to make sure that we have a full 20 minutes for Lorraine. She has
some important updates. So we are going to ask everyone to stay on task here so
that we can have the best dissemination of information. I see Mike with the
DR. FITZMAURICE: Judy, I notice you didn’t mention anything about the
standards and interoperability initiatives project, its goals, its expected
output, and just longevity. But that’s not my question. My question is about
intellectual property. Providers use proprietary quality measures, they used
certified electronic health records standards.
Do they or their vendors have to pay, say, a quarter for each time they use
it in the course of the year? Do they have to register their identifying
information to each intellectual property owner? And that could be a lot of
them. Has ONC considered buying the US rights from the intellectual property
owners to aid adoption? And by the way, I liked your presentation.
MS. MURPHY: Thank you. No. We haven’t really talked about intellectual
property. In the case of SNOMED, for example, we have had a great partner with
NLM and NIH. I know that there are some other things that they are considering
doing, but in terms of ONC, no.
Now most of the proliferation of our standards is through the S&I
framework work, and I didn’t specifically address that here because that is
pretty nitty gritty kind of work. But that is public. All the work is public,
not just the participation but the proliferation of the standards as well. So
there is no individual ownership. But I am guessing you are not necessarily
talking about that work. You are talking about some of the real —
DR. FITZMAURICE: Part of that work, too — I am a player in the S&I
framework. But it’s how to increase adoption by reducing the costs, or making
the costs known to everybody.
MS. MURPHY: I do understand your point. And as I mentioned, the S&I
framework is developing standards right now at a rate that we cannot get the
adoption. And so that’s why I am saying we really have to spend time now. They
are going to keep developing, because the adoption part of it is coming through
a different arm of ONC.
But we really do need to focus on adopting what has already been developed.
And that’s why I focused on that today.
DR. FRANCIS: This is just a question about whether you have any
distributional data about adopters. So I would love to know just a little bit
more about whether there are regional distributions of interest, whether there
are rural, urban, whether there are concerns that there are populations being
MS. MURPHY: All of our stuff, I think all, maybe there is some stuff that is
not, is on healthdata.gov. And there has actually been a proliferation of folks
that have picked it up and sorted it six ways from Sunday. With just very
little Googling, you could find a by vendor, you could find by state. Some of
the stuff we proliferate ourselves and actually post it out there in that way.
And we being CMS and ONC. In other ways, it has been picked up by the
industry. We don’t publish vendor specific data, but of course we have that
data. And so we publish the dataset, and then the industry takes it and does
their little sortings and creatings of different kinds of spreadsheets and
And the vendor data I think was posted probably about 30 days ago now. And
there is just a proliferation of folks that picked up on that and showed that,
as just one example. And so we are probably more into providing the data, and
letting people do — doing the different cuts.
But on the same case, in things like rural, we do look at that so that we
understand whether our regional extension centers, for example, are
specifically doing the work that we need done in those areas.
DR. FRANCIS: So maybe you could give me a substantive answer. Is there any
reason — I have seen a lot of discussion — for concern that there will
actually be a widening of racial and ethnic diversity with respect to quality
of care and so on, as meaningful use moves along?
MS. MURPHY: You know, it is on everybody’s agenda to keep track of that. As
I mentioned, that link to quality directly right now is not as tight as we
would need it to be, to know even directionally where that is going.
Your point is well taken, though, that as we start to put in health IT, if
there is a direct link, which we believe that there is, and if the health IT is
not put in fairly, that would actually create more disparities.
So do we know that data today? No. Do we need to keep our eye on that ball?
Yes. And I think again, 2012-2013 will be important years to be doing that in.
DR. COHEN: Thank you. This was really comprehensive. I am interested in the
SHARP research projects related to patient centered cognitive support. One of
the things the Quality Subcommittee is focusing on is measures that matter, and
the focus here is on providing better information to consumers, to help them
make better decisions.
My comment is can you hook us up with the specific projects so that we know
what they are doing, and perhaps get feedback from their results.
MS. MURPHY: Absolutely. That would make sense. I think, again, if I sound
pragmatic it is because this is a pragmatic year and my roots are pragmatic,
and that’s why Farzad brought me in.
One of the things is that the SHARP grants have kind of been holed up doing
their research, and we do need to create some hooks now into the community that
are tighter than maybe were last year. And so I would be delighted to be able
to do that.
DR. GREEN: I want to thank you, Judy, for a great report. I would like to
revisit something the committee heard a couple of years ago that relates to
this monitoring of uptake and adoption. So maintenance of certification from
the physician perspective is here to stay. It covers all medical specialties.
There is variation within the specialties about where they are in adoption.
To go to Leslie’s question, I want to illustrate how this might be useful to
the ONC. The American Board of Family Medicine is the second largest board in
the country. For the last seven years it has monitored self-reported adoption
of EHRs, and it has been a six to eight percent increment by year and it just
went over 60 percent in 2011.
These numbers are not huge, but you are talking about 8,000 to 10,000
physicians per year and it is not a sample in some ways. They can’t take their
exam unless they answer all the questions on the survey. So there are no
missing data, in answer to this question.
And it has been validated against NAMSA’s and the provider survey and the
numbers come out at the same spot. So it seems to me that ABMS represents
perhaps an unseized partner for you, to go to work on monitoring this uptake in
a way that could be very accurate.
When they are subdivided by demographics of where they practice, Leslie’s
question, from a family physician point of view, family physicians who practice
in two settings are trailing — rural underserved areas, and if they practice
in a community or a neighborhood that qualifies for having community health
centers and that sort of stuff.
The good news in that is they catch up in about 18 months in this training
program. They are coming in behind but they are training.
DR. HORNBROOK: Judy, thank you for your presentation. This is just a very
technical question. Do you have any Blue Sky idea about when we could expect
smart phone texting and images to be secured so that people can have secure
conversations with their doctor, with their smart phones?
MS. MURPHY: I have no idea. I could try to follow up. There are other people
in the room that I know follow standards. Jim?
MR. J. SCANLON: I have no idea. I don’t think so.
MS. MURPHY: It will be in our lifetime. How about that?
DR. CARR: Thank you so much, Judy, and now Lorraine, to hear from CMS.
MS. DOO: Judy, also thank you. And one good thing is that Judy covered a lot
of the things that I would say on meaningful use. It was very helpful actually,
it was great. CMS staff were at HIMSS along with ONC.
And I remember we had gotten one — there was this frantic activity of
trying to get these regulations up to the Federal Register so that they could
be on display at least before the end of the conference.
I think we were all in about 25 different sessions that we were supposed to
be speaking at, all on meaningful use and certification. So it really was kind
of a race to the finish. I remember at one point I got a note from a CMS person
who said, well I’m sitting on stage at the Town Hall, there are two of us from
CMS and there about 11 people from ONC.
And she was saying, so between us, we surely can cover all of this
information. It was a tremendous joint effort and a lot of support. So that
said, we did indeed get to display our proposed rule on the stage two, which as
you know the secretary had signaled late last year, or maybe in the fall, that
instead of being the 2013 that it would be in 2014.
That is really based on a tremendous amount of experience now with the
meaningful use program, with the issues pertaining to the incentives. And I
think the important distinction, that even though we collaborate with ONC on
many, many components of the regulations, we are actually getting the
registrations and attestations of the meaningful use, and going through that
process with the providers.
So in terms of data on the numbers and some of the breakouts, you also can
get that from the CMS website. I am assuming that there is going to be a link
to that. So Leslie, for some of the questions that you have in terms of the
states and who is doing it in the states, much of that will be there, as well
as the amount of money that has been paid out for the eligible professionals
DR. CARR: You guys refresh that monthly. Just to a point, they refresh that
monthly. So if you really do want to keep up, that is probably the best source
of truth. Whereas healthdata.gov is kind of like the big data sets, as compared
to distilled down the way you guys do it so nicely.
MS. DOO: And sometimes I think, there is a sense of when can we have it,
when can we have it. We are trying to make it as accurate as we can so that
there aren’t misperceptions of what actually is happening. And a couple of the
things in our proposed rule, we have listened very intently and met with lots
and lots of physician groups for some pretty specific issues that were
presented over the past year.
One of those was something related to a hardship exemption, so that those
providers that either have issues with internet access, the type of data that
they are collecting, the type of patients that they see, that there is some new
language in there specific to hardship exemptions.
There is also language in there based on specialties, and so some of the
criteria have been modified to accommodate what those specialty groups see and
do, so that they will not be excluded from the program. And then in an effort
— and I think we have all talked about what is not so much harmonic
convergence as these storms that are coming up, of so much happening over the
next couple of years.
But in an attempt to align some of the reporting requirements with PQRS,
ePrescribing, the accountable care organizations, even the other requirements
from the Affordable Care Act, the Medicaid Reauthorization Program — all of
these things coming on a convergence together, and the proposed rule tries to
accommodate some of that.
We, like ONC, are going to be putting our rules out also in a word format,
so that you will have both the Federal Register, with the three columns, which
drives everybody batty when they are trying to actually reference things, but
also in Word so if people needed to cut and paste and do something with the
tables in their own machinations, that will be available.
And I do love the idea of the comment templates, and I wish the Federal
Register would adopt that. It would make it easier for industry, and certainly
much easier for the folks that are going to be reading all of those. And the
deadline for public comments is May 7th. We know that we will get most of them
on May 7th, and there will be a mad flurry to look at them.
So we obviously encourage all the input that we can get, so it really can be
the best final rule that is possible. None of us have been idling, so a couple
of other issues for CMS, and particularly for the Office of eHealth Standards
and Services, and Medicare and Medicaid, is that thing called 5010 and D.0 and
The compliance date was January 1st. Last fall we did get input that there
were some issues with testing and a concern about readiness. And so within the
authority that we have, because we can’t change compliance dates, we authorized
if you will something called enforcement discretion.
And that really enabled all covered entities to continue working through
their testing issues without fear that a complaint would somehow result in
either a penalty or a sanction, or having to create a corrective action plan in
the midst of trying to come up with those final things.
We have seen progress, and both the private sector health plans as well as
Medicare are seeing now upwards of about 90 percent of the transactions coming
in in the updated versions. We have heard very, very few issues with the
pharmacy standards. We have heard many more related to the other
administrative, the other claims, not as much with eligibility or some of the
other actual transactions. Mostly with claims and with the remittance advice.
So Medicare has been working very actively with the MACs, the Medicare
Administrative Contractors. There were some issues with telephone access,
availability. They were being inundated as people were trying to finish their
testing, and get assistance on understanding rejections. They have rectified
some of that by adding staff and adding hours.
We in our office in OESS have been asking folks who are having technical
issues and either can’t get through or can’t work things out with the other
payers, to let us know. So trying to take away this fear of retaliation, or
somehow the fear that a complaint will result in a negative effect, is, rather
than using this complaint, to say let us help provide some technical
And we have gotten some of those in. We have been able to do some
interventions in a good way. And we actually have decided to coordinate with
WEDI on a fast-paced technical assistance forum. I think what we are going to
try and do is, every week have a web-based call where people can be on the
phone but they will have submitted their issues in advance through e-mail.
We will have subject matter experts looking at them to see if there are some
general things happening we can get answers to. It will also give people
protection, though, of privacy, if they don’t want to say I am having an issue
with insurer X, that we can address that, and make the phone calls, and see if
we can’t get the parties together.
So again, trying to get rid of this fear and help solve the problems. We had
our first meeting related to that earlier this week, and that actually should
start next week. And if any of you are hearing things, please feel free to call
me because I will get people to the right place. We think it is working.
We have been trying to work through some issues with the American Medical
Association, MGMA. We have gotten some letters from Congress people. Obviously
we take those as seriously as if anyone was calling us personally and made
outreach to those providers, to try to understand what the issues are.
So it is moving along. We have been asked to extend the enforcement
discretion period. We are not likely to do that. And I think part of what we
want to promote more, and we will do this on our website, is ask us for help.
If we keep delaying a date, it becomes meaningless.
But rather, let’s figure out a way to make the process work better, to
figure out a way to have the testing work better, to identify these issues
collaboratively and not just keep changing dates around when we never really
are fixing the issues at hand. So that is likely to be the outcome of that, and
just sharing with this group what the response will be. But we are not just
saying figure it out. We want to be 100 percent engaged.
On the Affordable Care Act, there are a couple of regulations coming out.
The interim final rule with comment, adopting a standard for electronic funds
transfer, which the Standards Committee in NCVHS recommended to us, did come
out at the end of last year. Comments are due on March 12th. So if you do have
comments, including we have already received some saying it was beautifully
written, which you never get.
That is a testament to this young man, Matthew Albright, who I think really
did an outstanding job of making something pretty darned complicated,
understandable. So March 12th, if you have comments, yea or nay.
We are also working on the proposed rule for the plan identifier, and that
is expected to come out the end of this quarter, and that will be a proposed
rule. There is also an interim final rule with comment adopting operating rules
for the electronic funds transfer and electronic remittance advice. That is on
schedule to come out by July 1st. And again, that will have a 60 day comment
And then the compliance certification of explaining what we mean by, and
what the legislation meant by certification of compliance for health plans. So
that it is specific to health plans, having to certify that yes, they can
conduct all of the transactions in the appropriate format, using appropriate
standards and using the adopted operating rules.
The first date that goes into effect is actually the 31st of December in
2013. There are probably going to be — it’s January 1st of 2014. We think
there was just a date issue. But that will be coming out also this year, so
quite a bit of regulations for everybody to be reviewing.
And then we have had lots of elephants in rooms lately, that thing called
ICD-10, which I know somebody will ask about so I might as well address. There
has been some concern, obviously, about all of these things that are required
of industry. They are all coming together, and they are all coming together
around the same time.
When this regulation was published in 2009, I am not sure meaningful use was
a glimmer in anybody’s eye. Certainly the Affordable Care Act had not been
published at the time, ARRA hadn’t been. So what knew then, and what we know
now, certainly are different. But it doesn’t mean it is any less important. It
is tremendously important, and I think everybody agrees on that.
It is what is going to be possible because with something like ICD-10, we
cannot have enforcement discretion. There aren’t going to be options available
to us. I think the Secretary and the Administrator took the concerns,
particularly from the provider communities, very seriously, and did signal that
we will be looking at using the regulatory process to add some time for
There is very little I could say about when anything will actually happen,
but I think it is important for industry to understand that we are working on
that, that something will come out very soon. It will be this year, and it will
be sooner rather than later.
But we must go through the regulatory process. We have already gotten — as
you can imagine, industry is all over the place on this. This is exactly the
time we need to really figure out how to do a better simulation lab.
Or really how to make this testing work so that we don’t end up where we
were with 5010, to we are dead in the water, to I am going to have to redeploy
all my resources. All over the place, all legitimate. You can tell that people
have given this thought, even though it has not been a lot of time to do
We have been very lucky people have actually given us numbers. It has always
been very difficult to get actual data on what the costs would be and what the
ramifications would be. People have been very forthcoming, which is critically
important for us to be able to justify whatever the decisions will be.
So that is underway, and we are going to have some messaging on our website,
just so that people do not struggle, not knowing which way to turn. So there
will be another message coming out from the secretary with some specific
talking points on that. And I think that is probably the best I can do related
to ICD-10 here.
Do you want me to talk about the summit?
DR. CARR: Yes.
MS. DOO: In our world we work very closely with an incredibly dedicated
group of people from the standards organizations, operating rule entities, and
the code seq committees. They are all doing Herculean work to help health
information technology and EDI transform. That said, there is a tremendous
amount of both personal and professional investment in the work.
And what has happened is, I think in part because we have the standards,
there has been this new entity that has come along called Operating Rules,
which are new business requirements that the entities have to work with. And
industry has begun to express pretty emphatically, they don’t know what table
to be at.
And we can’t really say, well you need to be in all of them all of the time
because it isn’t possible, and again, when we just talk about the convergence
of all of the work. And it is very, very important for all of these
organizations to demonstrate that they are working together and collaborating,
and that they have a common purpose just as you do, and just as we do.
Because industry at the bottom line just wants to be successful in the most
efficient way they can be. So actually NCVHS had gotten some letters from a
couple of folks at industry saying frankly, this situation is untenable, we
just don’t understand who is going to be first, who is on top, who do we listen
to, where do we participate.
At CMS we had also gotten some phone calls and some emails, and so we
thought it would be a good idea to get the groups together. So rather than
trying to fix something outside of that, or to have a conversation with person
A and then another one with B, you get off, you feel great, and then the same
thing happens. We thought let’s get everyone in a room, be transparent and open
and build consensus on how we can move forward and have a win.
So CMS — I will use my OESS hat — held what we called a listening session,
or a summit, on Tuesday, actually, and we invited all of the leads from X12,
NCPDP, HL7, all the code committees, so the dental, CPT, ICD. And the operating
rule entities, CAQH CORE was there, NACHA was there. I hope I’m not leaving
anyone out. And Walter and Justine came.
The purpose was to really identify what we were hearing, be honest about it,
and come up with some strategies for going forward, knowing we wouldn’t fix
everything in one day. There also were some issues related to the American
Dental Association and the Code Content Committee that had been considered
disruptive by the health plan community because they have changed the voting
process and who is actually at the table for the votes. So there had been some
concerns about that.
So we just brought everyone together to work on this process. And I used
something, which probably you all know something about, called dynamic
governance. It is a little bit like a Robert Rules of Orders, or you could say
it’s even like the Quaker consensus process. But it ensures that you fully
respect everybody’s opinion and their perception and their needs.
And you essentially, in summary, you have a proposal and you go around the
room and you either get consent on what the proposal is, or if someone has a
paramount objection which they cannot live with, then they have the opportunity
to explain why they have that objection and what they could live with.
It doesn’t mean they will be ecstatically happy, but they can live with it.
So it really is for the good of the order. And so we had a proposal. I didn’t
bring the big sheet but I will try to rephrase it. But we went through this
process. It took us all day. It was very, very powerful. And I will tell you
that at the end of it, everybody applauded. For themselves and for what they
had done, to have this consent of a proposal for moving forward.
We have a lot of work ahead of us, but we have the right group. And I
applaud every one of them for being there for the full day. Some of them came
from Texas. But we had 100 percent participation and 100 percent engagement, I
will say, by every single person in the room. So hats off to them. Some folks
are here today, but it was a pretty remarkable process, and I am sure Walter
and Justine will have something to say.
But essentially, what we are going to now work on is — and it particularly
has to do with how the operating rules and the standards will work together —
a coordinating body that is really perceived as objective, to help develop and
coordinate the subject matter expertise that needs to be present to develop the
operating rules that will support the standards, so that there is synergy and
there is not conflict.
And I think that was kind of the biggest thing, the issue we have had, is
what are these things called operating rules, and what could they really do to
the standard, and really needing to understand that those things are
synergistic and will absolutely help the standard work most effectively.
And so now the task will be to put that in writing, everyone, send it out to
everybody and say this is what we wrote down with all of our cross-outs, and
then work on who that might be and what the criteria for the subject matter
expertise and the development of those rules will be.
Because we have very little time left. The next set of operating rules has
to be adopted by January 1st of 2014, and there is nothing written on a piece
of paper yet for what has to be, what they look like. So I thank the committee,
actually, for the support in having that listening session, and for doing it,
and frankly for being there for the whole day.
DR. CARR: Thank you, Lorraine. It was a pleasure really to be there. I
agree, the commitment of everyone in the room was palpable, and Lorraine’s
leadership was exceptional. So a great job. Paul?
DR. TANG: Thank you, Lorraine, for the complete update. I have a question on
IC-10, and it is really an is this being considered kind of a question. I am
known for my lumping, but I also am a lemonade person. So we have this delay,
for better or for worse. What can we do that can turn it into lemonade?
We all recognize that WHO never intended this as a classification system,
never intended it to be a diagnosis terminology, et cetera. And we already know
that, for example, HIT Standards committees anointed SNOMED as the clinical
terminology for diagnosis on the problem list.
Is the department considering the possibility of using a clinical
terminology for the diagnoses so that we can live out the richness of what the
electronic health records systems and its derivatives can bring to us about
And at the same time, the Department create the national mapping between
SNOMED and AC-10 for use in claims in epidemiologic research and studies. Is
that in the realm of possibility, given a delay? Is that being considered?
MS. DOO: Well it isn’t being considered yet, because we haven’t delved into
the details. But to be honest, the focus really is on do we do this, and what
does it look like, and what else — so let’s say it’s a year to 2014 — what
else is that on top of. What is the impact to industry, and how do we ensure —
and I have used this term, this simulation — how do we make sure we can do it.
So what can we do from that perspective? So that kind of suggestion would be
totally appropriate to come in as a comment.
DR. TANG: In a sense you talked about costs, you talked about can it be
done. It will be done because we have to, to catch up with the rest of the
world. But what could be the benefit side? I think you have heard a lot about
the cost side. Can we add the benefits, so that for something we have to do and
the pain we have to go through, can’t we get some additional benefit out of it?
That is the proposal.
DR. CARR: Right, and I think that would be appropriate to send.
DR. TANG: It sounds like something this committee might weigh in on, in
terms of providing —
DR. CARR: We have more discussion this afternoon on ICD-10, and so in the
interest of time I want folks to start thinking about this, and also some
guidance, actually, Lorraine, from you. I know you said that there will be an
update, an invitation for comment, and what that timing is. If that is coming
up very soon then that would be the appropriate time for us to comment. But we
should begin today, crafting what we are going to say. Is that right?
DR. DOO: Yes, it is imminent.
DR. CARR: Mike had a question, and then we will stop for a break, but
Lorraine and Judy are here if you want to speak with them directly. Mike?
DR. FITZMAURICE: Lorraine, this is really quite an accomplishment. I wasn’t
able to be there, since I was here. To put all the SDOs together and get their
cooperation and to get their agreement, I think CMS is really to be commended
Particularly, to get them to agree to start sharing their subject matter
I know that the SDOs have separate constituencies, separate subject matter
experts, although sometimes they travel across to SDOs. And they have separate
revenue sources, and that they need to cover their costs to stay alive.
So the question I have is, who will own the intellectual property that
results from their collaboration? Who will be able to charge users for its use?
Will it be the SDOs? The operating rule authorities? CMS? The public? Or some
MS. DOO: Right now I think the way the products work is that for the
standards organizations, so for the standards, there is a fee. And I think it
is a reduced fee if you are a member of the SDO. And NCPDP charges — HL7, I
don’t recall them charging.
When we first did the attachment specifications, I don’t think there was a
charge. And the operating rules are actually free, so those are available for
free on the website. I don’t know if they are planning to change that or not,
for as a revenue source, and to be consistent.
So each authoring entity, whether it is the standards group or the operating
rule or the code set, owns that product. And then how they are making it
available is based on what the regulations are for their organization.
DR. CARR: With that we will take a break. As has become our habit, we have a
lot of intense thinking to do, coming up with standards. So I would ask you, if
you would come back at 11:15 and be prepared to work hard until lunch. Thank
MR. J. SCANLON: Just a quick announcement. Marjorie and I were talking about
this. Several of you, your terms will be expiring in June, I think, and some of
you, we can’t reappoint you. You have had two terms already. So for those that
can be, I think we will probably, when we just reappoint, I think we will
reappoint for six months.
MS. GREENBERG: I think I have communicated with some of you about this. We
are only allowed to extend anybody’s term for a maximum of six months. It used
to be we could extend you until you were replaced, but they are always changing
the rules. So we can only extend you for six months.
We would like for anyone whose term is expiring, even if you have already
been on for two terms, we would like to extend everyone for that maximum six
months. You may not serve that long because you may be replaced. Jim and I have
been looking at the holes that will be left by those whose terms are expiring,
and generally we have some ability to recommend for a second term people who
are in a leadership position. So we are looking at that as well.
As you know, actually our chair has served two full terms. So this is major
here. What we would like to get — I don’t think we are going to do a Federal
Register notice, are we?
MR. J. SCANLON: No. Within HHS we will circulate a request for nominations,
based on the expertise, at least from within. We were not planning on a Federal
Register notice because we didn’t think we had enough to — you would get
everyone all stirred up, and then there are not that many appointments. But we
can reserve that as an option, I think.
MS. GREENBERG: If you have served one term, and your term is ending, and you
would like to be considered for a second term, please let us know. I think we
may have spoken to a few people already who had some interaction. But please
let us know. If you do not want us to extend you for the six months, because
you thought, yes, I thought I was going off in June and I have got other things
I have to do. Let us know that immediately, because we will be doing that very
And if you have suggestions of other people in your field, or that you feel
would fill some of the needs here of the committee, please provide those names
and resumes, if you have some, to provide them to Jim and me as well. This will
be a high priority for us between now, I think, and the June meeting, right?
MR. J. SCANLON: Yes.
DR. CARR: With that, I will turn it over to Walter and Judy. Take us through
the standards letters.
DR. SUAREZ: Thank you. Let me start by making a very brief comment about
what we talked about at the meeting on Tuesday. I wanted to also acknowledge
Lorraine’s leadership and her new dynamic governance approach to meetings,
because I think that was very, very helpful and very dynamic, and very good
actually, to have this deliberate approach of making sure that everybody’s
thoughts and points were raised and brought up and included in the discussion.
The other thing I want to make sure, for those that are on the line also, on
the phone, when Lorraine was explaining and describing the terms of the
agreement that the group came up with, maybe some of the words didn’t come
exactly the way we finalized it in the agreement.
So I just want to make sure that everybody on the phone or anybody that is
in the room understands that the terms of the agreement are noted in the notes
of the meeting, and that they will be used as it was defined in the agreement
on Tuesday. So if people heard something different, or perceived the message
that Lorraine was trying to provide to the committee out of her memory and her
notes, if they perceived it in a different way, it isn’t.
We are going to have a more detailed discussion about that this afternoon
during the Standards Subcommittee meeting, and make sure that the right wording
is described and presented. The ultimate point, I think, that Lorraine was
trying to make sure people understood was that there is a lot of interest and a
lot of commitment to collaboration and coordination and active engagement
across the various groups that have participated.
So that was an important element, I think, across the board, that certainly
there is going to be a need, as we will discuss later this afternoon in the
subcommittee, of a process to identify this authoring entity and the
coordinating, authoring entity responsible and ultimately accountable for the
development of the operating rules. But we will talk more about that later.
With that note, we have three letters that we wanted to present to the full
committee. These all come out of our November hearing on three different
topics. The first one is claim attachments, and the second one is the
provisions on the Section 10109 of the Affordable Care Act.
And the third one is regarding the maintenance process and the development
and maintenance of these standards and operating rules, which is a lot of what
we discussed earlier this week and will be talking about in the subcommittee.
So the first letter we want to talk about is claim attachments. At the
outset of this, an important point is that this letter does not contain any
formal recommendations. The purpose of having this hearing on claim attachments
in November of last year was to understand the status of the current practices
of claim attachments, and the priority claim attachment areas — what are the
most common attachments that organizations are exchanging today, what are the
methods that are being used today, and what is the status of development of the
So it was less clear where things are, and not really the formal hearing
that we are normally holding, to identify and then discuss the recommendation
for the actual standards. So this first letter about claim attachments is a
summary, really, of the observations and thoughts that were discussed during
the hearing about claim attachments.
We will be having a second hearing on claim attachments, probably early next
year. I don’t expect that it will be at the end of this year, but early next
year, to now bring back the industry and stakeholders, to provide testimony on
the standard that should be adopted for claim attachments, and the priority
areas and all those things that we discussed. So that is the background about
DR. CARR: Excuse me Walter, could you say in three sentences what is a claim
DR. SUAREZ: In three sentences what is a claim attachment? Well I can maybe
say it in less than that. A claim attachment is basically a supplemental
medical documentation that payers ask for, to appropriately adjudicate a health
DR. CARR: What would be an example of that?
DR. SUAREZ: What would be an example is an ambulance claim. A provider sends
a claim for an ambulance run. The payer says I need more documentation about
this before I can actually process the claim. That additional documentation is
what is known as a claim attachment. The additional documentation would say —
because in the standard claim, the 835, which many of you recall, the number
associated with the standard electronic health care claim is 835, for medical
The 835 tells a lot of the things that happen, but it doesn’t fail for
ambulance claim, the details about the actual transport and the mechanisms of
the origin, point of origin. Note, a lot of the things that are not normally in
a claim. And so that is what payers need, that additional supplemental medical
documentation, to process a claim.
DR. CARR: And the question we are being asked to address is?
DR. SUAREZ: The question we are being asked to address actually is something
that is part of the original HIPAA law in 1996. The law said that we have to
define standards for the electronic submission of claim attachments. We
identified standards for electronic submission of claims, of claim payments,
the other eight transactions. We never got to the claim attachments.
We started that process in early 2000 and a proposed rule was published in
2005. Some of you might remember, an NPRM was published in 2005 recommending a
set of standards for the electronic submission of claim attachments, and a
priority set of attachments for which those standards would apply — so five
That never got finalized, primarily because of the change in the industry in
terms of the evolution of electronic health records, and the coding and the
vocabulary standards, and other things that were being done. And really, the
evolution of the process of the attachments itself.
And so, in light of the Affordable Care Act, the Affordable Care Act has a
specific provision that says, a secretary now shall publish the final
regulations on adopting standards for claim attachments by January of 2014. So
that is the impetus. That is the reason why we are called to do this. The
secretary has to write the final regulations on 2014, defining the adopted
standard for claim attachments.
We now have to, as an advisory body, evaluate, recommend the standard that
should be adopted for claim attachments. That is what called for this. Now in
order to do that we thought, well, we don’t want to do it now and have a
hearing to identify the standards and recommend them, because they are still
So what we wanted to do in November was have a first hearing, again, to
understand the status of the industry working claim attachments, the current
practices and priority areas for attachments, and the current status of the
standards themselves. And then again, later, in early next year, we will have a
second hearing, at which, again, we will hear from the industry what are the
So that is the main concept behind this. There is a lot of new interest, of
course, on claim attachments, because they are medical documentation, clinical
information, and in light of the work and the new evolution of adoption of
electronic health records and standards for the electronic health records, and
standards for documenting medical information and exchanging medical
Then clearly a claim attachment is bringing now together these two worlds
that have been always separate, the administrative and the clinical, where the
EHR in a medical system, and the billing and financial system. Because now you
are generating a description of the medical information and sending that
electronically to a payer — not to another provider, to a payer.
But ultimately the same kind of medical documentation that is being sent
between providers for referrals, and for coordination of care, and transitions
of care, is the kind of same standard that is now being looked to, to be used,
to submit a claim, which is again, the definition is really supplemental
medical documentation to a claim.
So one of the main highlights that we heard in the hearing that we
documented was exactly that, the concept that a claim attachment is not any
more than administrative, simple transaction that was thought of in the early
days, and that it is really much more tied now to the electronic health record,
to the standard being used by providers in electronic health records and those
kinds of things.
So throughout the letter I think we try to synthesize, really, the findings.
There are many, many things being heard. It was very rich, very valuable,
thoughts or perspective, the payer’s perspective, and within the payers,
Medicare, Medicaid, private payers. We heard from providers, from hospital
systems, from long term care, pharmacy, others. And then the vendors and claim
So at the end, the main message basically is four or five points. Number
one, currently, claim attachments are still significantly being done on paper.
That is the first kind of outcome. There is a lot of claim attachment being
done today. Some estimates are between five and up to 20 percent of all claims
submitted require some sort of a claim attachment after the claim is sent.
There is different types of purposes for which a claim attachment is needed.
There are lots of priority areas, and a number of them were listed in the
findings. The most common ones are things that involve ambulance services, for
example, medical necessity documentation. So documentation of support of
certain medical services, or medical equipment devices, authorization requests
for referrals require quite a bit of those.
And then there were a number of other areas where claims attachments are
being requested, more administrative, like the third party liability,
determining third party liability when there is other payers involved. Things
like privacy documentation, certification of a patient, consumer privacy or
consent, I suppose, consent to be treated for certain procedures or those kinds
So there are a number of areas, and we identify them, we highlight them in
the notes of this letter. And so in summary, and I guess just to point out, and
page four, I guess, is where we summarize some of the main findings, there is a
strong support from the industry for the identification and adoption of an
electronic standard for claim attachments.
While there is interest in moving to electronic attachments, there is also
interest in reducing the number and the type of attachments that are needed.
Part of what we all discussed during the hearing that we thought was going to
be provided by having more refined codes to document conditions and procedures
performed on patients, like the new ICD-10, was an opportunity to see the
reduction of the need of attachments.
But now the claim would have much more detail and richer information about
what happened to the patient. So there is still interest in making sure that
while we want the electronic standard, we can find a way to minimize the number
and the amount of claim attachments, rather than going the other way around and
increasing the need and the number of attachments.
There were consensus and consistent support for the adoption of the
standards that are being developed by HL7 for the content of the message.
Remember, this is a medical documentation message that is very much the type of
message that HL7 is used to. And it is very, again, consistent with the
standards that have been adopted under meaningful use for the exchange between
EHRs of medical information.
And there were discussions, of course, in addition to the HL7 standard, as
we call it the wrapper, the envelope, to be used, then being developed by the
other standard development organization, X12, the ASC X12, that has developed a
standard to provide the envelope for communicating and routing through
electronic data interchange networks, an electronic health care claim
There were some areas that people thought would be helpful to identify as
priorities. And those are the areas that the standards development organization
is already working on in terms of what are the most common types of attachments
that are being done. We heard survey results from assessments in the industry
about which are the current standards, or the current priority transactions,
that require claim attachments.
There was also a strong recommendation to ensure that the standard allowed
for something that is known as the unsolicited claim attachment. In many cases,
providers know that every time they send this type of claim — and this is a
good example — every time they send that type of claim, every payer is going
to ask for an attachment.
So instead of waiting for the claim to go, the payer to send back a request,
and then the provider documenting the supplemental documentation, putting it in
a message and sending it in a separate time, and then having the risk of not
matching the right claim with the right attachment.
And those kind of issues, having the ability to send a claim, and along with
the claim, send an unsolicited attachment. That is why it is called
unsolicited, because we didn’t wait until the payers solicited the attachment.
We just sent it with the claim.
There was a strong support for having that be part of the standard. And
there was also strong support for making sure that not everything in the claim
attachment world can be coded, coded to the granular level.
There is going to be some instances where the attachment is going to be a
print image of something, a print image of a signed form by a patient. And that
is not at this point able to be coded. And so the two versions of a claim,
which are the coded, or what we used to call the computer variance, of a claim
And that means basically that the claim attachment will come all coded, and
the computer that receives the claim attachment will be able to open the
envelope and allocate all the coded information, and even take adjudication
actions out of that.
And then there is the non-computer version of it, what we call the human
variance. And this was a term that we used back in the 2005 regulations. There
was support to also consider that as a mechanism, because certain things will
have to be sent in the right envelope, and using the right standards, but in a
not coded structure document way, but in an unstructured document format.
So those were some of the main highlights. Again, it was a very rich and
very helpful hearing. And with this letter, what we wanted to do was document
the hearing, establish some of the original parameters that we heard during the
hearing, so that when we have the second hearing and we prepare the
recommendations, we build from this first hearing and some of the observations.
So I am going to stop there and see if there are any comments from people,
or any questions. I know we did receive a couple of comments from Executive
Committee members after we submitted this letter to the full committee. So we
will be taking and incorporating those.
Many of them were primarily wording edits. So we will be taking those during
this afternoon’s subcommittee and editing the letter and adding that.
MS. MILAM: I had already submitted a comment that you are referring to
around the minimum necessary requirement for the payment transactions, which I
think is fairly important. But I have a question.
I am wondering whether the privacy notices that are required by some laws —
I am thinking of federal, part II substance abuse, and some state HIV law —
that are required to travel with the PHI, wherever it goes, would that be part
of a claims attachment? And then if the answer is yes, do the standards around
claims attachments somehow impact the content of that privacy notice?
MS. DOO: That is a good question. The way we originally defined attachments,
and the way the X12 and HL7 is, it is information that is necessary to
adjudicate the claim. So that term of adjudicate the claim might be the
differentiator. So if they don’t need those documents to process it and pay it,
then it probably doesn’t travel with it and it probably wouldn’t be part of it.
Not for the purpose of the claim, but we should put that on a list of
questions to look at when we are really exploring what those attachments will
be, because they are in that process now.
DR. SUAREZ: The minimum necessary component of HIPAA will apply here. And
minimum necessary to adjudicate a claim is going to be a key term to ensure
that whatever data is being requested by payers is the minimum amount needed.
MS. MILAM: I guess I see the notice issue as something separate from the
minimum necessary issue, because the notice is required by specific state or
federal law with regard to the category of sensitive information. It is a
notice that is required to be put on the PHI that alerts the recipient that
they are receiving a certain category of PHI, and what they are allowed to do
in terms of their use and disclosure.
So it is required by law to travel with the PHI, although the plan may have
no use for it.
DR. SUAREZ: I think that would be necessarily a part of the element that
would need to be considered to go with the attachment. Clearly, if that
particular piece of information that is requested by the payer carries along
what probably could be seen as metadata, the concept of metadata, then yes, the
attachment should be able to support the ability to comply with those federal
and state laws. I expect that then it will be able to be carried into the
DR. W. SCANLON: I want to just say, I think what Walter gave us is an
incredible introduction to a topic that in my mind is maybe one of the most
important things that the Committee deals with over the next couple of years.
And it goes beyond the Standards Committee. It is the entire committee. And
Sally’s question is a personification of that, which is that there are privacy
issues, there are quality issues in terms of the data flows that we can
potentially have in the future.
And as Walter pointed out, it is a new world. It is not 1996. It is a world
in which we are going to have electronic health records, there is going to be
this information. Administrative simplification, the goal was to try and become
more efficient in the processing of claims. But we now are in a world where are
talking about pay per performance, quality measurement, et cetera.
And this is maybe a vehicle that is going to help us a lot in all these
different dimensions. The question is, how do we structure it in a way that
will work? I am thinking that it is very important that you all learn from what
Walter talked about, as well as this letter. It is not just a letter to the
secretary in some ways. It is informing the committee about this really
important topic for our future.
DR. TANG: What if we step back and make sure we know what the problem is to
solve, not the problem is that was to solve in 1980? Because I think the whole
notion of claims attachment is a paper-based concept. So I have a bill. I can’t
tell whether I should pay this so give me the medical record in a sense was
what the claims attachment was about.
Now to fast forward a little bit, the new world actually isn’t about fee for
service, it is — let’s think of the ACO. You actually don’t need that any
more. It is your job to do the right job, prove that you do the right job and
then get compensated for taking care of a patient.
In that sense, claims attachment per se is sort of an anachronism. So is
there a way to reconfigure the whole notion which is, should I pay this amount.
And you could prospectively, because all the plans, actually, they run the
computer systems that spit out the requests for the claims attachments, they
know prospectively what I need to know in order to pay this amount.
In the old world it is just, should I pay this lab test result. And we could
actually adjudicate that literally online using the electronic health record
system, which is very different from saying give me a whole lot more
information to digest and to spread the privacy risks around.
Alternatively, in the new world, which is still only a matter of a handful
of years away, this whole ACO notion, the pay for managing of population, it
then really makes claims attachments sort of an old concept. So how can we
There are two steps. One is, can’t we just prospectively, everybody know
what is appropriate and then literally just send that with it? Or have an
auditable way of detecting whether that was an appropriate charge? And then the
other leapfrog is, and do we really need claims attachment versus — look,
Kaiser doesn’t have claims attachment, for example. Is that even possible?
DR. SUAREZ: It is the old way of thinking, if you will, the concept of
claims, and claim attachments and all that. I totally agree with that. I think
the perspective is probably going to be more, there is still going to be a
payment to be made from a payer to an entity, a provider, and lets you the
The type of information that the payer is going to be wanting to see is
going to be different than a traditional claim, but it is still going to be
probably an encounter based medical documentation support, or some sort of a —
well, there are going to be some ideas of an aggregate and maybe even detailed
quality measures, outcomes, and all that, that is going to serve to that.
But that is going to be part of the discussion of how to we establish a
standard that can allow the transmission of medical information that is needed
to adjudicate a payment, however that payment is made. So I think they have
opportunities to look wider and think wider, is still within the context of
some of the things we have to do today, but the things that we need to do in
DR. CARR: Judy is next, but I just want to underscore both Bill and Paul’s
comments. It is thinking about where we will be, and the relevance of the
information needed. So I think this is exactly the role of the committee.
DR. WARREN: So I would like to frame the discussion in a slightly different
way, being the very pragmatic person that I am. Currently, we have fee for
service. This claims attachment standard is to support that process. And so
that is what the standard is. It is a very precise piece of information that
insurance companies need to have, in order to pay the provider.
And so let’s look at it from that perspective. Now that then travels within
the larger health information exchange framework, which has all the things that
Sally is talking about, of attending to privacy and everything else. If you
start looking at it as an onion, it starts getting bigger.
And then I do agree with Paul that this is a conversation — and also with
Bill — that this committee needs to have, is this larger one of how do we
transition from the current fee for service standards to ACOs. And ACOs aren’t
going to be out there for awhile, because we still just have pilots. And then
wherever health care reform is going to take us after that.
I think the conceptual dialog of how we go from where the electrons meet the
road to ship out the information in forms of ones and zeros — because even if
it is XML encoded, it is still a one and a zero that has to be sent — all the
way up to where does our health care information need to happen.
I think sometimes these kinds of letters stimulate the conversation, which
was certainly the purpose that Walter and I had when we put the letter
together, was to begin the dialog. I would like to say we still have the
standard to approve, we have some time. If there are key questions, like Sally
has given us key questions to make sure that we hear about in the hearings.
And then I think for planning for the Executive Subcommittee, we need to
have some quality time like two or three hours set aside to talk about some of
these issues, because they will inform all of us, and they will help bring in
the quality aspect, the population aspect, et cetera.
Because there is data in the claims attachment that could be very well used
for population health as well. All of the things that Jim talked about this
DR. CARR: So just to keep us on track, I have next Mike and then Bill. And
then maybe what I would like to do is bring it back to Walter, to say as we
look at this closing comment in this letter, do we want to foreshadow the
broader context in our closing comments about what we will be deliberating on
now and in the hearings ahead. So think about that, and next is Mike.
DR. FITZMAURICE: I think the letter is great. It is very explanatory. It is
as clear as you can get for something as complex as claims attachments. I raise
the issue that Section 1104 of ACA directs its Secretary to publish final regs
no later than January 1st, 2014.
We plan, looking at the concluding comments, to hold a second hearing in the
early part of 2013. And I assume that we would then put out a letter in June to
the Secretary. I am wondering if that is sufficient time, or do we need to move
that up earlier?
DR. SUAREZ: Didn’t we say the second part of 2013?
DR. CARR: Yes, you said you were going to have a hearing in 2013, not 2012.
DR. SUAREZ: Early 2013?
DR. FITZMAURICE: Yes, I said that. So if we get our recommendations out in
June, that’s not a whole lot of time before a regulation to be published in
DR. SUAREZ: I think they gave us, to try to hold a hearing, like in January
or February of the latest of next year, and then turn quickly around the letter
with recommendations to CMS, so that they can incorporate those into the
MS. DOO: That is right. That make sense. The trick will be for the industry,
for the standards organization and whoever does operating rules, to have
something to be able to share with you in that timeframe. And it is February,
or March, actually. So there is very little time.
DR. SUAREZ: That is what we have discussed with the standards of
organizations of the timeline, that the latest that we can hear and receive as
a deliverable the recommended standard is early 2013. And so that allows for
the balloting inside the SDO and all that.
DR. FITZMAURICE: So we are working fast, but Lorraine, if we get your
recommendations by June of 2013 —
DR. SUAREZ: No, it would be by March or April of 2013.
MS. DOO: We are going to try to do the letters faster.
DR. FITZMAURICE: Okay, but that would be sufficient time for you to work
hard to get the regulation out? So we are in synch?
MS. DOO: Yes. And there is much we can do on the regulation in preparation.
For example, much of the work on the preamble and impact analysis, we would be
working on. So we are not so dependent. Much of that can be done, and then it
is a matter of did we get the stamp, and what are the parameters.
DR. W. SCANLON: Just quickly, I think both Walter and Judy raises — Paul, I
am thinking that we may reach the point where we can use smart phones with
secure communications before we get to the point where we can ignore some of
this encounter information.
And I think even if we get to that point where between the payer and an ACO
there is a capitated payment, there is going to be the issue of the ACO and its
management. Some capitated organizations already now are doing claims
processing in the background, because they have to measure productivity and
It is just that we should be thinking where you want to be, but we also are
going to probably need a very robust interim strategy because it is going to
take us enough time to get there. It is kind of like, we are now trying to
finish the 1996 assignment for HIPAA. Let’s all worry about the current problem
DR. TANG: What about the prospectively determined algorithms for determining
the eligibility of a claim, the suitability of a claim, so that we don’t have
massive amount of information going, just because it is a claim attached? I am
trying to avoid that piece.
DR. W. SCANLON: I agree with you on that. My sense is that there should be a
purpose for the information that is transmitted. And part of it can be the
adjudication of the claim, which is clearly consistent with the law. But again,
the law was passed before we started to think about pay per performance quality
measurement, et cetera.
We may need to think about revisiting the statute and say we now have these
other needs which we have very clear responsibility for. Let’s get information
to do them well.
DR. CARR: So with that, I would like to follow on that we need to allocate
time for the full committee to really brainstorm about all of the implications
of claims attachments, the immediate need as well as the future uses. So let’s
make a note of that for another committee meeting. But then from there, with
this letter, will we be voting tomorrow?
DR. SUAREZ: Yes, we will be.
DR. CARR: So with that backdrop, and thank you for that clear summary, take
a look at the letter. If there is anything that we want to bring up tomorrow,
let Walter know, actually, sooner rather than later. Are we ready to go on to
DR. SUAREZ: I think we are ready, unless there is any —
DR. WARREN: Just one thing is, we are going to be making corrections to
these letters this afternoon. So the Subcommittee on Standards, and the
Executive Subcommittee, have seen these and given us feedback. We still have
Sallie’s to incorporate. If anybody else who has just seen this for the first
time has a big kneejerk, let us know before we meet this afternoon.
DR. SUAREZ: Thank you. So moving along, the second letter is a section in
10109 of the Affordable Care Act. Just to point to the two things that Justine
always wants us to make sure we cover, what is the ask, and what do we have to
do. Are those the two?
DR. CARR: That will do. What is the problem we are trying to address?
DR. SUAREZ: Why don’t we process what is the problem we are trying to
address, and then what we have to do. So this section in 10109 basically was a
supplementary section in the Affordable Care Act to the administrative
simplification provisions in Section 1104. That was the main section that
talked about operating rules and the new standard for health plan ID and the
claim attachment, and all that.
10109 said that the Secretary should consider other areas of administrative
processes and other areas in the financial and administrative sector of the
health care industry, to see if there were ways in which we can improve the
efficiency of those processes by adopting standards. And so the Secretary was
asked to look at that, and to consider input from NCVHS from the HIT Standards
and Policy Committee, and from the industry.
And to try to do that, actually, by January of 2012, basically receive the
input and provide some perspectives on this, or get the input by January of
2012. So consistent with that, NCVHS convened this hearing on Section 10109,
and the ask was basically, are there any ways in which we can improve the
efficiency of processes in five areas, basically. And that is what we cover in
the letter. I will talk about those five areas in a second.
We did got to the HIT Policy Committee and the HIT Standards Committee to
introduce this, and to ensure that there will be an opportunity to hear input
from them, and we offered to serve as the facilitator of that process rather
than having separate things go on from different committees.
We served as a facilitator of gathering this information. That is what we
did in the hearing in November, where we heard from the industry about four
processes. We didn’t cover the five areas, we covered four areas. The four
areas that we covered are at the bottom of the first page.
Are there ways to improve the efficiency of provider enrollment? Are there
ways to improve the efficiency by incorporating or requiring somehow other
aspects of the insurance sector — not just health but property and casualty,
which includes Workers Comp, the auto insurance and these other areas,
including them under HIPAA? Is there any way to do that?
The third one was, are there any ways to improve the efficiency of the audit
that we all face, whether it is providers being audited by Medicare, by payers,
whether it is payers being audited to Medicare, all these audits that happen in
the industry? And was there any way to improve the efficiency if we were to
adopt standards or some other things for the edits that are done by payers when
they receive a claim?
It is called a claim edit. As soon as the payer receives a claim, they have
procedures, electronic procedures and even sometimes manual procedures, to edit
that claim before it goes through adjudication. And ensure that we are not
doing C-sections on men, for example.
So those kind of edits are front edits, and there are other edits that are
done, and so explore the possibility of exploring those processes. Those were
the four areas that we covered in the hearing.
The fifth area was timeliness of payment rules. So consider whether health
plans should be required to publish their timeliness or payment rules. In other
words, having health plans published, basically, when are they and how quickly
and how timely they are paying claims. We did not cover that. We say that we
will be covering it later this year.
So we divided the hearing into these four sections. We invited industry
perspectives, and we, again, had an incredibly rich hearing. And the
overarching observation, which is on page two, was that clearly there are, and
we heard that there are, important opportunities and meaningful ways to
increase the efficiency and the simplification of these various processes,
through the standardization of those processes, establishing some standards,
even electronic standards.
But that these were very complex areas, that each of those four areas, even
the fifth one, they are very complex. They cannot really be addressed in a
matter of a few months, and coming up with some recommendations.
There needs to be a lot more work done into the future, and that is what we
are commenting, and we are pointing in this letter, that really this is just a
starting point to get a sense of what are some of the opportunities.
Really, when you are about to make a recommendation of this, which should be
the standard process for doing this, and this should be the actual electronic
standard that should be used, it is a very different level of recommendation,
and at NCVHS we didn’t feel that we were ready to make those types of
So the overarching recommendations on page two are, number one, a strategy
to establish for farther exploration in each of these four areas, in order to
develop recommendations for comprehensive improvement.
The strategy may include holding additional NCVHS hearing, and convening
multi stakeholder workgroups, whether under the auspices of NCVHS, as we are
about to launch a workgroup for other purposes, or whether someone else, and
then bringing back recommendations to NCVHS and to the Secretary.
While we are saying in the recommendations that NCVHS plans to recommend
such a strategy — so between now and June we are going to develop that
strategy and make a recommendation of that strategy by June of this year. We
point out that NCVHS is certainly prepared to facilitate the recommended
strategy as resources permit.
Because again, if we launch four different workgroups, plus the fifth one,
and this new one that we talked about earlier, there is going to be a resource
issue. We don’t have the funding the ONC has to create workgroups and power
teams and tiger teams, and spring school and summer comp and things like that.
We do not have the resources in NCVHS, and this will take a lot of
resources, each of these four areas. So those were the two recommendations. We
provide a brief description of the findings. I just want to point out just one
example, and this is something that had been mentioned. We had extensive
discussions within the subcommittee.
I do want to acknowledge Bill’s persistence in his points, and I really
appreciate — I think we all appreciate — your perspectives on how complex and
how important this is to be done in the right way.
We weren’t really before making recommendations. The initial idea was we
should create these workgroups. But I think we need to develop this strategy
and mature this strategy a little more, before it can be formally recommended.
I think that was Bill’s point. But we wanted to provide a description of the
findings of each of these four areas.
The one I wanted to mention is provider enrollment, just to give you a quick
example of that. When we started to think about provider enrollment, what is
provider enrollment, well, providers enroll in health plans, to be part of
their health plan network and be able to care for the enrollees of that health
plan. That is what is known normally as a provider enrollment process.
And there are standards to do that, even standard forms. A lot of that is
done by paper. You will have heard of the 855 form, I think it is, Medicare’s
enrollment form into the Payco system, which is the system that maintains
information about all the providers that are enrolled to be providers in
Every health plan has the exact same thing, but it has different forms. It
has different data that is collected. And in reality, when we started looking
at it, 80 percent of the data is the same, in some cases more.
And there are a few extra data elements that are asked. And the question
was, can we standardize that process, is there ways and tools and other things
that can help standardize and simplify the process. And we heard during the
hearings yesterday, there are certainly ways to find improvements in this.
Enrollment is not just enrolling myself as a provider into a health plan. I
have five or six other types of enrollment. I can enroll as a provider into the
electronic fund transfer exchange with a provider. I have to enroll for other
things. There are different things that I can enroll as a provider.
So the provider enrollment became a much larger, again, concept, the scope
itself. And the opportunity is even greater. So that is what we thought it
would be very good to, again, launch, and for a deliberate process to try to
understand better each of these issues before any recommendations can be made.
I am going to stop there and see if there are any reactions. Bill, do you
want to highlight a few of your points?
DR. W. SCANLON: First of all I have to say it is very nice to characterize
what I was doing as persistence as opposed to just stubbornness. And my
principle issue was that being on the Standards Subcommittee is like drinking
from a fire hose.
And I think Walter and Judy and Lorraine are the only people that are
capable of doing it. They manage to deal with the volume that comes before that
committee in an incredibly effective way. And the outcome is commensurate with
their skill and their knowledge.
So being not of that category, I am looking at — we had 50 people, I think
roughly 50 people testify in November. I am looking into, after recovering from
that day and a half, looking at these testimonies, and the transcript at times,
and trying to sort this out and coming up with more questions than answers.
And my concern was that we really do need to think about how do we move
forward on all these fronts in a responsible way. So this notion that we have
now, that we really need to have a strategy that takes into account the
sources, and probably ultimately starts to think about priorities — I hate to
say this on a transcript, but every date that Congress has ever come up with
isn’t necessarily one that is feasible.
And so, we need to consider what is the most important thing for the
resources that are available to focus on, so that that job is done right. And
things of lesser priority, it may not be so critical if it is six months later,
12 months later, that those are actually addressed.
I think that was the source of my persistence, and I think we are in a good
place right now. We are recognizing the complexity of all these issues, and
again, bringing them before the full committee so that you can be more informed
participants as we move forward.
DR. CARR: Thank you very much. I think that it is certainly true that these
standards issues are very technical, very challenging. But every member of this
committee is accountable. When we send a letter, it represents our collective
And so there is no question too fundamental or too simplistic or off the
reservation, because those are often exactly — I am not talking about you,
Bill — we are asked to use the collective wisdom of this committee.
The committee is charged with standards, but most of the people on the
committee don’t deal with standards, but they deal with process and they deal
with the consequences of these standards. And therefore, it is our
responsibility to diligently learn, ask questions, and deliberate.
So just to hearken back, in terms of creating a time not just to understand
claims but really, as you say, for us as a committee and with Jim’s input as
well, to look at our resources, the demands, and the priorities, to ensure that
we meet the expectations, within reason.
DR. FRANCIS: This is a silly question, maybe. So one of the things that I
have never had a good purchase on is how much of this is like redecorating a
room? The analogy that I want to bring to mind is that, when you redecorate a
room you don’t put the carpet down first, before you paint the walls.
So I don’t have a sense of how much in any of these things really need to
happen before other things. That is part of where I have been really confused
about the whole question of what do we delay and when. And that goes back to
DR. WARREN: So, Justine, can I respond to Leslie? I think she has got a
really terrific metaphor going. So standards are to decorating a room, knowing
where the light switch will go, and knowing where all of the electric socket
places will go and knowing where the window goes. So before you could even get
to using the room and putting in carpet and thinking about carpet, you have got
to have the foundational pieces that make that room worthy of being used. And
that is where a standard is.
MR. SOONTHORNSIMA: I think another driver for this recommendation is the
understanding of what the industry is going through right now. There is a
tremendous amount of foundational things that were being laid back in 1995, and
we have yet to complete implementation of those things, in addition to 5010,
So that is why we followed the wisdom of Bill, stepped back, re-look at the
big picture, coupled with all of the tremendous changes that the industry has
to go through and absorb. So we recognize that these are important issues, but
we really need to take some more time and strategize further, and not rush into
any sort of recommendation.
DR. TANG: Let me pick up on Judy’s metaphor. The trouble is, if we redesign
this room, and we already know that the highway is going to be built and we are
going to raze this house anyway, why should we do this? And one of the things
that Justine always says is, one of the things that NCVHS has had a track
record of is to sort of predict and talk about the vision, predict the future,
and try to offer some guidance.
This may be the wrong time to be redecorating the room that is going to be
razed. And maybe we should be talking about what is going to be required in the
new world. And the pace of the world is much faster than it was the last 10
years. I think we are going to get to where it’s not just ACO a la CMS. The
private sector has already got ACOs, this year.
So it is just something that is happening. And if we have the standards
chased to last years, we actually won’t have enough time to do — right now we
have a little lead time to do standards for three years from now. If we use the
three years to chase last year’s work, it is just not going to work. And it is
not what we have been able to contribute. So that is an example.
It is part of this whole claims attachment, but it is the how can we build
the systems and the data requirements. In some sense we almost should keep up
with meaningful use. We have set out these three stages, and we are already
working on stage three. And stage two has signals built in for stage three.
In some sense I think that is the kind of work we ought to be doing. We
should be writing in our letters the things that are going to happen and we are
going to need, three to five years from now.
You said there are no stupid questions, so hopefully there is no stupid
suggestion. The stupid suggestion is that claims attachment might be the .jpg
of the paper that is being floated around now. Why, because we need time to
specify the standards that are going to be needed in three to five years. And
that is a stupid suggestion, but it just might bypass some work.
Ob was saying all the work — well gosh, do we want to give them the work of
today and tomorrow at the same time? Or should we just skip yesterday and go to
tomorrow? End of stupid suggestion.
DR. SUAREZ: I think it is a fundamental element where we are historically
right now standing. And just to take the analogy farther, first of all, I don’t
think this is a room. This is more like an aircraft carrier. An aircraft
carrier, the good thing is they don’t have carpet so we can decorate them
differently. But still, the important thing is really, I think, this is like,
yes, we have a building, and the building is placed in this place, and there is
a highway right now going near that building.
And we are shifting and trying to adjust that building while the highway for
the future is being built someplace else, and we are not yet thinking about,
well, so how is that building going to look like near this new highway? So that
is the real challenge — we have an industry that is today moving in a
particular way. We have a lot of reform and a lot of new directions where we
are going to go.
And we have to be thinking about how all this is being done in that new
world. And we will have to work to make sure that when the new world exists,
which is, well, probably 2014, if things continue to go the way they are — we
are going to have to have a mechanism to support that new world while the old
world is still moving, and try to begin to move everybody out.
It is really the challenge. We are going to see claims going today and for
the next few years probably. And in the future, the new world is probably going
to say what is a claim again? We are not talking about those terms. So we need
to begin to think about that future absolutely. I totally agree.
DR. WALKER: This is Jim. I think we are talking cost benefit analysis.
People are doing it now. What would be the benefit to the people who are doing
it now in a sub-optimal way for us to pile onto them all the costs of doing it
in a more optimal way? And what is our best guess about how long it would take
to do that more optimal way and get it into production, and how long they would
actually use it once it was in production?
And, if we let them vote, would they say we would rather cripple along the
way we are now than do all of this stuff? And so it seems to me we could
contribute at least a high level cost benefit analysis, and make some estimates
about what the benefits and what the costs are, and then make a recommendation
out of that, that other people could get some use out of.
DR. WARREN: My comment is, in responding to Paul and his recommendations,
that you always have to plan for the future but you have to live today. And so
everybody at their place today had a timeline of ICD-10 and everything that
happened to it.
Note that we made the recommendation and the rule was passed to start with
ICD-10. We started in the nineties. We made incremental approaches in the speed
that industry told us they could handle it. We passed the rule — I believe it
was 2008, 2009 — that it be required in 2013, which is what industry told us
they wanted. And that is all of industry. So everybody was there and agreed to
Unfortunately, Meaningful Use came out during the process of that.
DR. CARR: No adverbs, no adjectives. (laughter)
DR. WARREN: Well it depends on your perspective, doesn’t it? When you were
part of the table, trying to very carefully craft. But I think the lessons that
we learned in ICD-11, we need to re-look at, and apply to what Paul is talking
about. Because we do have these two timeframes that we live in, the one that we
are in today and the one that we have to build towards the future.
And I think we need to look at some of those lessons and decide how better
we can craft these than what we did with ICD-10.
DR. TANG: So if we truly did go to .pdf, I think that actually is a better
world than the world we have now. It gets there quickly, it gets there
reliably, it gets there in a way that you can sort and manage while we paint
the new world that is going to be enormously, not only better but more
But we are going to need time to paint the new world. So I think there is
something we can do immediately.
DR. SUAREZ: The standard for a claim attachment — just to use the same time
that we have been using — the standard for that is actually taking a regular
paper and saying, now there are two ways that you can send it. One, convert it
to a .pdf, so it’s quickly you get it there.
The other one is, for some things where we already know that the electronic
health record is maintaining that data in a structured coded way, you can
create the structured coded message and go beyond one step from the simple
.pdf, and for five areas send the actual coded message.
So we are already kind of building that transition from the current
standard, which is paper, to the .pdf version, and on top of it to, for some
areas, the coded structured version that then can be sent to the payer. And
then you can in the future really see the structured coded way would be the —
DR. CARR: We are going to have Paul and then Lorraine. We will continue
until we are done, and shorten lunch. Just want to let you know that. Paul?
DR. TANG: The only tweak I would make on what you just suggested would be,
yes, we want to get the coded data. But the leapfrog would be, let’s do only
the coded data necessary to answer this question right at that time. That is
And if we work on that, specifying how do you tell me what you need with the
charge, and I send it right there, they will adjudicate it instantly in a
sense. Do you see what I am saying? But that is not being asked for. Right now
we are asking for, how do we codify paper?
And yes, we were answering that question, but the question was asked in
1996. Let’s just update it by a decade. And that is the predictive thing that I
would just —
MS. DOO: I would only say, and this is probably a little bit naïve,
that yes, we were supposed to do, actually claims attachment we were supposed
to do 36 months after the passage of the law. And we got there in 2005, which
was a few more years than 36 months. But I think what we heard in that, and
what we saw in ACA when they reiterated certain things, these are the current
So provider enrollment is a pain point. It has been, and it is. And are the
challenges we can’t be nimble enough, it seems, to make changes ourselves
within industry. We are sort of waiting for a hearing and we are waiting for a
mandate instead of industry saying, you know what, we could solve this problem
together. And none of us really knows how to make that happen.
DR. SUAREZ: The third letter. And this is incredibly building upon the
discussion we just had. The third letter is addressing the point that I think
Lorraine discussed earlier in her report. This wasn’t really something that we
were asked to do, but we ourselves decided this was an important thing to do.
And it was to understand, now that we have standards and implementation
specifications, and operating rules, we needed to understand what is the
maintenance process and what is the right, appropriate mechanism and pathway to
ensure that process of maintaining the standards and updating them and moving
forward, is consistent and is predictable, and is not a known way of
So we started to hold hearings. Actually, this was, I believe, our third
hearing, with SDOs, with Standard Development Organizations, with the authoring
entities of operating rules, and with the rest of the industry — providers,
payers, clearing houses, and others — to really understand how is the current
process working, how is it not working, where do we need to improve it, and how
can we move forward improving it.
And that is what we try to do. So we were going to make a letter into three
sections. The first one was, a summary of the X12 standards. In general we deal
with primarily two types of major standard bodies really, the X12 standards
that define the standards for the claims and the claim payment and enrollment,
eligibility, all these transactions. We also deal with the National Council on
Prescription Drug Programs, NCPDP, for the pharmacy electronic message
standards, the pharmacy standards.
And then we deal with the operating rules, authoring entities for the
operating rules that are being developed. So we divided the letter into those
three groups. What is the maintenance and update process of X12 standards and
implementation specifications, and how are things being done, what are the
processes for operating rules in terms of developing and maintaining those
And then the same thing for the pharmacy standards. And basically just to
jump to the conclusion part, I think, in page four at the bottom, we document
here that we continue to hear that there are issues affecting the current
process, including barriers to participation.
There are still issues about participating in the process itself, lack of
stakeholder representation, many small, rural providers are not yet quite
engaged in the actual process. And many providers, even, across the board, are
not engaged in the process of the standards development. And then increased
complexity of the standards, and the operating rule maintenance process.
So at that point we continued to see some of the challenges between SDOs,
between data content communities, between the operating rule authoring entities
— a lot of it because of confusion and lack of clarification perhaps. And
still, becoming familiar with a new way of thinking about standards.
So basically the idea was, we needed to find a way to begin transforming, if
you will, the way in which we as an industry maintain standards, and have a
much more coordinated approach across all the ACOs, across the authoring
entities of the operating rules, and across data content committees.
So that was kind of the ultimate goal, and I think that was the
recommendations on page five that we made. The first recommendation is
something that actually already happened, and you heard the response.
We were making a recommendation here that the Secretary should encourage the
industry to immediately convene an intensive working group that involved SDOs,
operating rule authoring entities, members of the DSMO, to fully evaluate the
strength and the weaknesses of the current process, with an expressed goal of
submitting specific recommendations back to NCVHS and HHS within six months.
So that work and that group, already started by CMS convening it and taking
the lead on that. And we really appreciate that because I think it was amazing
to see how everybody that participated in that were saying, we really needed
this meeting. We needed someplace to start and it was great to see that CMS
took the lead in facilitating that meeting.
The second recommendation is about working with SDOs and operating rules,
authoring entities, to establish an expedited modification and adoption process
for changes that need to be done in an emergency manner, expedience changes to
the standards and operating rules.
So what we are experiencing is that just in general this year we implemented
5010. A lot of the implementations, it was the very first time 5010 was being
done by those systems. There was no testing done, as we, NCVHS, recommended.
There was no either internal testing, or in many cases joint testing, testing
with trading partners.
And so what we are finding is a lot of reports in these first few months, in
the last two months, basically, since this has been started, of issues and
problems. In some cases serious problems.
So we need to find a mechanism to quickly address and solve those, whether
it is an operating rule change, whether it is a standard change, or whether it
is some other change in terms of the understanding of the use of the standard.
So that is a second recommendation, is to develop that kind of a formal
And the third recommendation requires that before a new standard is adopted,
or a new version of an existing standard, testing is completed, and findings of
that testing is included in the information of the recommendations to NCVHS for
consideration to adopt.
Because what we are finding, again, is that in many cases the standards are
being developed, and quickly put together. Not quickly put together in the
sense of there is a process and the SDO goes through balloting and a lot of
that. But at the end of the day, the actual testing of that standard doesn’t
happen until the actual standard is implemented.
That is what we are experiencing in these last two months. People are
testing how 5010 is being done, and trying to fix and address the issues. And
that should not be the case. The clear requirement should be, before a standard
is recommended, it had been tested and it had demonstrated the benefit, the
value, the improvements.
And then we can say okay, we evaluated the standard. It was developed in an
inclusive, open process, it went through the balloting, it addressed all the
concerns, and then it was tested and here is the demonstration of the test.
So those were the three recommendations at this point. And I know out of the
meeting from Tuesday, there is going to be already a number of things that are
going to be happening over the next few months about this process. I will stop
there and see if there are any questions or comments. Lorraine or Judy, did you
want to say anything?
MS. DOO: I will just add, I did get funding this year to do a pilot test of
the standards, so we could develop a process for doing that. So when they come
forward to you in future, you will know that it has got something that says it
works, and it truly represents industry’s needs.
DR. CARR: As we think about tomorrow, Walter, and all the deliverables
mapping to you, I just want to review and make sure we are on the same page.
Number one is a timeline on that RFP for the coordinating body, and the
decision making coming out of the summit. I think we really need to think work
backwards from due date to how that process will map out.
Second, we will have these letters. Third is, we need a recommendation for a
discussion, as we have talked about, for the issues, about how we prioritize.
And coincident with that, I know on one of our standard subcommittee calls, you
had outlined a huge body of work that maps to standards.
And I think that we need to hear from you, need to have, want to have, nice
to have, kind of the things that we can do because it was extraordinary. And
even as we talk about these other things, it is even more demanding. So when we
come back tomorrow, that will be a lot on you. But I think it will help us
With that we will adjourn our morning session now for lunch. Folks who want
to talk about that workgroup can convene. Thank you everyone. We will come back
on time, as outlined in our calendar, 1:30.
(Whereupon, a luncheon recess was taken at 12:30 p.m.)
A F T E R N O O N S E S S I O N (1:30 p.m.)
DR. CARR: Welcome back from lunch. Now I look forward to hearing from Larry,
Sallie and Vickie. Who is starting?
MS. MILAM: I think we are going to have Vickie help us with the
DR. CARR: You want to just frame up what are we about to hear? Larry or
Sallie, while Vickie is getting settled, just frame this for the audience
DR. GREEN: Socioeconomic status is important. We need to work on it more.
DR. CARR: So we have a workshop that is coming up March 8th and 9th. Who are
the participants in the workshop?
DR. GREEN: We are about to see that.
DR. MAYS: I am looking for a version of this. I can talk you through this,
or I am just looking for an old version that I have to put up.
DR. GREEN: While she is doing that I would suggest one thing is, we ask Jim
or Marjorie, either one or both, to again say where this work fits with what is
going on in the HHS.
MR. J. SCANLON: We talked this morning about the Affordable Care Act and the
data collection standards that were included as part of the Affordable Care
Act. That was generally in the area of public health and health disparities,
and it was part of the concept of looking at disparities in health status and
health care use and so on among vulnerable populations.
As I said, we adopted the first set of standards, and they are now being
implemented. But the Secretary has the authority to adopt additional standards
for these purposes as well, and so she, after some consultation, a lot of folks
said well, why don’t you look at socioeconomic status as the next set of areas
where you can look at, is there potential for standardization.
This is not something you would normally find in medical records, I will
grant. But the insurance company probably has it. So what we are looking for
here is, what is the current state of the art, and the collection of social and
economic data on federal surveys to start with, with a view towards what are
the best practices and is there the potential for recommending best practices,
or even a standard.
A standard, as you know, has a much higher level of meanings than simply a
good question or a good way of asking questions, or a good practice. The
standard, at least should that be adapted as a winging it kind of situation,
there should be a lot of evidence that it works, that it serves people’s
purposes. It is going to be around for awhile and it has some enduring nature,
and fits the goal.
Whenever you adopt a standard, you take away freedom and you take away some
variation, which is not always a good thing in research. So when you decide you
are going to do a standard, you have to do it in a way that is this.
So what we are asking the committee to look at, we decided the best thing
would be to look at the state of the art and the state of the practice, at
least in federal surveys now, in terms of what is collected relating to social
and economic status. So that would include income, educational level, probably
occupation and industry, although we don’t use those as much, and any other
variables that are — and people have done a lot of work here in the past in
terms of what would be part of that concept.
There used to be a focus on combining them into some index. I don’t think
that happens as much any more, although maybe in the universities it does. It
doesn’t really happen in the real world in federal surveys. It is amazing how
much folks want you to be transparent. And when you use an index, they can’t
see the original variables and it worries them a lot.
We had an interesting situation at HHS where we used some factor analysis
and some other multivariate techniques to try to commit the best way to
designate medical shortage areas. From a research perspective, researchers
And when we published it, everyone said we know what we used to do, the
number of doctors, the number of facilities. We don’t understand what you are
doing now. And they literally had to withdraw the regulation, because no one
understood factor analysis.
So again, we just have to be careful about when we look at this area. But
what the idea would be, through the committee’s own deliberations, through some
state of the art reading, through some public hearings, which is a quick way to
gather the information, get a sense of what is being used now.
Certainly the federal surveys, we have a pretty good sense, but they don’t
all do it the same way. They can’t even all generate federal poverty levels. We
do need some standardization there. With the idea of a minimum standard that
could be expanded for additional granularity, that could be collapsed as
needed. But anyway, what standard exists, it may be that SDOs and HL7 and
others have looked at this part of a demographic dataset previously. We would
want to know that.
But again, we are asking the committee, particularly through the
subcommittee on population health, to look at this and come back to this with
some state of the art assessment, and then what are the potential areas and
opportunities for standardization.
If you came back and said it is not a good area for standardization for
various reasons, that would be fine. But we do want to see — if you said
standardize income questions, occupation is already standardized, and so is
industry, and educational level would benefit from some standardization —
again, we could deal with that.
But I think there is a sense that we are missing something that has occurred
in thinking now, in terms of stratification analysis. We certainly don’t want
to miss it if there are some better ways of doing this. So the committee has
arranged a hearing, and I think Larry and Vickie are going to tell us where we
DR. MAYS: Thank you for that. I think that Jim has actually from this
morning given us an introduction. So the first set of slides I am going to go
through fairly quickly, because what Jim talked about this morning is actually
what we were just going to put on the table. In terms of talking about it, the
4302, as you can see they have already talked about this in terms of
establishing the standards for race, ethnicity, sex, primary language and
disability status. We are trying to do the same thing here.
In terms of the requirements, part of why we put this up is so that — I
know as soon as we want to talk about SES, people have a lot of places they
want to go with it. But Jim has been very clear about the focus that we need to
have rather than trying to create other things.
So again, here, what we are talking about is the standards, what we have to
do. It is for use by any federally conducted or supported health care surveys.
We have to keep in mind that whatever we come up with has to be practical. It
also cannot be super expensive. We have to think about the issue of burden.
Where will this go in terms of standards development? It is actually the HHS
data counsel that has been drafting the recommendations for the data standards.
There is a set of federal data standards that they are actually trying to
maintain. We have Susan Queen, who is in your office, who has been working with
us to keep us pretty focused.
So we need to make sure that, for example, the kinds of things that we
recommended that we know that they will work well in national surveys. We need
to make sure that we have heard from some of our partners, particularly those
with the surveys, and at some point in time, some solicitation of public
comments. We have talked about that, but not worked out where we want to go in
terms of that.
This is just the background in terms of the implementation of the standards.
I think I will skip this. And we actually can get to here, which is the purpose
of the hearing. The three things that we have decided to focus on, we have done
a fair amount of discussion with people in the field. And when you bring up
social-economic status, which is going to change in many ways and I think it is
a really good time for us to think about whether or not they should be changed
as to what is actually measured.
But in a standard way, what is the standard in the field? It really is to
talk about income, education and occupation. So that is what we are going to be
focusing on. We are going to also try and determine what are the best practices
for collecting this information, describe what those measures look like, how
well they work.
One of the things that we are adding is to try and get a sense of how well
do these measures work, for whom. And when we combine that with what the
measures are used for — in the surveys, for example, there is a range of
things. Sometimes it is used for program eligibility, other times it is used to
determine disparities. Other times it may be used to define household income
for participation in programs.
So we are really going to try and understand how well it works for whom. The
other is that what we are told to do is a minimum. So we are very aware that we
are trying to figure out across several different surveys that do many
different things, what a minimum might be. And we are also clear that it may
not work for all in exactly the same way.
One of the issues as we talked to people is that when you raise the issues
of SES they tell you right away, we don’t collect SES. They are very clear that
they collect those three variables.
So we always have to remember that what we are coming back to is providing
guidance on those variables that make up SES, as opposed to the concept of SES,
because there is no survey that actually does that. And again, I have said
these are the standard components for it.
A little bit about what we are going to be talking about in terms of these.
Let’s start with income. What we already know is that income is measured
differently across the surveys. I think that we are going to face quite a
challenge to actually come up with a minimum for that, because the surveys have
such different reasons that they need the type of detail that they might need.
So we will come up with what is a minimum, realizing that the surveys are
going to differ in the quality of the way in which they are measuring income.
We will struggle with issues of looking at individual versus household, single
income versus combined household income, the issue of poverty versus
Particularly when you talk to the census, they tell you that they are
mandated by OMB to actually give you measures of poverty. I don’t know how many
of you have listened to the controversies that are going on now in terms of
some of the new approaches that are being taken to the measurement of poverty.
So we will actually have a speaker, Connie Citro, who can talk a little bit
about the changes in the way in which the federal government has started
defining poverty. They have some implications for use and quality of data in
terms of SES.
DR. HORNBROOK: Vickie, as you are starting down this pathway here, as I
talked to you earlier it was the question of whether motivating this out of
regulations of HHS is the most appropriate way to deal with it at this
committee level. I was thinking if you look at the World Health definition of
health, economic opportunity is a phrase in that definition.
So you could motivate concern about health status and economic opportunity
and socioeconomic status from that definition rather than just the bureaucratic
one. So it says for some reason, it puts a bigger vision on why we are doing
this for the committee, but I don’t know how the committee chair feels about
MR. J. SCANLON: These are data collection standards, and it won’t be by
regulation. They will be adopted among federal agencies, actually it’s among
HHS agencies. Again, if there is a standard — there may not be — they are
already collecting this data, and some of it is being analyzed, some of it is
Some of the income data, I can understand why income, people need a lot of
variability because of their own programmatic requirements for income data. But
in many cases on the surveys, the information, you actually have a fair number
of degrees of freedom about how it is asked.
And again, the issue here is — this comes up in the Affordable Care Act in
the context of disparities, and identifying and minimizing disparities. So the
first set was race-ethnicity disparities. The others are other vulnerable
populations, persons with disabilities you could argue.
There are an infinite number of vulnerable populations, rural populations,
children. But I think income and education particularly are probably viewed as
the most explanatory, in a lot of the disparities research. And I think that is
why there is a desire for collecting this.
And this in no way minimizes or expands or interacts with what the WHO
definition of health or anything like that. This is purely in the context of
survey measures that we use. Can we standardize, and is there a standard that
we can agree on? And it won’t be a regulation.
Nobody else will have to do this other than our surveys. This is purely for
federal use at this point. Does that help?
DR. MAYS: Also, let me just make the comment about there is short term and
long term. And I will show you a bit of what we think the long term, broader
issues will be. But let me try and get through this quickly.
DR. CARR: Vickie, give me a little timeframe. We were initially scheduled to
go till 2:00, which would be 10 minutes.
DR. MAYS: I am going to whip through the rest of these a little faster.
Anyway, it includes the income. You can see some of the thinks we are worried
about. Linkages, we are worried about it is the variable that is most difficult
to actually get. Education, we are probably going to be focusing on the issues
of the difference between educational attainment versus education, the quality
of the education variable, how it is that education is used in terms of social
We have been told that part of what we have to worry about as it is measured
currently, is it picking up all these new forms of education such as
certificates, et cetera. So that will be a part of what will be discussed.
Occupations is probably the most difficult. It is probably the most
underutilized to some extent. It has most difficulty. We have even had that in
terms of how much work is being done on that. So we will talk about that, and
as well, talk about the issue of the linkages there.
So in the frame of the hearing, the potential for STEP is the short term,
which is really looking at individual family, which is really looking at those
three things. I think in a longer — which I may not have that slide — the
longer view is that we would do exactly what Mark was discussing.
Questions, and I think Jim has almost outlined the way in which we are
asking these questions — how is it defined, current practices, best practices,
how is it used. We are really trying to deal with things like use, usage. We
are trying to deal with quality of the data. We are trying to deal with what
the gaps are, how this drives poverty.
Can we apply a standard or homogenize the measures across national surveys?
Again, I don’t think we are going to know that until we get into the workshop,
into the hearing. This is our goals, and that is to come up with that
recommendation of minimum standard.
What we will discuss in the population subcommittee is — and we have begun
to think about this. One of the things I can say that I am learning from
watching the other groups, and also Justine kind of reminds us, is that we need
to know as we go into this what it is that we are actually going to give to our
client, which is Jim.
So we have begun to talk about what the letter should look like. There will
be a letter with a set of recommendations. And there is where I think going
beyond the minimum to say some of the issues that we have learned, to also talk
about it more broadly and to talk about what we think would be some other next
steps, is where that would go.
In that letter we also want to try and talk about data quality and gaps. We
want to point towards more a dynamic approach in terms of thinking about where
we want to go. So I think part of what, Mark, you are asking, we will try and
do. Just to give you a sense, and most of you probably know this, these are
surveys that are on consideration.
We have major surveys coming in to talk with us, like MEPS, and NHIS, and we
have the census and ACS coming in. We will have on our second go-round somebody
for the National Center for Education Statistics we will probably be working
with. These are other surveys that are important to think about.
Timeline, just so that you have a sense of that — we have a workshop March
8th and 9th. In April we will have a session that we will convene. It is set to
be online. The purpose of that will be for us, after we have heard this first
hearing, to figure out if there are any things that we think that we have
missed, we are going to pull it in by doing a webinar.
During the month of May we will work on the report, and then during June we
will send it through to the Executive Committee, and then onto the full
committee for the end of June meeting. And then in Populations after we reach
that point, what we would talk about are more looking at the long term and
dynamic metrics for measurement of SES.
And also, what we think is really important to talk about is the issue of
linkages. That is probably the big piece. We can see from pushing that linkages
are where people want to go, and that is where the innovation is going to be.
For example, in education, if you could link into data on standardized test
scores then you would raise the quality, for example, of the education
measures. So there are ways in which we are going to raise some of the issues
around innovative linkages.
The next thing we would talk about would be SES in terms of administrative
records and electronic health records.
DR. CARR: First let me say thank you. This is perfect. It sets the stage, it
educates the full committee, we know what to look for, and it is very exciting.
So thank you very much for putting this together.
DR. WALKER: Just in terms of usability and minimum necessary in terms of
minimum necessary data, I would recommend that the subcommittee have a goal of
having education and income and five categories within each, and justify very
carefully anything more than that.
DR. MAYS: Is there a particular reason? Or should I talk to you offline?
DR. WALKER: Yes. It is short enough and usable. I think everyone
experiences, you all the time get pick lists for income and education. Everyone
that does pick lists like that is used to them, and they are pretty clear, and
you don’t say oh god, I don’t know what to say.
Occupation, I have never seen an occupation pick list that I said, oh, that
captured me. Income and education, unless there is evidence that occupation
adds additional information — and that is what I mean by justification. If you
could say, look, you get 30 percent better information if you ask occupation.
Well then maybe you’d be worth fooling with. But if you get two percent, or it
doesn’t actually improve your ability to understand the people you are talking
about any better, then just don’t do it.
DR. WARREN: I was glad to see that you had on future the linkages to EHR.
Because as I think about the work that nurses do, when we work with our
patients to figure out how to teach them to manage their conditions and get
them ready for discharges, and then follow them throughout the venues of care
— it is that critically important.
We know what their income levels are, we know what their education levels
are. And probably everything else that is on the list. And people are looking
for places to look at standards. I know when I have looked for standards in
this domain, I went back to the IOM report that looked at all of this, because
it was best one that was out there, and had the best data.
So I have constructed various EHR items based on that report. So as you are
looking at this, then look at the linkages. If we could get those standards in
the EHR, now you have got them to where you can pull them out and attach them
to the surveys. So it is back to that future thing that Paul was talking about.
It’s let’s look and see how we can pull the data up. And don’t just look at
the data elements for surveys only. Where else can they be used and for what
DR. MAYS: Just one comment. Bob Kaplan, who isn’t here, although he sits on
this committee, he actually had some beginning work that they are doing at NIH
And so hopefully — I have been kind of encouraging him, to see if there is
a point at which he might want to discuss that, at least in populations, but
maybe even for the full committee. So there are some very interesting things
occurring on that front.
MS. QUEEN: One comment about using a categorical variable for collecting
income is that you really can’t do poverty estimates. So we need to have as
great granularity as possible even though there are challenges, along with the
household size and composition.
We don’t want to be like the CPS, the annual supplement, that has something
like 50 items on income. We don’t want that. But we unfortunately do try to get
more of a greater granularity.
DR. FRANCIS: I just have a quick question. I seem to recall that at one
point you were talking about questions about vulnerability to decline. So for
example, in figuring out socioeconomic status, it’s not just how much income is
coming into the household, but what is your risk of losing that income, or do
you have health insurance, do you own a home, what are your buffers.
I just wanted to know whether there is any thinking about trying to include
any of that, or is it just too hard?
DR. MAYS: I think for a minimum standard it is very difficult to do that.
But in terms of the perspective — and unfortunately I don’t have the template
that Bruce made in here, just a little piece of it. And in the template, when
we do the bigger picture of what is short term, longer term and dynamic, in
order to be able to really talk about those things we are interested in things
like health insurance, where you live.
There is a variety of things that are in that template. But to get through
to just the minimum, we really are going to stick to what is standard practice
in the field. But we are looking towards making a case for better ways to
MS. QUEEN: Vickie, it is Susan again. In our background document we were
including a number of variables across the survey, such as home ownership,
characteristics of the home, assets, debts, viability, because there was quite
a broad array of information from the various surveys on all these things.
MR. J. SCANLON: Remember, this is not to deal with all the information we
should be collecting or are collecting. This was a very specific requirement
that we begin to standardize some of what we are collecting. And I think
probably justifiably, the focus was on a core demographic dataset, the first
elements of which we talk about previously.
And the second one would be, just by popular demand, social and economic
status, particularly income and education were the first sort of voluntary sets
that the Secretary asked us to do. The goal here would be, if it is desirable,
a standard way of asking income questions, a minimum standard.
So that you might have five standard categories and depending on what you
needed it for, you could expand the granularity. But it would depend on why you
needed it and what you thought you would get. The same thing with educational
And interestingly enough, we do a lot of analysis of disparities and other
things by various variables. It is frequently the case that income and
education often turn out to have the most explanatory power. They are not only
— and you combine it with race and sex and age — you are pretty much getting
the variables that explain most of the variance.
We don’t know where this is going to lead us. But I think if this could be
standardized, this would be a great help.
DR. COHEN: Thank you, Vickie. That was great. There are two focuses for what
we are doing. One is what Jim just said, coming up with a minimum dataset that
would be useful now. But the second focus really is, once we have the experts
here, to discuss the vision of where we need to take these conceptually and
So I think both things are going to be exciting, one, for what we can do
currently, and just as important is to begin thinking about where we need to go
in this whole area. So I think that is a really robust design, keeping in mind
Paul’s and the committee’s notion of visioning about not what just fits today
but where this all needs to be.
DR. WALKER: I would just propose that we think of these questions as a
health care intervention. And the appropriate way to do health care
interventions is to study them and prove that they have more benefit than risk
or cause, and then start using them. And not to start using them because it
sounds like it might be a good idea.
DR. COHEN: I am going to refrain from responding. But we can talk later.
DR. CARR: Thank you.
MS. GREENBERG: Remember that this is about a standard initially for federal
surveys. And even then, your voluntary survey respondent has his or her limits.
And there are a lot of other things that are being asked in these surveys as
well. But you can ask certain types of questions.
First, it is voluntary. It also is a one-time thing. You can ask certain
types of questions and probe more in some surveys, not in all, than you can if
you are thinking about, well what are we going to collect in administrative
data or electronic health records. Clearly, that requires a lot of additional
burden and other concerns.
DR. WALKER: But I would posit that even in that setting we shouldn’t ask
questions we don’t know what we are going to do with the answers to.
MS. GREENBERG: I agree totally with that, yes.
DR. COHEN: I think it is a researchable issue. I don’t know what is more
important for an individual’s health, his particular income, his family’s
income, or the stress caused by how different his economic circumstances are
compared to his neighborhood. So I don’t know that we have answers to these
questions yet with respect to your fundamental point, what has the biggest
impact on health outcomes.
DR. GREEN: So SES matters. It needs more work, and we hope to see you next
DR. CARR: Great work. I look forward to hearing what you find. Now I want to
welcome Floyd. Floyd Eisenberg is Senior Vice President of Health Information
Technology at the National Quality Forum. And Floyd, thank you. Good to see
you. Just as a framing comment, we talked at our last meeting about ICD-10. Who
knew? And talked about the downstream implications.
Because we have built so much of our quality measurement on ICD-9, we raised
the question as to how will that be mapped, and how will measurement continue
seamlessly with the transition. And so thank you for coming today, Floyd.
I will also point out at your place Marjorie has prepared the history of
ICD-10 development timeline. We are talking about ICD-10 and mapping from
ICD-9. Anyway, what we have at your place goes back to 1993 to the present
history of ICD-10 and the work that has been done it, and particular work done
by the committee.
MS. GREENBERG: It actually goes back to 1979, when 9 CM, which we are
currently using, was implemented. But this was a request from Linda Kloss, who
is a member of the committee but wasn’t able to be here. She suggested I update
some other timelines we had. That is where it came from.
DR. CARR: I think the point is well made. This has been a topic that has
been around for more than a quarter of a century. So more on that. But going
back to where we were, what we knew in November, that with the imminent arrival
of ICD-10, what would become of quality measures. So Floyd, welcome and thank
you for coming.
MR. EISENBERG: I am happy to. And yes, I do predate ICD-9 then, I guess.
Let’s not worry about that. So I actually provided some information. NQF did
have an expert panel providing a consensus report for ICD-10, CM and PCS coding
maintenance operational guidance for measure developers published in 2010.
And I will refer to the timeline related to that in the slides. But I did
provide the report to you as well. I will also refer to, on page eight, it
identifies best practices for measure developers to assign code lists or value
sets to their measures using ICD-10.
So what I will be talking about is our timeline and our documentation today
about ICD-10 requirements, and I will have some additional comments regarding
quality measures. The transition, as we have asked for, started requiring
ICD-10 for all new measure submissions starting in October 2011.
We had anticipated requiring ICD-10 for all. So it was accepted as of
October, required. But it was anticipated that as of October 1, 2013, all HIPAA
related transactions that were used for the measures would be required for
measure implementation. We are now awaiting a new date from the Secretary for
when that is implemented.
And the expectation that we would no longer accept ICD-9 for measures coming
in for endorsement, maintenance or update as of January 1, 2014, again, we are
not certain about the date with the new announcement, but that is the plan.
So the next slide basically provides a timeline starting in March 2010. We
put out our code maintenance report in January. We started accepting measures
that contain ICD-10 as well as SNOMED. And as of October, 2011, we have
required that all new measures and all updates and all maintained measures come
back in with ICD-10 value sets.
Starting then between now and December, 2013, again, we require I-9 and 10,
and at such point as the Secretary determines I-9 is no longer acceptable, we
will no longer accept I-9 in the measures being submitted. So that is the
process for measures for endorsement. So the question then is, so what is it we
ask our measure developers to provide to us?
And this is related to the best practices that you will see in the
DR. SUAREZ: Floyd so you said as of October of last year the measures have
to come with both ICD-9 and ICD-10?
MR. EISENBERG: Yes.
DR. SUAREZ: What has been the experience? Or are you going to talk about
MR. EISENBERG: The experience is, they are coming in with I-10. I can give
you a little more experience also from a retooling project in which we were
involved and what I understand about what some contractors from HHS are now
doing for e-measures, electronic measures. So of the 113 that we assisted
measure developers to retool, all conditions were provided in I-9, which they
They also provided us in I-10, a diagnosis list for a few procedure lists.
They weren’t quite familiar with doing that yet on procedures back in 2010. But
all did provide I-10 diagnoses, CM lists. And they updated those again in 2011
for those 113 measures.
That was less problematic than some measure developers trying to provide
SNOMED value sets for the same concepts, where they weren’t familiar with
SNOMED. So, almost every one of the ones that uses a diagnosis or condition,
has an ICD-10 in the e-measures as part of the submission.
DR. SUAREZ: And the submission is done by providers?
MR. EISENBERG: The decision of how to develop the list of codes in ICD-10 in
the retooling project was done by the measure developers themselves. Some of
them do it routinely with their providers, with their experts, and come up with
the list. And some of them did it on their own for retooling. When it comes in
for endorsement or maintenance, we do recommend the use of best practice.
And there are six steps. So just refer to them: that they convene their
clinical experts as well as coders, who understand the terminology, that they
determine the intent of what they are trying to say, then develop the
appropriate list, either using a conversion tool — and I believe Vivian Auld
will be talking about conversion tools that NLM is working on now and has
available — or do just create a brand new list of codes for the ICD-10.
The next step is, assess for material change comparing the two lists. And
that can be done either as face validity, but preferably to test the measure
against a set of data that have dual lists, I-9 and I-10 coding. That is not
always possible, but that is the preferred. And then, solicit comments, and
then update the measures. So there is a whole set of best practice we
DR. COHEN: Who does this cover? Who is required to adhere?
MR. EISENBERG: This is for all measures being submitted to NQF for
endorsement. Measure developers. So anyone submitting for endorsement, existing
measures coming back for maintenance, all measures now go through a three-year
measure maintenance cycle. Or, if there is an update off cycle from the three
years, they are to present them with their updates.
But it is only measures that are coming through NQF for endorsement. The
guidance is out there for anyone to use. But the only requirement is for those
submitted for endorsement.
DR. SUAREZ: If a provider wants to submit the data using this measure, they
can use ICD-9, or now ICD-10, because it is already defined by the developer of
the measure, right? That is what the intent is?
MR. EISENBERG: They could. We have not assessed all receiving organizations
to see if they would know how to interpret it. But if it were calculated
locally, it could be in 9 or 10. Or potentially SNOMED, for those that have
SNOMED value sets in them as well. So the first one they do submit, we ask them
to tell us what is your intent. And I will describe that briefly. What are the
codes, and what was the process you used to submit them. And the reason for
that is so they can be evaluated by the Steering Committee.
So if we look at the intent, was the intent just to take the measure as is,
and convert it to add a new code set in, a code list in ICD-10 where you have
one in 9? Or, was it to take advantage of more specific granularity of the
codes? An example would be asthma, where it is an easy to understand and common
example in ICD-9. I can indicate a patient has asthma.
In ICD-10 I can indicate persistent asthma, but in 9 I can’t. So number two
is, I am taking more advantage of the better codes to get better information.
But my intent in the measure hasn’t changed. The third is, I am also at the
same time changing the intent of the measure. I really want as a measure
developer to get to more information, and identify more.
And I couldn’t before, so I am changing even the intent of the measure. And
I might pick up a different population, and I might be looking for
interventions I couldn’t otherwise identify before, so I have a new intent. So
that is what we are asking them to tell us up front.
Of course, in the analysis, the Steering Committee that reviews this, which
is a committee of peers that are selected from those who submit names to
evaluate the measures, will also determine is this a material change or not.
The next step, of course, is to look at the codes. So we asked for all the
codes in 9 and 10. We asked them to provide the code descriptors or
definitions, so folks can read them.
And also, if you had a conversion table to provide the table as well, so if
a conversion was used. So all of this information, again, is reviewed by the
Steering Committee, and basically where is it used in your measure, in the
numerator or the denominator exclusions, or wherever.
The third thing we asked for is, we do want to have the names and
credentials of the experts who were reviewing this and evaluated and created
this new list, a summary of the stakeholder comments, because we do expect that
they have had this out for comment, and results of any testing, whether that
was face validity or in the field testing. Because that gets evaluated as part
of endorsement, to see if the intent of the measure is achieved by use of it.
So that is our current process. And this is just a link to where on our
website we have the report. But I did provide it to you, and I have an
electronic copy I can provide as well for those who need it. So that is the
Again, in the retooling process, in what I believe is going on now with all
of the HHS contractors creating electronic measures, because of recommendations
by the HIT Standards Committee providing for procedures, SNOMED, ICD-10 and
ICD-9, the PCS for those, too, and for diagnoses, all three as well.
We did that to the extent we could in 2010, but not everyone was yet
familiar with SNOMED, or ICD-10 PCS. So that is where they are now.
DR. CARR: Just a question and then we will go around the room. What does the
continuity look like, with the change from ICD-9 to ICD-10? If you take a
dataset and apply the ICD-9 description of the measure, and then you compare
that with ICD-10, is there continuity? Is it a parallel response?
Or is there divergence, as you get to greater specificity or granularity
with ICD-10? Do you then get a bimodal population? Maybe one group scored 85
percent, but now you go back with the new ICD-10, and suddenly you have one
group at 70 percent, and one group at 95 percent?
MR. EISENBERG: At this point we are just starting to review, get some of
those measures coming through Steering Committees? So we don’t have all those
data. But I can say that, again, some of these will actually increase the
specificity of who I can look for and perhaps, if we use the persistent asthma
example, lower the number in my denominator, because I am only really
interested in persistent rather than all asthmatics.
And that would change the intent of the measure actually. So that has to go
back through review. So I can’t tell you percentage wise what percent
difference this is, at this point. And a lot of it depends on testing. I don’t
know that we have had any specific dual testing where patients are coded both
ways and they are testing against that.
DR. CARR: So as we are adding patient safety indicators, complication
indicators, all publicly reported, I think that what prompted part of the need
for this discussion was a request that we understand what we can expect with
that change in coding. Because the significance, the implications, the many
ways that it is used, pay for performance being the obvious one, but public
reporting, all kinds of quality assessments depend on that.
And I think that if there was a very disruptive segregation of people who
once looked the same, I think we would want to understand that. And then also,
understand the impact of familiarity. It’s the documentation, it’s the new
coders in a new system, and then the measure. So a lot of variables, but a lot
riding on it in the end.
MR. EISENBERG: We don’t have the data yet. And I would also have to say that
you might have the same questions if we moved to SNOMED, because that is,
again, a change in documentation.
DR. CARR: But I think we need to be preparing for what we are going to find.
We have so much emphasis on measurement of quality, and so much of it resides
on administrative data, that to suddenly flip a switch and have everybody
re-shuffle in where they appear for reasons that may or may not be related to
quality, I think that we would want to anticipate that.
We talked this morning about pilots, when we make changes. So that we are
aware of the vulnerabilities and we can address them before we flip the switch
DR. WALKER: A quick background. One of the good things is that ICD-10 is a
barbaric language clinically, and one of the large tasks that most
organizations that work hard at this spend time on, hundreds of person hours,
is creating presentation layer diagnoses and other clinical things that are not
ICD-10. So we have been hiding ICD-10 from our clinicians, all our clinicians,
for years, very carefully.
DR. CARR: Using SNOMED?
DR. WALKER: No, we just make up our own names. We just say diabetes type II
MS. GREENBERG: You mean ICD-9?
DR. WALKER: Yes, ICD-9.
MS. GREENBERG: You said ICD-10.
DR. WALKER: I’m sorry, ICD-9 is almost un-interpretable to clinicians. If
you give doctors ICD-9 straight, they have trouble knowing what it means. And
that has been true forever. And so it is unlikely that clinicians will see big
changes whenever the switch is flipped.
ICD-10 and SNOMED are both prone to be more clinically interpretable without
translation into a presentation layer. So we may use them a little bit more in
native form. But that should not be an issue. There will be plenty of others.
DR. TANG: A couple of other things. One is the comment I made earlier about
maybe this is an opportunity, given the delay, that we could rethink how do we
best use standards of some type having to do with clinical concepts. And is it
going to a classification system in an EHR, or going to clinical terminology,
and having that mapped in the background to a classification system.
That feeds into our data mission of this group. The other comment is, in
addition to what Justine mentioned, when we changed the code sets as we moved
from paper based systems with coding in the back end, to EHRs with coding on
the front end by the physician, there is likely to be a change literally in the
So applying the same rules to a new coder, which in this case, in the
ambulatory setting, is going to be the physician, you may end up with very
different results in that way, too. So it still supports your notion of the
industry really needs this information on what does happen with these two major
things. One, a change in a coder, basically, to a physician. And the other is a
change in the coding system.
DR. COHEN: When Vital Statistics moved from ICD-9 to ICD-10, there was a lot
of work done on developing comparability ratios so that trend analyses could
continue reasonably well. Is there any work going on around comparability
ratios between ICD-9 CM and ICD-10 CM?
MR. EISENBERG: We are looking for comparisons, when people bring their
measures in. We just haven’t seen enough yet to give you any data on that.
DR. COHEN: Do you know of any plans to generate comparability ratios on wide
sets of data?
MR. EISENBERG: As far as comparability ratios, I don’t know. I do know that
since one project that is definitely doing this is those that are being
retooled for Meaningful Use, stage two, where they might already have been in
existence, of course there is another variable. You are now looking for data in
a clinical EHR, that you were looking elsewhere before.
They are asking for testing by the measure developers and the contractors.
That is either just beginning or soon to begin. We don’t have any results from
that yet. And I am not sure how detailed that will be. That is all under HHS
work. I don’t know the answers to that yet.
DR. CARR: Marjorie did you want to add a response?
MS. GREENBERG: Yes, and I may call on Donna Pickett as well, who is here. It
is true that we did comparability, this is standard practice, and mortality
data, the comparability ratios, when we went from 8 to 9, 7 to 8, 8 to 9, 9 to
And in order to do that, you actually need to have the same information
coded both ways. And we can do that with mortality data because we get the
text, which then is mostly coded in the automated system, but some of it ends
up having to be coded manually.
We did do that when we went from 8 to 9 CM, with the hospital discharge
survey. We were able to do that because we were still getting a lot of the data
that we were coding. Now we get almost — NCHS, when I said we — now we get
almost exclusively coded data.
So we don’t have the records from which we can code both ways. We did do a
pilot several years ago with AHIMA in which health information management
professionals, coders, did code records using both 9 CM and 10 CM.
But for the mortality data we do the entire 2 million records, or however
many there are. You need quite a large volume of data, which would be quite a
bit beyond what was done in that study, to be able to do actual comparability
We looked into, several years ago, whether we could try to do this again,
from going from 9 CM to 10 CM. And we had a study, a contract, et cetera. But
it was determined that certainly NCHS couldn’t do that without getting —
because we weren’t getting the actual diagnoses or the actual texts. We were
only getting codes.
So we have the general equivalence maps. We do have information from the
mortality comparability ratio. But of course, the rules for coding mortality
are quite different from coding morbidity. We don’t have the capacity to do the
actual comparability ratios.
I know that CMS has plans, obviously, to evaluate the data. They are not
going to make any decisions, once we do start, if we do start collecting data
in 10 CM, it would be a few years before they had enough data to be able to
then make comparisons. But even then, I don’t know of a large scale study in
which text will be coded in both ways, that would allow those comparability
There are other things, though, being planned, and we can report on those
DR. FITZMAURICE: Floyd thank you so much for a very clear and simple
presentation of how NQF is making the transition from 9 to 10. I am going to
pick on the most complex of the slides, which is the ICD-10 timeline.
From October to December 2013, all measures submitted are specified in ICD-9
CM and IC D-10 CM or PCS. Does that mean that for procedures, the quality
measures will be using ICD-10 PCS as opposed to some other procedure coding
MR. EISENBERG: Yes, instead of ICD-9 PCS?
DR. FITZMAURICE: Instead of maybe CPT?
MR. EISENBERG: Well, I think they also accept CPT. And, I guess when I saw
the NPRM for certification, it also includes CPT and HCPCS as well. So I think
we will accept all. But we do require ICD-10 PCS.
DR. CARR: I am going to ask another question. Doctors write things down, and
coders code them, in the hospital. In the offices, it is the admin, nurse,
wife, whoever is there, billing specialist.
DR. TANG: In the ambulatory setting, the docs code, in an EHR. In some
DR. CARR: In some settings. I think it is all of the above. Traditionally,
at least in the institutions I have been in, if a coder has a question they
will query, was this a sense of this, a sharp debridement, whatever it is.
Actually going back now about five years ago, I think there is a role of
educating coders to the measures that are used for quality.
So that in addition to assuring that the resources consumed are properly
represented, I think that there needs to be attention to these codes that drive
the quality evaluation. And I look to AHRQ as a resource, in terms of the
coding for quality guidelines.
So that when we choose a measure, endorse a measure, begin to use it, that
we close the loop so that the coders understand the importance of the
specificity such that they would go back and ask for clarifying information to
achieve appropriate specificity on those codes. Judy?
DR. WARREN: Part of my question is, I think that is a great idea. AHIMA has
been a great partner, and it is a shame that Dan and his colleagues are not
here. This is one of the first, I think they have missed. But they have been
invaluable in working with HIMSS and AHA in putting forward ICD-10.
But the other thing that kind of keeps niggling at the back of my head, and
Paul was referring to this, and so was Jim, is when we held the testimony about
ICD-10, we really tried to clarify who would be impacted.
Meaning, who is going to actually deal with the code? And pretty much what
we found out is the people that deal with the code are going to be the EHR
implementers of the organization, trying to figure out what table structure to
put it on so they can hide the words from the clinicians.
And I have built plenty of those tables. Let them call it what they want to
call it in the back end. It is this. And that will be on the report. We heard a
lot from some of the physician management organizations and from some of the
small practice. We particularly brought in physicians from one and two man
practice plans, to find out what they did.
And what we heard uniformly for them is, none of them dealt with the codes.
They either had the wife, the secretary, someone in the family, or they jobbed
that out to a service that they paid for. So now, I am beginning to wonder,
don’t we need to have some best practices for how to manage these standardized
terminologies that are coming out.
I get very concerned, because my own special area is SNOMED, and I
continually see publications that come out, that talk about how bad SNOMED is.
And then the examples they give are in error, because they didn’t know how to
MS. GREENBERG: We see that with ICD all the time.
DR. WARREN: I am sure. I have talked over the years with that. And I
remember one of the first NCVHS meetings I came to, we had someone from CMS who
said that they had done a study that periodically they looked at what the codes
were, and there was only about a 50 percent agreement between the staff and the
coders of the documents that came in, which really kind of surprised me. I
thought we would be a little higher than a 50 percent agreement. It is like so
much else the NCVHS does. We start pulling the string and you find all the
So I think maybe what we should look at somewhere thinking about it, is best
practices to help people deal with the standard terminologies. So we get the
quality right. We get the population health right. We get the billing right. We
get all the surveys right, and all of those. Otherwise, we are still going to
be in a Tower of Babel.
MR. EISENBERG: I fully agree with that, Judy. I think if we could also make
the case that there are some measure developers, in selecting their lists of
codes or their value sets, that are very well versed in SNOMED or ICD-9. And
others that are not.
And I think the same thing you mentioned on the practice side, to make sure
that there is an appropriate best practice and people are aware, is also on the
query development side, is needed as well.
DR. CARR: Floyd, thank you very much. And now we are going to hear from
Vivian Auld, from NLM, no stranger to the committee.
MS. AULD: So thank you for having us back. We wanted to talk about mapping
from ICD, both 9 CM and 10 CM, and also, just plain 10, and the maps to SNOMED
CT. Just to remind people, in case you are new to this, NLM’s role, the reason
that we are here talking about this, is that we are the distributor of HIPAA
classifications and code sets, and terminologies that are required for
electronic health records certification. And those are distributed to the UMLS.
We are a designated body within HHS for coordinating clinical terminologies,
including the relevant mappings. And we deal with three primary terminologies.
For LOINC we are a major funder, RxNorm, we are the developer, and SNOMED we
are the payer of the US membership in the IHTSDO that allows free use within
But today I am going to talk about mappings, specifically from SNOMED CT to
ICD-9 CM, to ICD-10, the international version, to 10 CM, and then the other
direction from 9 CM to SNOMED.
And what we are really trying to do is narrow it down so that we are
providing maps that are specifically addressing problems that people are
encountering today, rather than trying to map everything because nobody in
their right mind would need to use it, and nobody can afford it. Instead, just
focus on specific problems.
So the first problem is people who need to transition from the use of ICD-9
CM for problems. Right now, today, there is already available from the IHTSDO a
conceptual map that goes from SNOMED CT to ICD-9 CM. This predates the creation
of the IHTSDO. I think this map has been there, I don’t even know how long,
more than 10 years.
But it is essentially an equivalence map. Right now, because there had been
the announcement that we were transitioning in 2013, the IHTSDO, through the
advice of NLM, has pulled back. And they are only doing minimal maintenance on
that map so that it matches the updates to SNOMED.
As we hear more about the new date for 10 CM, then we will regroup and
figure out whether or not that needs more maintenance in the future. Available
from NOM right now, you can have synonymous maps that are automatically created
through the UMLS Metathesaurus. Because of the way that we create the UMLS, you
automatically have links at a concept level between everything that is in
So, if there is a SNOMED code and an ICD-9-CM code and an ICD-10-CM code
that all mean the same thing, they are all linked under the same concept unique
identifier. Automatically done. And then by April of this year we are going to
have a trial map ready for evaluation that will look at some of the heavily
used ICD-9-CM codes and map those back to SNOMED.
This is a harder direction to go, because you are going from general to
specific. And it is not always a clean conversion. So we are using CMS data,
the codes that are used in the Medicare program, to inform us of what we need
to focus on for those. And I don’t think I have the numbers on what we are
planning to use for that.
But only about roughly 40 percent of those will not be a one to one mapping.
It is just the nature of the beast.
DR. WALKER: Just for clarify, so it might be prostate cancer in ICD-9, but
SNOMED has five different varieties of that. And so that is why it is not one
to one. But they are still clinically rational, just not one to one.
PARTICIPANT: One is the terminology. The others are classifications.
MS. AULD: Precisely. The second problem is people who need to use SNOMED CT
to generate encounter diagnoses in ICD-10-CM or 9-CM for billing and
statistics. And here again, you can use the same math from the IHTSDO, the
conceptual map from SNOMED to ICD-9-CM. You can again use the mapping that is
automatically available through UMLS Metathesaurus from NLM.
We do have an old map from SNOMED CT to ICD-9-CM that was specifically
created to support reimbursement. It is a rule-based map, but it was created in
2008 and put up as a test. And then it has not been maintained.
And actually yesterday Betsy was at the HIT Standards Committee and she
asked for advice, do you think we should now start maintaining it? So that is
under debate as to what is going to happen with that. It very much was used as
a proof of concept that fed into future work that was done by the IHTSDO, and
then successively by NLM on subsequent maps.
Just yesterday, literally yesterday, we published a new map that goes from
SNOMED CT to ICD-10-CM. This is the preview of the map. There are 7,277 SNOMED
concepts represented in this preview, and it is up on our site, and I will
spend some time talking about that. And in July of this year we will have
expanded that to the first full release, with about 15,000 codes.
We have very much leveraged, as I said, the work that we did on the SNOMED
to ICD-9-CM work, the work that the IHTSDO has put into creation of a map from
SNOMED to ICD-10, the international version, the tools that were developed for
that map by the UK National Health Service, or leveraged by NLM.
The processing was created by NLM and given back to the IHTSDO. So there has
been a lot of give and take, and it has been a really rewarding experience. It
has allowed us to benefit from an international expertise, rather than just the
DR. TANG: Just a quick question on the 15,000. Did you do it also by the
frequently used codes, and then about what is the coverage.
MS. AULD: So this is a screen shot of the page that describes the map and
where you can go to get it, and the URL is posted there. There is also a news
announcement on the NLM’s front page. And if you have problems getting it, just
write to me. I can send you the link.
The intended use of the map. This is one thing that we learned from the
9-CM rules based reimbursement map, is people really did not understand what we
intended to do. They took the map to test, and they gave it to their coders.
And they said here, just use it. It is not meant to be used that way. It is
meant to be embedded in your EHR for real time, interactive ICD code
And Kin Wah Fund, from NLM’s Lister Hill Center, he and his co-workers, have
developed a tool called I-MAGIC that does a beautiful job of showing you how
this could work. And I will show that to you in a minute.
The map is also intended to assist coding professionals by suggesting ICD
codes based on SNOMED CT encoded problem list entries. So it is intended that
your systems experts will take this map and embed it into your system so that
it can provide an interface. It is not intended just to hand to coders. You can
do it if you want. It is not a good idea.
So I keep talking about the two maps, the relationship between the SNOMED
CT, the ICD-10 map, and the SNOMED CT ICD-10-CM map. The 10 map was a joint
effort between the IHTSDO and the World Health Organization. NLM was a very
active participant in the IHTSDO special interest group that was working on
Other participants included folks from the UK, Canada, Sweden, Australia and
New Zealand. And that project was funded by IHTSDO with a considerable amount
of volunteer effort. The map to 10-CM was led by NLM, funded by NLM, and we had
quite a bit of help from NCVHS. David Berglund was instrumental in making this
And the methodologies and tools were based on the 10 map. And then we put
all the processes in place, and we have automated those, so it makes it easier
for everyone. One thing that is going to happen in the future is that the
IHTSDO is now recognizing the fact that if you handle maintenance of a map
through a contract process, you really end up having to pay more than you
should have to.
So they are looking into putting in place a mapping service that will be
available, not only to handle IHTSDO mapping priorities, but also available for
the members. So that if we have a map that we need in short order to have
created, we can provide the funds, or maybe we won’t have to provide the funds.
It depends. But they would be able to assist us in getting that up and running
much sooner than we would be able to do if we had to go out for a contract.
The priorities for the phase one of both of these projects, for the 10-CM it
was based on NLM’s CORE problem list subset. And the SNOMED CT concepts that
corresponded to the top ranking ICD-10 codes that were used in five countries.
And if you want to know what five countries, I will have to go back and ask and
tell you later. I forgot to find out earlier.
The 10-CM map is also based on NLM’s CORE problem list, plus we benefited
from the materials that were provided by Kaiser Permanente in their Convergent
Medical Terminology. They had provided their top 2,500 problem list that they
use in their system. And they have also provided problem lists for several
subspecialties, including cardiology, neurology, mental health and
The top 25 did not include a mapping to 10-CM, but the subsequent
contributions have included that. So we have taken that and used that to
enhance the map that we are working on. And it is allowing us to go through
much faster, because they have a very robust system in place for doing their
maps. So again, we are benefiting from all these people coming to the table.
The scope for the maps — any concepts in SNOMED CT hierarchies of clinical
findings, events, and situations with explicit context. And there is a total of
about 110,000 SNOMED CT concepts that are within scope for this map. SNOMED is
roughly 300,000 to 350,000 concepts. So it is only about one-third that would
even possibly be relevant.
And we are willing to bet that not all of that would be used. So we are
trying to focus on what would be used, and then set up a situation where you
can request additional mappings as you find that you need them.
The progress of the work — as I said, the 10 map. The preview publication
of that map had 5,000 concepts and it came out in September of this last year.
The final publication of the phase one map will be about 20,000 concepts, and
it should be coming out in April of 2012. And as I said, for our map, the 10-CM
map, we just released the first 7,000 concepts, and we will have 15,000 by
I just wanted to give you a few screen shots from the I-MAGIC tool that
allows you to look at this map, because it is very cool. It is intended as a
demonstration, and we have this set up so that under the name you have some
canned searches, patients who are already saved in here so you can take a look
at how they work.
And so I chose — well first, this is a flow chart that shows how the logic
goes through. So I chose the male adult one, and this patient has a series of
diagnoses or problems that are listed in SNOMED. And if you click the Get ICD
Codes button, then it will go ahead and it will show you for each of these
SNOMED problems, these are the related ICD-10-CM codes that match that.
And it is designed so that you can, where it is appropriate, it will tell
you that you can refine the problem, get something. We have given you the
general, but there might be something more specific that is relevant, so you
can click on that button. And it will give you additional possibilities of what
you could choose to make it closer.
The other thing is that if there is ICD notes that might apply, that might,
again, make you get something a little bit closer to what you want, you can
click on that button and it gives you the additional information, so that you
know whether or not it applies.
And this was just another example of the patient. And there is the URL at
the top of this page. So that is all I had to tell you. Questions?
DR. TANG: My question was how much does the 15,000 cover? What percent of
the total diagnoses does the 15,000 cover?
DR. WALKER: I don’t know for sure, Paul, but I think the 15,000 is supposed
to be about the number of billing codes that a fairly large organization would
MS. AULD: It is intended to cover about 95 percent. All of the information
that we are getting shows that if you cover about 95 percent, then you have
most everything. And the rest are just outliers. And there isn’t enough
evidence to show that it would be worthwhile to do them.
DR. GREEN: I want to refer to what I heard Paul Tang say this morning about
lemonade and that he revised this afternoon, and possibly get myself into a
little trouble here. When I look at Marjorie’s history to 93, which she quickly
extended to the seventies, what are we doing here?
Pretend for a minute that I am a physician. Now I am an old physician. I
would ask you to just trust me. Diabetes isn’t what it used to be. Hepatitis is
not what it used to be. There are a hell of a lot of things that aren’t what
they used to be, and it doesn’t hold a candle to how it is going to change as
the genetic stuff does the association work.
And it is my personal belief — it is just a belief — that an international
classification of diseases prepared to characterize the causes of death may not
hold up. It is so much fun to be able to talk with someone about
classification, like the folks at this table, because most people have no clue
what we are talking about.
But all classifications, by definition, are arbitrary. All of them. There is
no exception, as far as I can determine. And the International Classification
of Diseases is arbitrary. And what we do is, we organize our concepts in terms
and words, and then classifications are about ordering principles that let us
feel like we understand what the hell we are doing, and what is going on.
And then sometimes in some countries we use them to adjudicate claims. I
think this — going back to Paul and the word lemonade — I just want to
confess despair almost in hearing that the pause button is being hit, we are
not going to move toward — I have no confidence we are going to move toward
ICD-10 any time soon.
I really don’t like practicing medicine absent 60,000 additional concepts or
so, or maybe it is 35,000 or so. Who cares whether it is 35,000 that we are
missing or 60,000 when you get right down to it. What you are interested in is
taking care of patients.
So when I look at this mapping of SNOMED, it is sort of like the alphabet
and the words. That looks to me as if we are getting quite close to being able
to write whatever patient novel we want to, with this terminology. And with
computing capacity that already exists, I just wonder if it is time to
I just was confessing that I am frustrated, and my question really is, is it
time to reconsider what the future of ICD is, and how we get to it.
DR. WALKER: I agree with you. I was talking with Betsy Humphreys, who leads
all of this at NLM, yesterday. And I said it seems to me that what an adaptive
organization would do, would be translate from ICD-9 to SNOMED, which actually
is a clinical vocabulary, which is far better curated, which has the ability to
add new concepts in a way that doesn’t destroy the old system. And only
translate what you have to translate into 10 to drop the bill. And she said
So I think that is the thing to keep in mind, that we have unfortunately
gotten ourselves fixated on 10, and in meaningful use it is not the end game.
The end game is SNOMED. And ICD-10 will not be acceptable for quality measures
in MU3. Correct? It is just SNOMED.
So that the thing to do for all of us, and somehow to communicate it to the
industry, is that SNOMED is the game. And SNOMED, while all models are flawed,
it is far and away the best model there is. And just regard 10, and then when
11 replaces 10, if we do it right, then it is just a matter of saying, okay, so
we have got to translate these 15,000 SNOMED codes into 11 now.
And it becomes work, and it will be a pain in the neck. But it doesn’t
become the next cataclysm. So I think you are right, but there actually is a
solution that is right there for us to use.
DR. WARREN: Just to make you feel real good, Larry, they are already working
on 11. Fortunately the Chair of that workgroup is Chris Chute, who is at Mayo.
They are re-doing the entire classification structure to come more in alignment
with SNOMED. Other work that is going on are, there are classification systems
because people are addressing how are we going to bring in the genetic detail
that we need into the EHRs. Chris is involved in some of that work as well.
Plus, work on looking at phenotype classifications and things like that.
So as our knowledge expands, we have to grapple with how are we going to
deal with this in our EHRs, and provide that information at the point of care,
where the decision needs to be made, in a way that the old guys can apprehend
it and make good decisions. Or the new guys. I worry about the new guys more
than that because I am around med students.
MS. GREENBERG: I could either say something, or I could just retire today.
But those of you who know me, know that I am not going to take the latter
approach. So let me first of all thank NLM. Last night I received a very
exciting e-mail from Betsy about the progress they have made in the mapping,
between SNOMED to 10-CM. And not only the mapping but an electronic tool. And I
was playing with it at some ungodly hour.
So I think the future to some degree, when we have data in SNOMED, to be
able to map it to 10-CM, is here. I think to spend, personally, since everyone
is worried about using our time wisely, to spend a lot of time on the mapping
to 9-CM I think is a very poor use of resources. 9-CM was created originally in
1979. It is hopelessly outdated in many areas.
We have a handful of people in this country really working on these areas.
That may be part of the problem, as Judy said. What are our best practices? But
compared to much smaller countries we have just a handful of people, including
at NLM. They had to scratch and dig in order to get the resources to do what
they have done.
And I think the committee was helpful with that the last time that Betsy was
here, and she said there was no funding for this activity. And I am very
pleased that now there is. But I defer to you as to what recommendations you
want to make.
But I do feel that investing a lot of time into a mapping of SNOMED, a
highly granular terminology, to a much less granular and outdated version of
ICD, is a very poor choice of resources. So that would speak to probably not
staying with ICD-9-CM forever. But again, I defer to you on your recommendation
in that regard.
There are a lot of uses for these classifications in epidemiology, in
research, in health care statistics. IT is not just a question of dropping a
bill, although I realize that many people like to see this. It is just a
billing code. I think Ob could probably tell me that you make a lot of use of
coded data beyond just making a payment.
And I think that in the future, if we have fully coded SNOMED data from
electronic health records, then this type of mapping — and improved mapping,
and certainly with the next generation of ICD it should be improved as well —
that will be the solution. But, it is not just a question of the bill.
And right now, there are absolutely no — not even best practices, but there
is no roadmap for researchers, epidemiologists, statisticians, et cetera,
federal statistical agencies, to get data from electronic health records. It is
all a whiff. And so I really hope you will keep that in mind before you suggest
just tossing the 100 and some years of work with ICD out the window.
I could say more but I will stop there. But I do think that IHTSDO has
agreed that there is a role for ICD, and ICD, WHO has agreed that there is a
role for SNOMED, and that the two of them have agreed to work together and are
doing that, and you have just heard a demonstration of that.
And that is where we should be going, rather than just assuming that there
isn’t a role for an international classification, which is used now, by 120 or
more countries. Right now we can’t even compare our data to those countries,
except for mortality.
DR. CARR: I want to just clarify, because I was playing with that I-MAGIC
thing. Isn’t that the presentation layer? You write in whatever you want, and
then it maps behind the scenes.
MS. GREENBERG: Do we even need a presentation layer to get to SNOMED?
DR. CARR: That is what I am saying.
MS. GREENBERG: That’s what Jill is saying.
DR. CARR: Well that is what this I-MAGIC is. You just type in something, it
gives you a pick list, you pick it and you are done. And you can also map it to
MS. GREENBERG: Because these aren’t one to one, it does require some
intervention of an intelligent human.
DR. CARR: I am just saying I as a physician was able to do headache, find
migraine, and then find the ICD-10 code for that. It took me a second.
MS. GREENBERG: Physicians shouldn’t have to use it.
DR. WALKER: It says you need more information here. So it does that part
also. If it doesn’t have enough information, it prompts for more.
DR. TANG: Just a question. The words that show up in I-MAGIC, those are
actually UMLS concepts, correct, because they are human real? So I suspect they
are really at the conscious level, which is how human doctors deal with things.
And then it gets turned into a SNOMED map that goes on to ICD-10.
DR. CARR: If we can pull it off to just demonstrate it, I think it will be
DR. SUAREZ: From a standards perspective, Jim, I don’t know if that is
right, but you mentioned that the quality measures are not going to be thinking
ICD any more. I think the standard is that, starting with Meaningful Use II,
SNOMED is the only vocabulary for coding problem lists. But NQF and the quality
measures are going to continue, except on using both SNOMED and ICD. And they
are going to be dropping ICD-9.
DR. WALKER: Yes, ICD-9 drops, but ICD-10 does at some point.
DR. SUAREZ: I haven’t seen that. It could be that later on they drop it.
DR. WALKER: The point is that all of the people who are experts, and Betsy
and others, regard SNOMED as the CORE, as far more clinical granularity, which
will extend to research and mortality also. That is why 11 is being aligned
with SNOMED. So that in effect there will be one code, increasingly, as 11 is
designed to mesh with SNOMED.
DR. SUAREZ: Sure. So one of the realities is that in this country we are
going to continue to depend on it. And we discussed this earlier. You at the
beginning of this entire meeting described the major data initiative that HHS,
have all the various strategies, a lot of them depend on something called
Now we have been trying to stay ahead and beyond claims. But at the end of
the day they are going to depend on encounter data. Encounter data is going to
be reported electronically. The electronic standard to report that encounter
data, today is the standard that has been used to submit claims. It is the same
And that standard does not support the submission of SNOMED codes in the
vocabulary. It supports only the submission of ICD. In fact, the purpose of
moving to 5010, is to really be able to handle ICD-10 rather than ICD-9.
So I think the reality is that we are going to have to handle the billing
aspects, and the administrative aspects of health care, using ICD. I do feel
that because now the push of ICD-10 to who knows when, there is a value in
trying to create a map from SNOMED CT to ICD-9-CM, particularly because now we
are going to stick with ICD-9-CM for a longer while, and EHRs are going to
begin moving to SNOMED.
So now they are going to have to go from SNOMED to ICD-9, to be able to bill
in today’s environment and for the next probably four years or who knows how
long more — five years, maybe, depending on how long the delay is on ICD-10.
What I really wanted to ask is, Vivian, on your slide number two, the map
issue number two, you mentioned that the pilots are — one of the pilots is the
NLM pilot, to map SNOMED CT to ICD-10-CM. And you mentioned that the purposes
and the benefits are really — you put it in billing/statistics. Right? But I
haven’t seen really the piloting of this for billing purposes.
I think a lot of the piloting and the testing of this mapping is being done,
and significantly so, for statistical analysis purposes. But my question, and I
guess the interest, and in fact the interest because of the Standards
subcommittees, now we are going to be looking at, are we piloting, are we
testing this? Are we making sure that this is working?
Would it be possible, and can it be expected, that some of the testing will
involve actually not just providers trying to pull out the SNOMED data and to
see ICD-10 and then reporting it to be used for statistical analysis. But
actually involving payers to allow the testing of the actual billing part of
it, is that part of the scope? Or is that beyond the scope of this pilot that
NLM is doing?
MS. AULD: Yes and no. Essentially this map was just posted yesterday. And
all of our resources have been in just producing the map, producing this
preview version and getting the final version out by June. So we don’t have in
place specific plans for pilot testing it. It is something that we have to go
back and look at. And we are going to do it in conjunction with ONC and NCHS
and others. But we don’t have specific plans in place.
DR. SUAREZ: I would really encourage the process, you and all of us,
basically to try to identify that and be able to show how this is to be done,
or can be done, or should be done for billing purposes. And I think the idea of
ultimately creating a template standard for being able to map SNOMED CT to
ICD-10-CM for billing purposes is going to be probably one of the most valuable
things we can see into the future.
DR. TANG: So I saw it for the first time. What I was surprised at, you can
even type in abbreviations, like hypertension, and it will come up with — or
heart failure. It’s a little hard to do on this. So CHF comes up with a number
of things. And what is interesting — I do think this is mapped in UMLS to the
concept. Concepts are written by health professionals. So we can relate to
And then you select one and get the ICD code. And what happens — I’m making
this up but this is what is happening — it just maps to the SNOMED term that
is mapped to that concept, and then mapped over to ICD-10-CM. And that is what
you get. I can deal with it.
DR. CARR: Exactly. So who invented this I-MAGIC?
DR. CARR: This solves the problem that we are dealing with. We have had a
lot of conversation of many different aspects. But clinicians want to write
words that they are familiar with. ICD people who need population things may
need ICD-10. EMRs need SNOMED.
But if you as a physician or clinician can just write the words you know,
and then choose from six, that is it. We are done.
MS. AULD: Keep in mind that the data that is in here is the preview map that
we just released yesterday. So there will be much more information come June,
when we have the full map available.
MR. SOONTHORNSIMA: Did you say that the mapping is done at 90 percent?
MS. AULD: I think it is 95 percent, yes. To 10-CM.
DR. TANG: But the main thing is, this is the map that is going to be
maintained by the federal government. So it has longevity.
DR. CARR: It is standardized, yes.
DR. SUAREZ: Maybe part of the issue is really, so this tool needs to now be
somehow incorporating to the EHR itself, to be able to deal with that. Maybe
that should be a requirement of a state’s EHR in the next meaningful use page
two, is we are creating capability. The EHR should be capable of doing this or
doing that. They should be able to incorporate a translator from SNOMED to
MS. GREENBERG: I have been saying this for years.
DR. CARR: This solves everything.
MS. GREENBERG: I think the intent is that this would be maintained by the
DR. CARR: It is standardized. It gets rid of these proprietary things such
as certain places that make presentation layers, we all have the same
presentation layer. It can map to ICD-10, yes.
DR. SUAREZ: This is very much — and I know you mentioned it — building on
the thing that Kaiser Permanente had with their Convergent Medical Terminology,
which is a tool that maps backwards and forwards from different vocabulary
standards, even SNOMED-LOINC is even involved in that. So yes, I think this is
the kind of example of tools that demystifies the complexities and the issues
around cross-walking SNOMED to ICD.
MR. SOONTHORNSIMA: So the elephant in the room perhaps is this ICD-10 date.
What is the new — the date has shifted from October 2013.
DR. CARR: The question is what is the reason for delaying it, if this is the
reason for delaying it.
MR. SOONTHORNSIMA: This is one. I am sure this is not the only one, but this
is a major barrier, I think we can begin to explore further to see whether we
can check this one off the list, and then tackle other barriers. Because I
think the sooner we delay, this start and stop, it is costing the industry a
significant amount of effort and dollars.
DR. CARR: So a major voice was the physician voice.
DR. TANG: So there still has to be the reconfiguration, essentially the
retooling of the systems themselves. That has to be done. But the main thing is
to take advantage of this delay, because in some sense people do have to mind
shift. They are on the track, October 2013. And let’s mind shift in the
direction of the future, which is, let’s coat it in clinical terminology. I
think what we have been asking for is this kind of thing, in fact, maintained
by the federal government.
There are these proprietary systems that are actually distracting. But
NCVHS’s role is to mind shift to the future direction. That is a huge thing we
can do, I think. Right now they are preoccupied with how to deal with the shift
in actually 10, but it just happens to give us a pause to make some lemonade
DR. SUAREZ: I do think we have a very critical role, and it is a very time
constrained role. The Standards subcommittee normally holds a hearing to
evaluate the progress that the industry is making in terms of adopting these
new standards. So we were going to, at least finding a hearing around June on
the status of 5010, and then the ICD-10. And every year we have that.
We might need to consider, based on the timeline that CMS is working on in
terms of any regulations that they need to write, to modify the date of ICD-10.
But I think this particular opportunity will give us a chance to engage the
industry as a whole, and understand not only how these tools can help but also,
what are all the other issues around compliance and transitioning and
And then be able to formulate a series of recommendations that might help
the regulator to frame the timeline by which this movement in terms of the date
could happen. And then, what would need to happen in order to make sure that
the timeline is met, and what are the critical steps that you need to take —
including this testing, and including the incorporation of these type of
elements into electronic health record capabilities.
MS. DOO: I think assuming we do have the hearing on the update, it probably
behooves us or the committee to have some really specific questions about what
exactly is being done with these simulations, with the labs, with crosswalks
So that they don’t come in and just say — and this is no offense to anyone
— just we are working on it, we are X percent with testing. I think we are
DR. CARR: What was concerning at the last hearing we heard, were how many of
these crosswalks were being developed.
DR. SUAREZ: And the reliability of some of the vendor-based crosswalks, the
formula, the black box, kind of rule behind it that didn’t really —
DR. CARR: I think very important is for us to understand what were the
things that caused the Secretary to stop? If this teaching doctors ICD-10, was
one — obviously that was one, this takes care of that. What were the other
DR. SUAREZ: I am sure each of us probably have our own opinions about what
happened. My own perspective was clearly a particular constituency, in this
case the providers represented by the American Medical Association, were
raising clearly important concerns about the cost, and where is the benefit for
DR. CARR: The cost of?
DR. SUAREZ: The cost of doing the transition, of modifying all their systems
to be able to pull out and produce an ICD-10 compliant bill.
DR. WALKER: This is work, and the vast majority of small care delivery
organizations were not going to make, weren’t even started. I think 40 percent,
some staggering percentage, hadn’t even started. I think the other thing that
would be interesting to know is if CMS thought that they would be ready to
receive ICD-10 bills October 1st. I think it is a combination of those two.
I have talked to insurance executives, private insurance executives, and
they tell me that they all just regard it as a certainty, even if we put it
back a year. But particularly October 2013, there would be organizations that
did not make the deadline, and private insurers would have to be able to accept
9 and 10 for some indefinitely long period of time. They just regard that as
reality, and are just preparing for it.
So it is very hard. This is work that most organizations don’t have
resources for. It is not just that it is conceptually hard. It is just hard
hard. It is work.
DR. SUAREZ: This was all documented in our letter to the Secretary last
year, out of the hearing. Because in the hearing we heard exactly that.
Medicaid, there were 40 or — I know it was a very high number where what they
call, high risk of not meeting ICD-10. And we documented that, and the
principle responding out of that, the Secretary should — and this is a year
ago, so when we were writing that we were two years ahead of the supposed
deadline for compliance.
We said, that is why it is important to ensure that the deadline is
maintained but now that we need to make a lot more emphasis on transitioning
and working toward that, and all that. Clearly as Jim pointed out, nobody was
really thinking about it.
MR. J. SCANLON: And it isn’t just ICD-10. If it were only that it would be
arguable. But I think it is the perception among the providers that with health
reform, with all of the incentives and penalties with the accountable care
organizations, with all the things that are underway that ultimately impact the
physician’s office, when you put all of them together it was a very daunting
It wasn’t so much that ICD was worse than this or that. ICD in itself was
likened to a Y2K situation for physicians’ offices. But it wasn’t just that ICD
was tough and everything else was easy. I think the cumulative perception is
that with all of these things, can’t we just slow down a little bit. And ICD
was the one where, for whatever reason —
But remember, this is a final rule. We can’t just announce that we are
delaying. We have to go through a formal process of rulemaking, just as we did
to establish debate. So within the weeks ahead, HHS will be issuing a proposed
And all of the comments for and against, and the nature of the delay and the
amount of the delay, and all these other factors, could be brought into the
public common process and would have to be weighed, would have to be weighed in
the ultimate decision.
Then a final rule would be issued that would go whatever way the policy was
taking us. So it is not an instantaneous sort of thing. But it was cumulative.
I think all of you were following that. This was not ICD-10 alone. This was —
you name them.
If you lined up matrices of all of the health initiatives impacting doctors
and physicians and health insurance plans and so on, it was daunting. And I
think people were just saying, well how can we sort through this.
DR. WARREN: So the question I have is, when we held hearings on this back in
2007 and ’08, I remember sitting in the room and we asked all these things. So
what do we need to put into place to make this easy for adoption. And we
recommended every single thing that we were told.
And I brought up the fact, don’t always consider just the clinicians, think
about the faculty, think about the curricula, let’s start the pipeline early
and bring them forward. And one of the things that we heard was, once the date
is set do not change it.
And the reason for that was, all the other HIPAA standards, everybody knew
that those deadlines would drift if they even hiccupped or said it was hard.
And they said if we want to go to this in future, hang tough and be in there.
Part of me is saying, we did everything we were told, everything that people
wanted, we are hearing the same comments, there are no new comments coming
forward with why this is hard. And yet we are drifting again. So what does that
do as we start trying to look at how do we plan the future, how do we migrate
towards new health care reform, if everybody knows we are going to come right
up to the deadline and then go, oh, excuse me, we will extend it — how will we
ever get to the preferred future that Paul keeps saying.
DR. WALKER: It is not true that people weren’t told. Several of us on the
Standards Committee said very early in its life — I don’t remember when —
that we should create a project map for a 100-bed hospital or a five-person
DR. WARREN: We did.
DR. WALKER: I have never seen it.
DR. WARREN: AHIMA had one. HIMSS had one.
DR. WALKER: But the point is, if you looked at the project map and you hired
a project manager, they would commit suicide. And we have never created a
project map that made sense in terms of the resources of a 100-bed hospital or
a five-person practice.
And you take 5010 and ICD-10 and Meaningful Use I and II, and all of the
other things and say, okay, what would a really high performance project team,
how long would it take them. And it doesn’t make a rational project map. It
DR. WARREN: So the problem that you bring in that we did not know about when
we did all that, we brought in industry partners that had the expertise to
create these timelines, to create the projected maps that were all over the
internet. Workshops were done. If you go back and look at the history, in the
early 2008s, 2009s, people were getting ready for the 2013 deadline.
Then we had ACA. And so we had lots of other deadlines. Then we have had
health care reform and we have had other things, and lots more deadlines. And I
think we are coming with, a lot of federal government is coming up with more
standards and more things.
So maybe it is back to Bill’s concern that he has been chastising us with,
because we have really got to take a look at what the bandwidth is. Maybe we
are way over the bandwidth, and we just need to slow everything down, including
Meaningful Use, until we can catch up. We can’t put the carpet and the stuff
on. And we are certainly not going to build the roads.
PARTICIPANT: We slowed Meaningful Use down a little bit anyway.
DR. FRANCIS: What is sad to me about this is not the question of whether it
gets slowed down, or whether you adhere to a date. But the perception, which I
am not sure how much reality there is behind it, that all of these demands are
separate, that there is no interrelationship between doing this.
So we have got to put all of our energy into meaningful use. And then we put
all of our energy into 5010, and then we put all of our energy into — I have
heard people actually say, we can’t do ICD-10 because we are doing something
else. And if there is any way, and I am not the techie type, but if there is
any way to figure out how to make it look coordinated in a real way.
DR. WALKER: But see, that is the problem. It doesn’t. If the date stayed the
same, we would make it, no question. But we have had a team working probably 10
FTE, something like that, working for probably 18 months. Meaningful use has
cost us around $750,000 a year, ever since it was first announced.
We have made all the dates, we will make all the rest of them. But the point
is that we are not the organization of interest. The problem is, you have got
to have a simultaneous 5010 team, ICD-10 team, Meaningful Use team, all going
at the same time. And that is why it is just completely beyond the capacity of
any remotely average organization.
MS. GREENBERG: I have a question that is — I don’t know whether Larry maybe
can answer it, but in line with what Leslie was saying about the fact that
these things should relate to each other. And then also, as Jim was saying and
others noted, that at least the latest cry for help, or for delay — although I
have to say that I think the letter is publicly available, the letter that came
from the physician community.
It actually did not ask for a delay. I think it asked for both Speaker
Boehner and the Secretary to stop the implementation. So that is what was
requested. That is what the letter said. It did not ask for a delay, it said to
But that letter and of course the trade press and everything, have quoted
this cost estimate that was actually, I believe, done for the American Academy
of Family Physicians, as well as maybe some other group as well — and I am
sure you have all seen this quote — of something like $87,000 for a three or
smaller physician office to implement.
And remember, they are only implementing 10-CM. They don’t actually
implement 10 PCS. That is only for hospitals. So it is actually, ironically, it
seems that it should be the easiest for them. But they have the fewest
resources, too. So I understand that.
But if you go find that study, which I did, half of that cost is because
they would have to improve their documentation to support 10-CM. And my
question is, if the documentation of these physician offices is not good enough
for 10-CM, how is it going to support SNOMED CT? Or the granularity of
electronic health records? Am I missing something here?
I think half of the cost is just to improve the documentation. This is a
bigger problem than 10-CM. I don’t understand how it is going to support the
granularity of electronic health records.
DR. WALKER: SNOMED is a well designed categorization. There are a number of
clinical prediction rules. If you want to know whether you should put a patient
with community acquired pneumonia in the ICU or send him home on PO meds, there
is a clinical prediction rule that has been validated to risk stratify patients
so you can do that intelligently, far better than any human can do it.
But one of the elements is neoplastic disease. Another is kidney disease.
When we code something like that in ICD-10, we would have 250 or 300 ICD-9
codes that we would put in for neoplastic disease, because you can’t say
neoplastic disease, and so forth.
So that one of the things is, you need to be able to say very high level,
general things like neoplastic disease, because that is the only thing that
works in that validated clinical prediction rule. And then there are other
times you need a specific type of glomerular nephritis.
The point is, with SNOMED you can do all of those things in a way that
doesn’t require putting 250 codes in by hand every time you want to say
diabetes or something. And so that is a lot of what is going on there.
MS. GREENBERG: Well I do think in the mapping, we should try to improve both
SNOMED and whatever we are mapping it with. ICD — there are things we learn on
both sides. But it seems like there is a flaw there in the decision support,
because it should then accept any neoplastic code. If you are saying that is
all you know —
DR. WALKER: But ICD-9 doesn’t support that. If it were SNOMED you could say
that, it’s any neoplastic code, and SNOMED would know all of the 500, that oh,
that’s a neoplastic disease, that counts. ICD-9 does not know that. You have to
tell it any one of these 500 individual things.
MS. GREENBERG: This is for another discussion, obviously. But I think this
is solvable. I do think we are dealing with a problem of documentation that is
PARTICIPANT: If they all had 500 IT people, they could solve it.
DR. TANG: I sort of would want to ground us in the realities. And it is not
a law. It is health reform. And the reform wasn’t made up by Congress. It is
the economic parasite on the rest of our country, like education in California
that has been decimated by the health care clause. That is the barrier. That is
the time we have to be. It is not high tech.
So that is sort of the different kind of reality we just have to do. And
there is no way to do this on paper. So that, I think, grounds us in why do we
have to do it, work so hard in such a short period of time.
The others, I guess the delay — we didn’t advise it, but it is water under
the bridge, and let’s just crank the lemonade out. And so the concrete piece
is, let’s find a process that says how are we going to form a recommendation
that may advise the Secretary, and she can decide whatever.
But at least she has input other than it just costs too much or it’s too
hard. Let’s just try to find some way to make the country better, after the
effort we already have to spend.
DR. CARR: Perfectly said. Thank you. We are going to adjourn for today, and
the subcommittees that are meeting are Population and Standards.
(Whereupon, the Full Committee adjourned at 3:40 p.m.)