[This Transcript is Unedited]
DEPARTMENT OF HEALTH AND HUMAN SERVICES
NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS
June 21, 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 – JustineCarr
- Updates from the Department – James Scanlon
- CMS – Robert Tagalicod, Denise Buenning
- ONC – Jodi Daniel
- Privacy – Joy Pritts
- Population Health — SES Workshop Letter – Larry Green, Sallie Milam
- Quality Letter — Steps to Improve Support for Consumers’ Decision Making – Paul Tang
- Privacy Subcommittee — Briefing from April 17-18 Hearing – Linda Kloss
- New Working Group on Data Access & Use — Plans and Status – Justine Carr, James Scanlon
- Standards Briefing on ACA Section 10109 – Walter Suarez
M O R N I N G S E S S I O N
DR. CARR: Welcome. I give you a very warm welcome to this very warm day here
in Washington, day one of the National Committee on Vital and Health
Statistics. I am Justine Carr, Steward Health Care, chair of the committee. No
MR. SCANLON: Good morning, everyone. I’m Jim Scanlon, executive staff
director for the full committee and Deputy Assistant Secretary for Planning and
Evaluation at HHS.
MS. MILAM: Good morning. Sallie Milam, with the West Virginia Health Care
Authority, member of the full committee, co-chair with Larry of Population
Health. No conflicts.
DR. GREEN: Larry Green, University of Colorado, member of the committee,
co-chair with Sallie. No conflicts.
DR. TANG: Paul Tang, Palo Alto Medical Foundation, member of the committee.
DR. COHEN: Bruce Cohen, Massachusetts Department of Public Health, member of
the full committee. No conflicts.
MR. BURKE: Jack Burke, Harvard Pilgrim Health Care in Boston, member of the
full committee, member of the Population Subcommittee, member of the Privacy
Subcommittee. No conflicts.
MR. TAGALICOD: Robert Tagalicod, director, Office of E-Health Standards and
Services at CMS and the CMS liaison.
MS. BUENNING: Denise Buenning, acting deputy director, Office of E-Health
Standards and Services, CMS.
MS. PRITTS: Joy Pritts, chief privacy officer, Office of the National
Coordinator for Health Information Technology, HHS.
MS. DOO: Lorraine Doo, Office of E-Health Standards and Services, and lead
staff to the Standards Subcommittee.
DR. SCANLON: Bill Scanlon, National Health Policy Forum, member of the
committee and member of the Standards Subcommittee. No conflict.
DR. MAYS: Vickie Mays, University of California Los Angeles, member of the
full committee and member of Population and Privacy. No conflict.
MS. KLOSS: Linda Kloss, health information management consultant, member of
the committee, co-chair of the Privacy Committee. No conflicts.
DR. WARREN: Judy Warren, University of Kansas School of Nursing, member of
the committee. No conflicts.
MR. SOONTHORNSIMA: Ob Soonthornsima, Blue Cross and Blue Shield of
Louisiana, member of the committee, co-chair of the Standards Committee. No
DR. SUAREZ: Good morning, everyone. My name is Walter Suarez. I’m a member
of the committee and co-chair of the Standards Subcommittee. No conflict.
MS. GREENBERG: Good morning. I’m Marjorie Greenberg, from the National
Center for Health Statistics, CDC, and executive secretary to the committee.
MS. SQUIRE: Marietta Squire, staff to the committee.
MS. JONES: Katherine Jones, CDC, National Center for Health Statistics, and
staff to the committee.
MS. KHAN: Hetty Khan, National Center for Health Statistics, CDC, staff to
the Subcommittee on Privacy.
MS. COOPER: Nicole Cooper, staff to the committee.
DR. LAZARUS: Steve Lazarus, Boundary Information Group, representing CAQH
MS. FOERSTER: Rachel Foerster, Boundary Information Group, representing CAQH
MS. QUEEN: Susan Queen, from ASPE, lead staff to the Subcommittee on
MR. QUINN: Matt Quinn, from NIST, lead staff to the Quality Subcommittee.
MS. JACKSON: Debbie Jackson, National Center for Health Statistics,
MS. KANAAN: Susan Kanaan, writer for the committee.
MS. THOMASHAUER: Robin Thomashauer, CAQH.
MS. LOHSE: Gwen Lohse, CAQH.
MS. THORNTON: Jeanette Thornton, America’s Health Insurance Plans.
MS. BANKS: Tammy Banks, American Medical Association.
MR. BIZZARO: Tom Bizzaro, First Databank.
MS. WEIKER: Margaret Weiker, Hewlett Packard, representing ASC X12.
MR. WANG: Derek Wang, Social Security Administration.
MR. DECARLO: Mike DeCarlo, Blue Cross and Blue Shield Association.
MR. CROWLEY: Kenyon Crowley, Center for Health Information & Decision
Systems, University of Maryland, Working Group on Data Access and Use.
DR. CARR: With that, I just want to make one opening comment.
We have been extremely busy lately, over the last couple of weeks. We had an
intervening meeting in May and generated a letter from that. I know we got
feedback from a number of sources about the concern that there’s a tradeoff.
Meeting the timeline has a tradeoff of foreshortening some of our usual
processes. We heard that loud and clear. I want to reiterate our commitment to
collaboration and transparency. We’re going to take time this afternoon — I
think it’s identified as talking about the claims attachment, but I think we’re
going to broaden it to talk about, in general, how we manage the competing
demands in the available time and how we prioritize. So we’ll have some
discussion this afternoon.
We’ll also be teeing up an Executive Subcommittee retreat for a day one time
this summer to help us continue to think about how we meet our expectations and
maintain our tradition of openness, collaboration, transparency.
With that, let me turn it over to Jim.
MR. SCANLON: Thank you, Justine. Good morning, everyone.
Let me bring everyone up to date with a number of developments in the data
policy area at HHS. A number of exciting things are happening, particularly in
the area of our Health Data Initiative, which we started about two years ago.
But first I’ll plow a little bit over previous ground. In the area of policy
and planning, I think I mentioned previously that we updated our strategic plan
with four overall goals related to transforming health care, prevention, a
number of other areas. I think I have shared material with you. Maybe we’ll
just make it a standard part of the briefing book. Of course, we would be
willing to brief the full committee on any of those, on the strategic plan
generally for HHS.
In addition, we have strategic plans in specific areas. Let me just mention
a few of them:
- Tobacco control is one of them, obviously a very large area.
- HIV prevention and control.
- Health care-acquired infections — a big area where we’re trying to work on
some prevention issues.
- Health care disparities. We have a very extensive plan on reducing racial
and ethnic minorities in health.
- Of course, you are familiar with the National Quality Strategy and the
National Prevention Strategy.
In addition, we have initiatives related to global health, on the human
services side, early childhood development, as well as health, and two other
more recent ones, an Alzheimer’s plan and a strategic plan for persons with
As I said previously, each of the plans, as a general rule, has to have some
goals and has to have some action steps, something you can actually observe
that you are doing, and it has to have some measures that indicate how close or
how far you are from reaching those goals. So we generally try to have some
data and measures as part of each strategy.
Again, we would be happy to brief the full committee or any of the
subcommittees on any one of these specific areas as well.
I’m going to talk a little bit now about our data standards from the ACA,
our Health System Measurement Project, which we launched about a month ago, a
data strategy, which the Data Council was asked to pull together — I’ll
provide some of the highlights there — and then some developments relating to
our overall Health Data Initiative, the goal of which, as you will remember, is
to make the data that we do have in HHS available more broadly to communities,
to patients, to providers, plans — to everyone — for choice and for decision
Just to remind everyone, one of the sections of the Affordable Care Act —
and I would remind everybody that the Secretary has the authority to do this
anyway — the Act called for the Secretary to adopt and promulgate
data-collection standards for race, ethnicity, sex, language, and disability.
In essence, these are demographic data items. These were to be standards that
would apply when someone is asked to self-identify.
As you’ll remember, we went through a process of looking at what standards
are used in HHS and other federal surveys now, what is the best practice, what
seems to work. We put together some potential models. We took public comment.
Then, back in the fall, the Secretary adopted our recommendations, which were
developed by the HHS Data Council.
Those standards, which are now available on the HHS website — we now have
data-collection standards for HHS surveys, the surveys that we sponsor relating
to those items.
As you all know, we were asked to look, as potential areas for the future,
at the whole area of SES, socioeconomic measures, position, and status. We
asked the committee here to take a look at — sort of an environmental scan in
terms of what the best thinking and the best practice there now is. I know the
committee is working on a letter summarizing the results from the workshop.
I shouldn’t mention the word “workshop” anymore. We’re forbidden
to have conferences or workshops or to send people to international meetings.
So we’re looking forward to that.
At the same time, though, HHS leadership asked us to look at those same data
items — sex, race, ethnicity, language, and disability — for data standards
that would apply to administrative data. Remember, with surveys and research,
you have a fair amount of flexibility, because you control the content of the
interview or whatever it might be, and you generally have the person in front
of you or some knowledge about the person in front of you to get the
information. With administrative data, there are many fewer degrees of freedom.
No one is interested in adding a lot items to any administrative forms or
processes. So here we are looking at what would make sense from current
practice, what’s being done. First we look at HIPAA, obviously. We look at HL7.
We look at where there might be standards in these areas already. We look at
standards in terms of the way in which our agencies are already collecting this
kind of data in their own administrative data systems — so Medicare claims,
Medicaid claims, Ryan White claims data, things like that, surveillance data as
Surveillance data sort of falls between the survey research side and the
administrative side. It just depends on what exactly the focus is.
At any rate, we’ll be looking at whether there are standards for these data
items that we can recommend to the Secretary to standardize these basically
demographic items across our administrative data systems as well.
I think I mentioned previously that we’re trying to make some headway on
getting better data on vulnerable populations. One of the areas is the LGBT
population. We already have under way at the National Center for Health
Statistics a plan to — and we’re testing these now, in the third phase of
testing — ask questions about sexual orientation in the National Health
Interview Survey. If this works out, it would be included in the January cycle
of the survey. It’s something that a lot of folks have been asking HHS to do
for quite a while.
Let me turn to our Health System Measurement Project. This I think I gave
you previously. The project is now available on the HHS website. In this case
the focus is ten domains or content areas that folks would want to look at in
terms of health systems change, not just health reform, but looking at system
performance and change in the US health system over time. We have ten overall
content domains, which I’ll read to you quickly, and then we have about five to
ten indicators — these were consensus indicators in HHS — in each of those
domains. You would be looking at changes every year or every six months in
terms of how the US health system is moving.
The domain areas — and I can give you more detail. We can brief the
committee. In fact, it’s available on our ASPE website.
- One of the areas is access to care. There are a number of measures of
access to care.
- Cost and affordability.
- Coverage, insurance coverage.
- Quality of care.
- Vulnerable populations.
- Workforce issues.
- Innovation. We’re trying to get some measures of innovation in the
- Population health.
- Health information technology.
For each of those areas, we had a consensus process to try to develop
indicators. We will look at them as they change, probably annually, but in some
cases — health insurance coverage, for example — the information becomes
available semiannually, and we’ll be looking at those as well.
Again, I’ll give the URL for the initiative. It’s available.
We’re managing it out of our office in ASPE now. We’re going to let it go
for a while with these measures, but we will be looking down the road to what
additional measures people would like to see and whether there are other
domains. We finally converted this from an internal system to an
outward-looking system, and I think there’s a lot of interest now.
This is not a policy website, by the way. We’re not trying to explain
changes. We’re just trying to describe the changes. When folks are looking for
the major indicators, it’s not quite as nice as the economic indicators. But
when you are looking for indicators, at least there is one place to start.
For each of the data sources, there is a link to the data source, the
questionnaires, the original data. You can call up maps. You can call up
various breakdowns — race, ethnicity, and so on. If we have the data, we have
made it available that way, so you can build your own tables and build your own
maps — again, part of an initiative to get our data out.
We also put together a separate guide, available on the HHS website, to our
HHS surveys and major data systems, probably two dozen major data systems. We
have more individual systems, but these are the ones that folks normally look
to, major surveys, the Medicare data, the Medicaid data, things like that. It’s
a data source that we have already been making available publicly. We have
descriptions. We have, in general, the methodology, the questionnaires, and
related information about that.
Let me say a couple of words about our HHS data strategy.
Leadership in HHS asked the Data Council to look at the portfolio of all of
the major data systems in HHS and to take a look critically and to look at the
obvious things, like efficiencies and coordination and so on — but to look at
how well we’re doing with meeting gaps that everyone has identified, including
future data gaps, and to also look at whether we are benefiting from the most
recent technology in these areas in terms of data-collection strategy and so
on. Are there techniques, are there practices and resources that are available
that HHS could take advantage of to speed the data up, improve the quality, and
all that in terms of cost savings as well? I’ll describe a little bit of what
we found and recommended there.
The other area was to ask, how can we align and integrate — at least align
— these various data resources, forces, and streams as we look forward?
Remember, at HIPAA we call it convergence. Now we’re referring to it as
alignment. Basically, it’s the idea that we have our population surveys and
surveillance, we have what is looking like less and less administrative data
supporting health insurance and so on, and then we have the potential of EHRs
adopted on a wide scale. Again, this is beyond the islands of excellence now.
This really would have to reach a fairly critical and broad-scale level.
As long as the thought is standardized data where quality can be ensured,
how do we bring these forces together? Are there ways that we could begin, not
to wait until it happens and then say, what shall we do, but to take steps now
in terms of pilot studies, data standards, various policies and principles, so
that we’re converging and aligning these systems so that as they do become
available, they are producing data not only for their initial purpose, but for
data secondary uses as well. Can we assure that they are more than just the sum
of their parts and we’re not conflicting?
We have a number of projects that the Data Council and others recommended
there. The nature of those is to assume certain things — what would you have
to do, for example, for the provider surveys in HHS to have them be able to
rely more systematically on EHRs? You couldn’t do that now. It’s just not in
large enough scale, and I’m not sure what you would ask to be reported out.
Secondly, for public health reporting, for example, what are the pilot
studies you have to do to try it out and actually work? Could you find some
health plans that have EHRs that would be willing to — where we could do some
pilot testing — and this is sort of where we are headed — to look at what the
plans now contain in EHRs and to what extent they meet the requirements of the
surveys in public health and research as well? So we’re heading those lines as
On the research and data side, it actually was very interesting. As a result
of the look we had, we’re going to be testing out the possibility of some Web
surveys and Web panels as augmentation for some of our surveys. Even for the
Health Interview Survey, we’re working now with the National Center for Health
Statistics on the possibility of a Web version of the Health Interview Survey.
Again, these have to be probability-based. Otherwise, it’s really not that
useful. You have to assure that it’s a probability sample, number one, before
you make estimates. But then it becomes quite a nice tool for quick turnaround
and to answer questions almost on an on-demand basis.
There are some examples in the industry. The worry is that they are not
probability-based and they have very low response rates. Again, you don’t know
what you have at the end of it.
It’s one thing for a company to do this. It’s different if you have to do it
for official health statistics, because then it’s not just a commercial
We also have several initiatives under way to speed up the release of HHS
data. Almost every one of our major surveys now has an early-release program.
For the Health Interview Survey, as I think you know, we have the early-release
program. We are releasing six-month data on a more or less quarterly basis, on
about half a dozen indicators, including health insurance coverage. The
provider surveys we have now moved up to where we have cut the time from field
data collection to what used to be more than a year and now it’s less than a
year. We have asked almost every one of our surveys — MEPS and others — to
try to speed this up.
On the administrative data side as well, there are several initiatives with
the Medicaid program to, rather than wait for the last state to get their last
records in before you start releasing the data, start making releases earlier.
There are several versions now of the MAX, Mini-MAX and Beta-MAX, that provide
this data on a faster basis as it comes in. You can often do a fair amount of
analysis. All you do is cut it off at that point. You don’t have to wait.
So very nice demonstrations under way.
We were asked to look also at commercial data resources and nontraditional
data sources. Many of our agencies purchase data from vendors who have
developed data sources over the years. IMS America, for example, has a lot of
data on retail pharmacy claims. Thomson Reuters and other places have all-payer
claims data. Nielsen has data on consumer behavior and so on. Many of our
agencies have access to this.
We’re trying to get a sense of what agencies have, what they are using it
for, and whether there is some way we can share the resources there. So we’ll
be looking at that and making some recommendations.
At any rate, I think everyone is excited about looking forward now on
applying some of the latest technology, Web surveys and so on, to our surveys
and data collection.
Just a few words on integration and alignment. Obviously, much of this
starts with data standards. We’re trying to get some principles where people
agree to use data standards. If you are developing a new data system, we really
just insist that you turn to well-performing data standards and not develop a
one-off kind of approach each time.
Administrative and claims data: As I mentioned previously, we have good
resources with Medicare and Medicaid. We in ASPE have a project under way where
we’re trying to pull together a resource for all-payer claims. This would be
everybody other than Medicare and Medicaid, which has always been difficult.
There are companies that focus on this based on their clients’ data, but it’s
hard to get access to a large enough and representative enough sample. It turns
out to be much more difficult than anyone imagine.
I remember we had meetings and everyone said, oh, yes, the Society of
Actuaries will be happy to do that for you. It turns out — and it’s not a
technical issue. It’s just very difficult in terms of ownership,
standardization, what the model is for pulling this information together, for
what purposes. Now it turns out that the industry is actually — the Blues, for
example, did have an initiative where they pulled all together their data.
There are some other places that are pulling data as well. But it’s very
tricky, and you have to decide what sort of model would really work best.
We have a number of pilots relating to electronic health records, public
health, survey data, and so on. We’re going to try to pull much of this
We’ll be spending more time on this today, but remember about two years ago,
with our open government plan, we created an emphasis — we called it our
Health Data Initiative, but the initiative is really the part of emphasizing
making the data available in an easier form, in an easier, accessible form, and
advertising the data that we have. I’m not sure we have more data. What we
wanted to do was get the data that we did have available. We do have a few
There are a number of initiatives here that fall under that rubric. Several
of you, I think, attended the Health Data Initiative Forum in early June,
called The Health Datapalooza, which we cosponsor along with some others. At
any rate, the idea there is to bring innovators and developers together with
our data experts to make our data more available and to promote useful,
innovative applications of our HHS data. There were a number of new
applications there. I think Todd, when he is here tomorrow, will talk a bit
more about that.
We have updated and expanded our Health Indicators Warehouse. This is where
we put our HHS indicator-type data. We have probably 1,100 indicators there.
Anyone who wants to use that data, it’s available for them in machine-readable
form for applications.
As I indicated, we just a month ago updated our HHS open government plan
again with an emphasis on data transparency and availability.
Finally, we’ll be starting today a working group under the auspices of the
full committee here on promoting data access and use. We have the first meeting
of that workgroup today.
Let me stop there and see if there are questions.
DR. COHEN: A couple questions. Very exciting stuff. Thanks, Jim, for
Are there timelines that the Secretary has for adopting standards around SES
data for administrative claims and surveillance systems?
That’s my first question.
MR. SCANLON: The first emphasis, the way the statute was written, was on
surveys. There was no statutory deadline, but it was a very high priority. We
really had those standards ready within about a 90-day period. But it took a
while to go through the public comment process and the review process. But
there was no date in the statute.
On additional SES items for surveys and research, that was the area where,
when we sought public comment, since the Secretary had the authority to adopt
standards in additional areas, we asked, what would be the most important area
you would recommend? Clearly, by far, most people were interested in SES. I
think it was more in terms of socioeconomic determinants of health. It wasn’t
so much demographic data in health. I think it was more social determinants.
Nevertheless, they were measures of SES.
We’ll take a look at what the committee is recommending and found. We can
move as quickly as the state of the science allows.
On demographic data items, these six items for administrative data, we knew
when we focused on surveys that there was an equal interest in the
administrative data. This is part of alignment and convergence as well. We
formed a workgroup, and we’re now meeting every other week. Within HHS, we’re
looking at what standards exist that we might consider — all the X12 and HL7
and other standards. We’re looking at recommendations from the IOM. We’re
looking at what kinds of data our administrative data systems are collecting
now. That’s probably a two-month or more project, to at least get the
DR. COHEN: It would be great to get the committee involved in those
discussions. I know there’s a lot of confusion at the state and local level
around these HHS standards that have already been adopted. People aren’t clear
how they apply and to what, although I know technically they are only targeting
federal survey data collection.
My other question: I know that NCHS is doing testing around GLB questions.
What’s the status of the listening sessions and progress around collecting
MR. SCANLON: There it’s tougher. On sexual orientation — this is the LGB
part of LGBT — we have actually moved along pretty well. NCHS is actually
testing those questions in their Health Interview Survey. That is the third
large field test that will be coming up shortly. If that works, we’ll be
including it in January.
On the transgender part of it, as Bruce indicated, we have had two listening
sessions with the research community and the community of experts in this area.
We have a couple of things going. We’re looking at the possibility of including
a question in the Behavioral Risk Factor Surveillance System.
But again, until these things are done, they are not — and in our
administrative data, we have a couple of systems that actually have been
collecting this already, different ways. We’re trying to see if there is a
preferred method there as well.
DR. CARR: I think what we are going to plan to do is take one or two more
questions and then move on to Rob. Walter and then Vickie. Then I think we’ll
go on to Rob.
DR. SUAREZ: A quick comment and a question. The comment is to the point
Bruce brought up about administrative data and standardization of SES. I think
it’s important to note that the standardization process for administrative data
that we have used in this country, at least through now, follows significantly
the standards-development organizations nationally and internationally, like
HL7 and X12. One would think that if we were to consider adding new data
elements related to socioeconomic indicators, that would be expected to be
captured in administrative data. When we saw administrative data, generally we
talk about claims and enrollment and all this data. Then it would have to go
down the process of standards-development organizations and be incorporated
into the actual transactions, which is a very established process.
That’s just a point of reference, I guess. Whereas on the survey side the
central provider of the survey can decide that we’re going to add these two
questions here and there, on the administrative side there’s a national open,
transparent process to do that.
MR. SCANLON: Frankly, in terms of the principles, the first thing you look
at is the X12 or the HL7 standards. Remember, these are for the data items, the
specific data items — race, ethnicity, sex, disability, and language. What
we’re finding is that in X12 there are some items there, and some of the areas
are just not covered well.
But that’s exactly right. We’re looking for what the industry has already
decided. We’re not interested in developing something entirely new.
DR. SUAREZ: The quick question is about –generally, the perspective
traditionally has been to aggregate data, to bring data together in sort of a
centralized mechanism and try to create these databases. Now the cloud is sort
of a new view of it. But there are new and innovative ways to do some of these
data connections. One of them has been developed through the work that the
S&I framework of the Office of the National Coordinator is doing, the
standards and interoperability framework. It’s called Query Health. In the
Query Health mode, instead of having a centralized database collecting all the
data and putting it into one place so that someone can do some analysis in it,
the idea is that you could just fire out queries to multiple sources and allow
the gathering of the data, particularly indicators, metrics, measures to be
analyzed, through a query.
I just wanted to point that out. I think the traditional way of thinking,
that we should bring together all the data in one place, is sort of breaking
MR. SCANLON: The focus on the standards is to get it standardized so that
you actually have something you can analyze. Again, whether it’s centralized or
whether it’s done locally — whatever the means of the query or the analysis is
— this is the measurement part of it. This is not the data-collection part of
it. But we have had in HHS for a long time now — it’s not just the ONC — we
have examples of query-based models. The FDA has a system, for example, with a
number of plans where they can send out a query to health plans. They can ask
about adverse effects associated with a specific drug or procedure or so on.
It’s not just a model. We actually have examples. It’s not as efficient and
it’s certainly not perfected yet. But it’s certainly a model that people are
DR. CARR: Vickie, quick question. We don’t want to shortchange CMS.
DR. MAYS: I think mine are pretty easy.
What comes under vulnerable populations, other than LGBT? Then in terms of
the collection of data on language, can we then look forward to some additional
languages, surveys actually being launched? Is there any innovation in terms of
the estimates of cell phone?
MR. SCANLON: Cell phones? Oh, yes. I’ll start with the last one. Every year
we collect data on who has cell phones in the Health Interview Survey. Every
year we sort of know who has computers, who has cell phones, as the basis for
additional survey research.
That’s how we are designing the Web survey, by the way. We’re looking at
findings from the Health Interview Survey about who actually has computers and
so on. We have always asked about various cell phones, so we can see that we’re
not getting biased results.
But the standard for language that we ended up developing, for surveys at
any rate — remember, these standards are a minimum — what we were encouraged
to do, for survey work, now — the issue seemed to be English proficiency, not
the individual language. On the other hand, for clinical work or administrative
data, you might need to know the specific language, so we added an expansion
where you ask about the specific languages. Spanish by far comes out —
nationally, it’s like 80 percent of the time.
For every state, we have the 10 or 20 top languages. We could put that on
DR. CARR: Thank you, Jim. That’s great. Very exciting things. I have to say,
the first time I knew you were venturing out into commercial was when we were
in the Humphrey Building and you said, do you want to see the command center?
We went in the command center, and it was CNN.
On to you, Rob.
DR. TAGALICOD: Thank you Justine. Thank you, Jim.
My name is Robert Tagalicod. I’m director for OESS at CMS. I’m here to give
the update at CMS. We have a few exciting things that we would like to report
Before I do, I want to apologize to ONC if I’m going to steal some of their
thunder. But we’re in this together. This is not simply a CMS endeavor. This is
both us and all of you folks.
On Tuesday, 6/19, CMS and ONC issued a press release. Essentially, over
100,000 health-care providers have been paid under Medicare and Medicaid EHR
incentive programs. Both CMS and ONC have been working hard together. Marilyn
Tavenner and Farzad Mostashari of ONC identified three months ago what that
target was for the end of calendar year 2012, which is 100,000 providers. We
are now beyond that.
We did look at trends. Farzad will know that I’m much more of a conservative
person. I like to look at trends and then do it very conservatively. I said,
we’ll get there by the end of the year. To my surprise — and I should have
yielded to him, because he’s much more free with the numbers and identified
that we would reach that in the middle of the year. He was right. So I would
like to say that we have done that, and we are even surpassing it. By the end
of May 2012, we actually gave $5.7 billion in payments. I think Jodi from ONC
can talk about the details.
We have limited time, so I’m not going to belabor that talking point. I’ll
yield to ONC.
I think we are well on our way. We are looking at the industry, the
so-called denominator of where we saw hospitals and where we saw ourselves in
terms of eligible providers. We’re essentially halfway there. Again, Jodi, I’m
sure, will go into the details.
Rather than focus on that, I’ll turn it over now to Denise to talk to you
about our front on admin simp and ICD-10 and 5010.
MS. BUENNING: Thank you.
Let’s start out with version 5010, since that is the one that’s coming up
sooner rather than later. I think you are all aware of some of the discussion
from yesterday regarding the challenges that the industry experienced with the
implementation of version 5010, so I won’t belabor those points. But I will
remind everyone that we did extend the enforcement discretion period for
version 5010. We did not change the compliance date, but we allowed some of the
noncompliant HIPAA transactions to go through until the end of March. In
response to industry need, we extended it again to the end of June. Basically,
what that meant was that providers, health plans, other HIPAA-covered entities
could transact noncompliant HIPAA transactions without fear of a penalty, under
our HIPAA enforcement process.
This allowed the providers and the vendors who were having difficulty doing
software installations or getting end-to-end testing completed additional time
to do so. It allowed us to keep the trains running, so to speak.
That enforcement discretion period is coming to an end on June
30th, at which point we will expect all HIPAA-covered entities to be
compliant and to be transacting in version 5010.
With regard to administration simplification, we have a number of
initiatives, as I think everybody knows, and some very challenging timelines
from Congress with regard to adoption of both standards and operating rules for
some of the transactions, as well as a health plan identifier. We talked a
little bit about this yesterday. With regard to the HPID, that was part of the
proposed rule that we published on April 17. It was also coupled with ICD-10,
as well as some changes to the national provider identifier. We have received
comments back. We will be preparing a final rule to address the comments that
we received. I can tell you, just in general, most of the comments regarding
HPID were clarifying in nature. They wanted some more information as to
examples of what enumeration would look like. So we will be addressing those in
a final rule that we hope to get out very soon.
Other administrative simplification areas we’re working on regarding
operating rules for the electronic funds transfer and remittance advice
standard. That actually was adopted in January of this year. It becomes
effective next year. Now, of course, we’re working on operating rules to
accompany that particular transaction. We have gotten a lot of really great
feedback from NCVHS, and we have depended on that, as well as industry and
stakeholder comments to craft an interim final rule with comments.
We’re working on health plan certification. That’s another area that I know
is of great interest. That is the requirement that health plans certify by the
end of 2013 and then again at the end of 2015 their compliance with some of the
standards and operating rules that have been designated in the standard. We are
working on that. We’re gathering intelligence on that right now in order to
craft a regulation. I think what we’re hearing from industry right now and what
we’re working towards is something along the lines of an attestation, something
very simple, very clean. Let’s make this as easy as possible. I know you have
heard that before, but we’re certainly going to work in that vein.
Finally, of course, ICD-10. I think we all heard yesterday that we received
565 comments on ICD-10. As we had with the 2008 proposed rule, comments are all
over. Some are advocating for staying the course for 2013. Others are saying
2014, which was the proposal that the Secretary put forward in the proposed
rule. Some are saying 2015. Some are saying to skip ICD-10 altogether and go to
11. Some are saying to split implementation between PCS and CM, between
We’re taking all that into consideration. We have categorized all of the
comments. Our subject-matter experts have weighed in. We will be going forward
with a recommendation to Administrator Tavenner within the next seven to ten
days. Of course, because this is an HHS initiative, it will go up to the
Secretary level for her review and consideration.
I think that’s pretty much, in a nutshell, where we are.
DR. TAGLICOD: I do have one more item. In the Federal Register notice
about a couple of weeks ago, I wanted to also share with you the creation of
the Office of Enterprise Management. It does change a little bit of our
relationship in terms of the Office of E-Health Standards and Services. The
Office of Enterprise Management was conceived as a consortium of offices, of
which the Office of E-Health Standards and Services is part. What is new to us
is not so much what is under our bailiwick, which is 5010, ICD-10, admin simp,
and HITECH, but we also consolidated, from an enterprise point of view, the
privacy piece at CMS so that there is one group that is overseeing that from an
enterprise point of view. When I say enterprise, I don’t mean just across CMS.
It is working, clearly, with the Department, ASPE, and with ONC and, of course,
the Office of Civil Rights, in order to address issues regarding privacy and
the use of data.
The three other offices that are part of this consortium, the Office of
Enterprise Management, are an existing office, the Office of Project Management
and Performance, a new office, the Office of Information Policy and Products
and Data, and the Office of Enterprise Business. We would work all together in
order to do, as our name indicates, from an enterprise point of view — and we
even say ourselves within CMS that the days of siloing are done. We need to do
a better job in terms of — and I think you have heard the words yesterday and
today — convergence, alignment, coordination, and integration.
Clearly, timing in life is everything. I think we’re looking at that
coordination and convergence of HITECH, admin simp, 5010, ICD-10, and, of
course, quality performance. Our aim is really to reduce the administrative
burden, not only to providers and industry writ large, but to ourselves within
the Department. There are just a lot of moving parts. The call for a kind of
strategic plan that unites all of this or converges everything, with discrete
milestones and timeframes and responsibilities, is needed. The creation of this
office is to meet that challenge.
I just wanted to let you know that we are looking at a strategic plan. This
cannot be done in isolation. We are not going to go to the ivory tower and do
that. The whole premise of this office is to do it, again, with public and
The other focus — and it got play yesterday — is, how can we facilitate
better communication and collaboration amongst all the parties? Whether that’s
payers and providers, as we were discussing yesterday, or just amongst the
federal stakeholders, state stakeholders, or industry, I think that’s what
we’re all about. But again, with the Department, with ONC here, that’s what we
are trying to do better.
I want to reemphasize our commitment to that public and private
What you can look forward to is that enterprise vision within CMS, within
HHS, including ONC and ASPE and others, and with industry. What are we looking
towards? It’s not just simply the data and the standards. We’re looking
towards, quite frankly — and maybe I should frame it as a question, because
it’s still a question. Are we looking towards 2017 in the new world of payment
reform? How do we get there in real time and in a mobile way?
I think the staff within CMS is looking towards that. But again, it’s a
question. How do we get there — well, first of all, let’s validate that that’s
the goal. I think all of us, as partners, need to address that. What are the
discrete milestones, whether we are talking about testing and ICD-10 or whether
we are talking about how we exchange data n this larger forum towards that end?
So we need to have a common understanding. Yesterday’s conversation was
about readiness. It’s more than just ICD-10 readiness. It’s readiness writ
large. Testing end to end, what does that mean, and the testing environment for
these different initiatives? What are some of the realistic milestones?
Again, we want to partner with everyone around the table, clearly with
NCVHS, in order to move the agenda forward. I think that’s one of the other
exciting things that I would like to put out there.
I’m very much, on Myers-Briggs, an ENTJ. I can talk at this level. There are
clearly folks who are much more the sensate folks, who will say, I need to know
exactly where we’re going. We need you folks in order to articulate what that
is so that we’re much more specific rather than pie-in-the-sky.
That’s what we have to report. Thank you.
DR. CARR: That is great. Thanks so much. This is very, very exciting.
Let’s hear from ONC.
MS. DANIEL: Thank you very much.
Yes, Rob did steal my thunder, but I’ll give a couple more exciting details
that go along with it.
Not only did we hit the 100,000 mark, but we actually hit 110,000. That
represents one in five of all eligible professionals in the United States. So
we’re at 20 percent, which is a really exciting place for us to be. This is, in
part, due to some of the really hard work that our regional extension centers
have done. They are currently working with 133,000 primary care physicians and
10,000 specialists. Among small practice providers in rural areas, there are
over 70 percent of those folks that are working with RECs and 74 percent of
critical access hospitals are working with RECs. So we’re really helping those
in some of the more challenging areas to become meaningful users.
With respect to hospitals, there are 2,400 hospitals that have been paid
through the EHR incentive programs, through CMS. That’s about 48 percent of all
eligible hospitals and critical-access hospitals — so really strong numbers
there as well.
That’s all I’ll do on the meaningful use update and the good news. It’s
always good to be the bearer of good news. We’re glad to be working with CMS in
I have a bunch of different things to cover. The first and probably most
important, beyond the update, is our regulatory efforts. We are actively
working, as is CMS, on our meaningful use stage 2 and state-of-certification
rules. We received plenty of comments, in the hundreds — of course, not nearly
as many as CMS has. We’re working through those and working collaboratively to
try to come out with those regulations as soon as possible.
That said, we’re already thinking about meaningful use stage 3, even though
we’re still working on our stage 2 regs. We have been working with our Health
ID Policy Committee, thanks in large part to the leadership of Paul Tang, in
thinking about stage 3 meaningful use and getting recommendations so that we
can start thinking about the policies, the standards, and make sure that we are
on track for our stage 3 process. They already had just this past month two
hearings in starting to think about stage 3, one on quality measures and
quality improvement and a second on patient-generated data. I think those are
both going to be two pretty exciting areas. We had great conversations and we
will have some interesting insights and recommendations from our committee. So
we’re looking forward to that.
Last on our regulatory front is our nationwide health information network
governance RFI that is currently in an open comment period. We extended the
comment period, we only had a 30-day comment period. Folks asked for more time,
so we extended it by a couple of weeks. That closes, I think, June 29. Still
time to get in your comments.
We had originally planned to put out an NPRM, a notice of proposed
rulemaking, but decided to do an RFI because we felt we really needed some more
input in thinking about the best way to do this, since it is such new ground
and it is such a new area and such developing area and an innovative area. The
HITECH Act called for ONC to establish a governance mechanism for a nationwide
health information network, which is what this is trying to implement.
In thinking about this, there are a couple of things that I’ll just
highlight from our RFI. One, our goal is trying to get to trusted exchange, a
pathway for trusted health information exchange. What we suggest in our RFI is
a voluntary program, not a mandatory program. It doesn’t mean that information
can’t be exchanged in other ways. It just means that this is a trusted way for
exchanging that information. Folks who are trying to share information with
those whom they don’t have a relationship with, a trusted relationship with,
have a channel for doing that in a way that meets certain requirements and
We are focused on the entities that facilitate electronic health information
exchange and looking at mechanisms for some rules of the road — we’re calling
them conditions for trust exchange — for those players who are facilitating
electronic health information exchange.
We also talked about a validation process to validate entities against those
conditions. We talk about a standards-classification process for identifying
standards and different phases of standards, what’s sort of ready for national
standards, and also talking about emerging and developing standards.
Once we get in the comments — and we have had a lot of rich discussions
from our Health IT Policy Committee and our Health IT Standards Committee, so I
expect that we’ll get lots of comments and lots of very substantive, thoughtful
input from folks out in the field — we will process those. Our goal is to then
turn that into a notice of proposed rulemaking as soon as we can. It would
definitely be after our standards and certification rules. We’re trying to get
those fast-tracked to meet timelines. We will be working feverishly over the
summer. We are definitely going to be the “Office of No Summer,” for
ONC we’ll change the “C” to an “S” — as we’re
working on both the standards rule and a governance proposed rule.
A couple of other topics I want to touch on. We are working a lot in the
areas of safety and usability. We commissioned, I’m sure all of you know, an
Institute of Medicine report on health IT and safety, which came out at the end
of last year. They called on ONC to develop a health IT safety surveillance and
action plan. We have been working a lot on trying to tease that out and figure
out what we can do in this space, aligned with the recommendations we heard
from IOM. We said we would put that out within the year. The report came out in
early November last year. You can figure out the timeline from there. Sometime
between now and early November, we will put out a plan.
One of the things that IOM identified as an important issue that was aligned
with safety was usability. We have also been looking a lot at usability. We
worked collaborative with NIST, who did a workshop bringing together vendors
and usability experts to talk about these issues. We also refer to
user-centered design and use of good manufacturing practices in our standards
NPRM. We understand the importance of usability on safety and trying to figure
out how we can improve usability of the products, both to make them easier for
providers to use and, even more importantly, to address safety concerns.
With respect to our S&I framework activity, we presented an update at
our Standards Committee meeting yesterday. Doug Fridsma, our director of — he
has a new title now. We have had a reorganization. He is our former director of
standards and interoperability. He’s now chief science officer or something
We are committed to enabling rapid standards development. We do have
authority and responsibility to support standards and certification criteria
development and adoption now and in the future. The S&I framework
initiative was funded with ARRA funds. We are looking at how to continue our
support and coordination around key initiatives in this important space even as
those funds are used and not necessarily renewed. We’re looking at how we can
use our current dollars to do that.
We just kicked off a Health eDecisions initiative, which is a CDS effort,
just in the last week. The goal is to take clinical practice guidelines and put
them in common standard formats that can be shared and consumed by electronic
health records in order to make CDS incorporation of guidelines easier by EHRs.
We’re working hard on many other S&I initiatives. A couple of the key
ones that folks seem to be very interested in are lab results and lab-order
interface. We also have been working on longitudinal coordination of care and
coordination of long-term post-acute care.
In the area of non-eligible professionals, we continue to hear from our
non-eligible professionals and we continue to try to support efforts to bring
them in with respect to health IT and health information exchange, despite
their not being eligible for incentive payments. The two particular areas that
we have been focused on are LCPAC, the long-term post-acute care folks, and
With respect to long-term post-acute care, we did have a roundtable about a
month and a half ago, and there was a white paper that came out with that.
We’re looking at what the next steps are in working with the long-term
post-acute care community. We have some state challenge efforts and then, of
course, the S&I initiatives.
With respect to behavioral health, we have a lot of work that has been done
now on behavioral health quality measures. We are going to similarly be doing a
roundtable on behavioral health and the incorporation of behavioral health in
primary care, in late July.
My last topic is on consumer e-health. This is really exciting. There are
some of us who have been around for some time in ONC and have been articulating
a need for focusing on consumer e-health and using health IT for consumer
engagement. It has not been the number-one priority since early in the ONC
days. What’s really exciting is that it has become a priority in a very quick
period of time over the last year, to the extent that we did just do this
reorganization and we created a new Office for Consumer eHealth. We will be
doing a national search for the director. Lygeia Ricciardi is now the acting
We’re coming up on the anniversary of the launch of our consumer e-health
program, which is September. We will have an anniversary of our launch event in
mid-September to talk about progress and some of the new things coming down the
pike. So keep an eye out for that.
Our pledge program now has over 375 pledges. We highlighted progress at the
HGI. There was actually a consumer track. This was the first time that we had a
consumer track at HGI. That was more evidence of this really becoming part of
the national dialogue on health IT.
We have a series of activities that either have been kicked off or we have
had some meetings that are kicking off some new activities. We have a cancer
care initiative, which is taking a focused-use case for engaging consumers and
families around cancer and cancer care. We are collaborating with the state of
Texas and some providers in the state of Texas. Our goal is to have a pilot and
an Apps challenge in this space. We just recently had a roundtable with NCI and
eHI on informing the research agenda on patient engagement with health IT for
So we have kind of a confluence of efforts with respect to use of health IT
for consumer engagement and cancer care.
We had a patient access summit at the White House, where we brought folks
together from the standards and technical world along with the patients to talk
about how to improve patient access to data. There were a couple of things that
were highlighted as next steps. One was to work on an auto blue button. Folks,
I’m assuming, know about the blue button efforts for download of patient
information. The goal is to make that come alive a little bit more and make an
auto blue button and put some standards around that as well. There was a lot of
discussion as well on patient ID and authentication with respect to patient
access to information.
We launched a blue button challenge, together with the VA, to challenge
developers to build on the current blue button feature that allows patients to
download their health information and share it with health-care providers and
caregivers. The challenge requires the development of a tool that will help
individuals use that health information and combine it with other types of
health information, like cost data, comparative health data, et cetera, to help
patients better understand their own health status and make informed decisions
with respect to their health care — trying to, again, make that data come
alive and be more useful and actionable once folks actually have access to the
Lastly, we have been focusing on how to help consumers better understand
health IT, the value of health IT, and ways that they can use technology to
support their own health and health care. We have a health IT animation that
will be coming live very soon. We actually did a preview at HGI, but it’s not
quite ready for primetime. Folks who were there got to see it.
We’re trying to figure out creative ways to get consumers to both understand
health IT and understand why their doctors might have an electronic health
record in their office and how it might be valuable to them and how it can help
them engage and interact and relate with their clinicians.
Lastly, we have been doing a series of video challenges, where we’re getting
patients and consumers to tell their own stories of how they are using health
IT to manage their own health and health care. We have done two of these so
far. The first one was a Healthy New Year Video Challenge. We have some videos
up on our website. The second was a Blood Pressure Video Challenge. There will
be a few more over the course of the year. We’re really looking at how we can
get others to be telling their stories so that it’s more real people talking to
real people rather than us talking to them about the value we see in all of
DR. CARR: What I would like to do is give Joy a chance to do your
presentation. Then we’ll take questions.
MS. PRITTS: Thank you.
I am going to start, before I go through my slide show, to talk a little bit
about some of the privacy and security issues in meaningful use and governance,
just to dive into the privacy and security elements a little bit more.
In meaningful use stage 1, there was a requirement that providers conduct a
security risk analysis, which, you all know, is currently required under the
HIPAA security rule and has been for years. This might seem redundant, but I
will assure you, it has had an enormous impact. As a practical matter, it has
alerted people that they should have been doing something for quite a while.
The response we have received on this particular issue has been — I don’t know
what adjective to use — appalling, but also encouraging that people actually
know now that they need to be doing this when they are installing their
electronic health record and on a regular basis.
We are also attempting to ensure that security is as easy for the providers
as possible. This is one of our goals. We understand that health-care providers
do not want to become health IT specialists. One of our goals is to try to
drive some of the security into the electronic health record itself, or
electronic health record systems, to alleviate some of that burden on the
providers. There are some security functionalities that were built into the
certification criteria for stage 1 and are being proposed for stage 2.
With respect to the governance request for information that Jodi was
referencing, a lot of the conditions for trusted exchange, of course, revolve
around privacy and security. In the security area in particular, our view on
that is that these large organizations that are going to be exchanging a vast
amount of health information should be adhering to somewhat similar standards.
The security rule itself is flexible and scalable, which is necessary when
you’re dealing with a huge range of providers and other entities sharing health
information. Yet at a certain point you look at entities that are fairly
similarly situated, which is what we are looking at governing with this rule,
and the proposal is that they should be adhering to somewhat similar standards
as they are going forward.
I’m going to start with some of the privacy and security policy
recommendations that you might have heard or seen sometime in the past and give
you an update on how some of those have been moving along. In particular, we
issued in March of this year a state health information exchange privacy and
security program information notice, which gives guidance to our grantees, our
recipients of federal funds, on how we think they should be proceeding and
looking forward on some of these key privacy and security issues. They
incorporate some recommendations that were made by this committee, as well as
the HIT Policy Committee in the past, particularly with respect to informed
choice to participate in this query exchange type of health information
exchange and strong provider authentication requirements.
There is a link there if you are interested in reading the entire PIN. It is
publicly available. If you use a Google search of “state program
information notice,” it comes up right away. That’s probably the easiest
way to find it rather than trying to check out that link. It’s really kind of
We are also working, on a big-policy level, with the administration on the
National Strategy for Trusted Identities in Cyberspace, also known as NSTIC,
for those of you who have not heard this particular government acronym yet. It
is essentially — I think a kind of crude way of putting it is that it’s almost
like one single sign-on, but on a very large scale. The idea is that providers
and individuals should be able to get ID-proofed once from a verifiable source
and be able to use the credential that they obtain in multiple settings. This
is a priority administrative initiative. It is being run out of NIH primarily.
We are working very closely with them. We are having a hearing with the Privacy
and Security Tiger Team on July 11. That will be looking at this issue of
provider ID proofing from a provider perspective. Then we will be taking the
patient perspective a little bit later in the summer.
The issues are very distinct. The challenges are a little different. So we
concluded that it would be necessary to have two separate hearings on that.
In the policy arena, where we coordinate with our other divisions within HHS
— some of them aren’t here for me to steal their thunder, so I will just put
them in a good light — the HITECH modifications to HIPAA — the final rule
went over to OMB on March 24, 2012. As a general rule, OMB has 90 days to act
on that. We can all do the math as to when we might think that these might come
out this year.
Some of the key provisions in here: This rule will finalize the break
notification rule. Probably the largest impact on the scope of HIPAA, I think,
is that it extends the use of disclosure provisions of the HIPAA privacy rule
to business associates, as well as most of the requirements of the HIPAA
security rule. This is really a widening of the impact of HIPAA and the
enforcement ability of the Office for Civil Rights
We have been working very closely with CMS over the last year, I would say,
in a very expedited way, on a lot of the Affordable Care Act rules that are
coming out that are establishing these new entities that will enable us to do
more analysis of health information and more sharing of health information,
holding providers more accountable and also enabling individuals to obtain
their health insurance.
I’ll run through these very quickly. The accountable care organizations —
the final rule that established them sets out that they may be business
associates, but it also says that individuals may opt out of having certain
identifiable information shared between some of the different providers. There
are definitely privacy provisions that have been baked into this.
In addition, the final rule was issued on the availability of Medicare data
for performance measurement of providers. This is the rule that creates these
things that are being called qualified entities. Some of these exist now, but
this is formally giving them a way to access more CMS data. They are not
considered business associates of CMS because they are not performing this
function on behalf of CMS. But they must have a rigorous data privacy and
security program in order to qualify to receive the Medicare data and they must
sign a very stringent data-use agreement.
There’s also a final rule establishing the health insurance exchanges, which
are different from the health information exchanges. These are the exchanges
where individuals will be able to obtain health insurance at the state level if
they are not insured through a group program. There are baked into these rules
requirements that the state health insurance exchanges must establish and
implement privacy and security standards that are consistent with fair
information practice principles. These entities are going to be exchanging
information from a number of different resources that have a number of
different types of privacy and security protections already in place, including
the IRS and SSA. The requirements on these have to be fairly high-level in
order to accommodate all of those different paradigms that already exist.
In addition to all of these policy initiatives, I would say, that are going
on, I would also like to give you a little bit of a snapshot into some of the
research and internal activities that my office, the Office of the Chief
Privacy Officer, is undertaking. When you see these, you will see that these
are also tied to a lot of the recommendations that have come out of this
committee in the past and the HIT Policy Committee. I will go through these
very quickly, given the time.
The first one is the Data Segmentation for Privacy Initiative. We used to
call this masking of data. What this is intended to do is to look at and assess
standards for sharing part of a medical record, but not all of the medical
record. It is not intended to create new policy. It is intended to help
facilitate the electronic implementation of existing laws, such as those that
apply to substance and alcohol abuse-related information for which there are
already very stringent standards.
This project is in process. It’s moving quite rapidly. It is to be wrapped
up this September. It is, again, focused on the standards element and looking
at some of the standards for metadata tagging that were recommended by the
Standards Committee, I believe, last year.
In addition, we have what we’re calling the eConsent Trial, which is a pilot
project which is focusing on some of this meaningful choice. The Policy
Committee and, I believe, this committee in the past have recommended that
individuals need to know how their information is being exchanged. It can’t
just be, here, sign this document. Having heard those recommendations, we also
heard from a lot of providers saying, you’re asking us to do a whole lot here.
This project is designed to find some alternative ways, other than just a
face-to-face with the provider, of potentially helping educate individuals and
assuring that they really do understand what their choices are. This is going
to be looking at things such as putting information on tablets or other forums
to educate individuals. It also has definite metrics involved in it to measure
whether the individual actually understood what it was that they agreed to and
whether they find this method of education to be beneficial.
That project is pretty much at the mid-mark now. All of it will be concluded
in March of next year.
We’ve also designed what we call our mobile health portfolio. It has a lot
of different components in it. The reason that we have focused on mobile health
is that I think anybody who has been reading the news has seen that the
adoption of some of this technology went from pretty flat about a year and a
half ago to just skyrocketing recently. There are also a lot of surveys that
have been done that have showed that although the use is skyrocketing, the
attention to security is not keeping pace. We have a number of different
projects that are ongoing in this area.
The first among those is working with OCR to develop a set of good practices
for ensuring security with mobile health technology. We had a roundtable, and
we are conducting other information gathering. We also had a project where we
tested smartphones and tablets out of the box to look at the security that was
built in. The second part of that will be showing individuals very basically
how you can add more security by doing manual configuration of these devices.
The education materials here will be in various formats, including short
videos and things of that nature. We have heard very strongly from the
community, the provider community that we work with, that they do not want any
more PDF products. They want other things. That’s how they learn, and so we’re
trying to accommodate them.
This project is well under way and should be concluded, I think, October 1,
2012, the end of this fiscal year.
As another part of this mobile health portfolio, we also have a consumer
attitudes survey. This was developed out of the Text4Health Taskforce, which
requested that we look at these aspects for them. We are working with other
divisions in HHS to help them answer some of the key questions they have with
This is particularly looking at text messaging, email, and Skype, and other
uses of apps from the consumer perspective, and how they feel. There is a
certain tradeoff that individuals are willing to make with respect to security
versus ease of use and things of that nature, and looking at some potential
safeguards for this information that are easily usable. That also has a
deadline of this year.
In addition, ONC has funded the SHARPS project, which is Strategic
Healthcare IT Advanced Research Projects on Security. With our reorganization,
these have been moved into the Office of the Chief Privacy Officer to oversee.
Two of the issues that they are looking at in particular are the security of
implantable medical devices and consumer attitudes regarding privacy of remote
monitoring devices, such as wristbands and things of that sort that measure
You can see that we are getting a pretty good view of the whole sphere
through a lot of different mechanisms here.
We also have as one of our big metrics in our strategic plan — all of HHS
has to have metrics now — one of ours is trying to measure consumer trust of
privacy and security over time as adoption increases. We have a survey that
will be fielded this year. It is pretty close to being approved. The idea here
is that we are going to be looking at this for the next four or five years to
see how consumer attitudes are changing about privacy and security over time as
implementation really goes live.
That’s the end.
DR. CARR: Tremendously productive, exciting, cutting-edge stuff across the
board, all of you. Really, this is one of the most exciting presentations we
have had from the various departments.
We have time for a couple of questions. I see Larry and Walter.
DR. GREEN: Wow. Amazing leadership.
Question: Of what use is NCVHS to you?
DR. BUENNING: NCVHS is of tremendous use to OESS and CMS in particular, in
that it provides the neutral and level forum for the exchange of information
that’s really critical to us in the formulation of not only policy, but also
our regulations. I know that all of you read our regulations very carefully,
and I know that all of you will recognize that oftentimes, number one, NCVHS is
required statutorily to advise the Secretary on recommendations that we use in
the formulation of our regulations. Number two, it provides a vast amount of
subject-matter expertise that we bring to bear.
So we very much value these forums, as well as the individual subcommittees
that form the foundation of the NCVHS. We think it’s absolutely invaluable.
DR. CARR: Let’s hear from all of you, because I think that’s a critical
question. Larry, thank you for asking it.
DR. TAGALICOD: More than the reiteration of what Denise has said, NCVHS is
seen as the honest broker. This is where the realpolitik of it is. This
is not a plus or a minus or a bad or a good. Entities, stakeholders have their
agenda, and they should. I think this is the convening place — and I had in my
head to say, the great convener, the great integrator — in order to help do
I think yesterday was very eye-opening. I think I need to attend those more
often and come earlier. It was a good place to hear certain things and where
people brought things together for us to consider. If you saw any of the staff
that were there, including myself, we were taking profuse notes. Then we were
having these offline conversations over dinner or breakfast this morning going,
how could we do this better, and how do we do this better with NCVHS, with all
the partners around the table?
That’s a further elucidation of what we think.
MS. DANIEL: It’s a great question to ask, particularly with ONC. I manage
our Health IT Policy and Health IT Standards Committees. It’s a challenge just
to even coordinate between those two. We actually manage both of them. We have
often had the question about how we can better work collaboratively with NCVHS.
I would say, personally, because I have been sort of focusing on those two
committees, I have not been focusing as much on NCVHS’s work. But there’s a lot
to be done. As you can hear from all this, there’s a lot going on. To the
extent that there are things that are important, but that we’re not taking up
in our advisory committees that we still need input on, it would probably be
useful to regularly — quarterly or even every six months — sit down and try
to think through what your agenda for the year is, what our agenda for the year
is, where there are overlaps and where there are opportunities.
Perhaps a take-home for me is that we should probably make sure that we do
have somebody from our staff who is here for the full meeting, not just giving
the updates, so that we can bring that back as well and think about how to
align the work better and how to get the most out of what NCVHS does best
versus what the Health IT Policy and Health IT Standards Committees can do
DR. CARR: Great idea. We welcome it.
MS. PRITTS: As you know, I think, we have populated the Privacy and Security
Workgroup with some members from NCVHS to make sure that there is some
coordination going on and to identify issues that they think might be better
addressed over here than over at the policy committee level.
What I think in particular, going to a specific example right now —
somebody was mentioning some of these standards, electronic fund transfer, for
example. The EFT and the ERA — I have to tell you, having worked on this issue
for a number of years, I’m scared about where that particular piece is going,
or concerned about it, because it has been kicked around since 1996 pretty
much, as to what the statute actually means on that one.
So there are specific things that are moving with standards that hardly even
come up to our radar screen that we really need to know about and that really
have, from my perspective, potentially serious privacy and security
implications. That particular aspect is something that I think this group is
really very well poised to help us with.
DR. CARR: I want to be respectful of our time. I’ll declare the break now,
but ask the speakers if you could stay around for a bit. I know there were a
couple of questions that folks had.
We’ll take a break now.
DR. CARR: A couple of housekeeping things.
DR. CARR: One very important thing. As you know, in our lifecycles here on
the committee, we have graduating classes. Two of our current graduating class
elected not to accept the option of being extended six months because of
competing responsibilities. Mark Hornbrook and Blackford Middleton, you’ll
notice, are not here with us today. I just want to take a moment to acknowledge
the tremendous contributions from both of them, both incredibly thoughtful,
broad-thinking, both big-picture and practical, who really always brought one
more idea to the discussion that made us think better and clearer and bigger.
Will we be bringing them back?
MS. GREENBERG: We certainly will be preparing the usual letter from the
Secretary telling them that we’ll be calling upon them in the future and
thanking them, and the certificates. I think probably September will be our
dinner for those members who are retiring in 2012. They will be welcome to come
back. I don’t know if they will be willing to. Otherwise, we’ll mail them the
DR. CARR: I just want to say that Harry Reynolds, had he continued his
projected tenure, would have been part of that graduating class leaving in
September. He has made a special request if he could be there, because
apparently he has some things to say.
MS. GREENBERG: And, actually, Simon Cohn — you never really go off this
committee, so just accept it — Simon Cohn was a presenter yesterday. He also
said he would like to try to come to that dinner.
DR. CARR: Is that right? Oh, how fun.
MS. GREENBERG: So we’ll have to look for a good place to have the September
DR. CARR: Great.
With that, we’re now ready to hear from Populations on your hearing and your
letter. We have the letter. You will give us an overview of the rationale
behind this, what we learned, what we thought, and what the letter says.
MS. MILAM: We will. In fact, we have a PowerPoint to assist us with
answering all of Justine’s questions.
I want to draw everyone’s attention — you do have a hard copy of the letter
in your material. It’s behind tab 3.
What we will try to do with this PowerPoint is answer all of the questions
about why we undertook this area, who it is for, and all of the relevant
issues, so that you can better understand the scope of the project, as well as
what we heard.
As Jim said this morning, the focus of the hearing was to better understand
socioeconomic status and see how far we could get with development of
recommendations. The hearing was held March 8-9. This came at the request of
the HHS Data Council.
HHS has authority to issue standards regarding data collection, as Jim
talked about this morning. It has been given a specific charge in the ACA, as
well as a general charge. The general charge is the part of the statute that
were keying off of. We identified from Jim the need to focus on recommendations
around SES data collection. We found it fairly complicated. I’ll say that.
We’re really excited to work in this area.
We had excellent testimony over two days. Unfortunately, I was not able to
participate in the hearing. I got to listen over the phone, but did not get to
be there in person. I found the slides really, really interesting. I enjoyed
At this point I will turn it over to Larry to talk about more of the detail.
DR. GREEN: This is a masterful understatement on this slide. After wallowing
in complexity, it comes down to the fact that the standardization effort should
focus on education, income, occupation, and family size and relationships.
That little sentence right up there is pretty potent. We’re pretty sure that
there is no gold standard. We don’t think there is, and we also think it’s
unlikely that there ever will be a one-size-fits-all reporting measurement for
those variables. The reasons for that are:
- Surveys have different purposes.
- Surveys also have different legislative authority.
- The populations of interest vary.
- Surveys are not actually conducted in the same manner, and their mode of
conduct. This has a big impact on what can be done.
Some of the findings and observations — I want to ask you indulgence for
personal privilege. Sitting through this, there was no greater learning on my
part from this wonderful exercise than the development of great respect for the
individuals who are members of the federal statistical community. There are
some very talented people here. They are remarkable individuals. They do good
work. They are classic examples of invisible leadership. One of the things I
learned was some humility. It’s very important to be careful with this, because
this is complex, and interagency collaboration and listening to these experts
is very important.
Also the role of the census surveys is very potent, very powerful. These SES
dimensions — we also learned that it’s not just a person’s or a community’s
socioeconomic status today, but these things fluctuate for individuals,
families, and communities. It is those changes and their effects that are
Then we learned about the key measures of SES.
To Justine Carr, what are we recommending and why? This is critical
territory, but it’s a very challenging area. From where we sit this morning,
we’re convinced that it’s too early for us to be writing a letter at this point
that makes very detailed recommendations about adopting specific standards. As
Jim said a couple of times, a standard is a very high bar to meet. We’re
recommending that HHS undertake some additional efforts to explore the gaps in
data collection and analysis for measuring SES on health surveys, particularly
to improve our understanding of health disparities.
These are the major challenges that can be anticipated here. We’ll just
leave those up there for a minute. For the people on the phone:
- Again, survey purpose and use may impact the standards.
- It’s important to distinguish the measurement of household and family.
Those are different things.
- There is under- and over-reporting of certain variables.
- There are missing-data problems.
- There are challenging imputation and allocation techniques.
- You often need additional information to provide meaning to the answer to
- For income, an example is that the measurement of retirement income is and
is going to be a growing challenge.
This is our last slide. Vickie Mays served as the hearing chair. Bruce Cohen
we also want to call out, and Jackie Lucas working with Susan Queen. These four
individuals, particularly Vickie — it would be hard to overstate their passion
and their expertise in this area. Vickie provided fantastic leadership for
this. Bruce has this way of sending you an email that sort of gives you
whiplash, but makes you think. Again, Vickie and Bruce are examples of the
talent and passion that really matter here. We thank them very, very much for
getting us where we have gotten so far.
I would like to invite both of them to make additional comments before we
open it up for your questions.
DR. MAYS: I think that the hearing was very insightful in terms of the task
that Jim faces, and that is that, for SES, the problem is the differences that
the various surveys need and trying to come up with a standard. I think one can
do it for some of the surveys, but not all of the surveys. If I had my
druthers, it would be even saying which surveys you could have standardization
for. I just don’t think we can do them for all of them.
But going forth in the future, I think this is probably one of the most
critical issues for the survey, now that we have gotten the other things under
way, because this is what the data keeps showing us is critical, especially for
the variable of education. It’s the wave of the future in terms of looking at
DR. COHEN: I don’t have much to add. A couple themes. Unfortunately, I was
unable to attend. Really, everyone has done phenomenal work in this area.
First of all, I had neglected historically focusing on occupation. That
reemerged as something that we need to consider, for a variety of reasons, not
only for environmental exposure kinds of issues, but as an important factor in
looking at socioeconomic status. I think the dynamic nature of this variable is
a key that Larry highlighted. With economic downturns and rapid changes and
increasing disparities socioeconomically in our country, I think SES is going
to play a greater role in looking at health in particular groups in our
population. It becomes even more critical that we develop appropriate ways to
quantify and measure this issue so that we can really target the populations
most in need.
MS. GREENBERG: First, I want to thank everybody, Vickie, Larry, Sallie,
Bruce, Susan Queen, Jackie Lucas, Debbie Jackson, who also helped pull this all
together. It takes a family.
I unfortunately missed the first day, but I was there the second day and
just found it very exciting. Specifically, I also was really glad to see
occupation get the attention that, I agree, it needs.
I just want to say that I think — and I don’t want the subcommittee to sell
itself short. I’m sorry I didn’t have a chance to participate in the
development of the slides, because I was in Spain. I think it’s true. The
subcommittee is calling attention to all of the additional work that needs to
be done and the complexities. Jim and I definitely thank Larry for his support
of the federal statistical community, who really came out in force. We held the
meeting at NCHS. So that was all very nice.
But if you look at the letter — and you will be doing that, obviously
— the subcommittee, on page 2, did provide some additional guidance on these
four variables: education, income, occupation, and family size and
relationships — some minimal areas for standardization. Although the letter
says that additional work is needed, I think the hearing and then the
distillation of those results by the subcommittee and the staff did move us
forward, with some additional recommendations.
It’s my understanding that, hopefully, this will be very helpful to Jim and
the Data Council and the Department as it explores trying to identify minimal
standards that can be promulgated.
Do you want to say anything about that, Jim?
MR. SCANLON: Let me thank, first of all, the full committee and all of the
members of the Population Subcommittee. I attended most of the hearing itself,
too. You always learn a lot, even when you think you know a lot about this
already. It just worked out perfectly.
What we really needed, most of all, was kind of a state-of-the-art review of
where we are and where we are heading, where there is agreement and where there
is not. We looked at, for example, how income data is already collected on HHS
surveys. It’s probably the one item that differs more among the surveys than
anything. With other demographic items, we’re fairly close, and we could
probably collapse it in some way.
We knew there wasn’t an obvious answer here. Sometimes, with the nice
analysis that the committee is giving us, it gives us a way to proceed. Even if
we’re a little far away from a standard — that’s a tough level to meet — we
could think about a little bit experimentation. We could think of a little bit
of best practice or preferred practice and gather a little more information and
see where we get.
There are some items, I think — education, for example — that are always
turning up as related to a lot of things. Whether it’s causal or whether it’s
associated with something else, it seems to make a big difference.
These findings are very helpful to us. We’ll take them to the Data Council’s
workgroup and have the agencies digest these and see where we could go next.
That’s exactly what we asked for.
DR. MAYS: I just want to make sure that we put it in the record to say thank
you to the Census Bureau. They really did yeoman’s work of turning out several
— I happened to see Bob Grove after that, so I thanked him personally. But I
just want it in the record. Not only did we get several people from Census to
come, but they did an excellent job. Thank you to the Census Bureau.
The other thing, Jim, is that you may want to keep an eye on Michael’s
Hout’s work. He got an NSF grant to study the measurement of SES in GSS. He is
looking at their measurements starting from the beginning all the way through.
They are looking to reengineer the General Social Science Survey. I talked to
him also a little bit later as well. I think over the span of the next couple
of years, he will have some information. So you may want him to talk with you
DR. CARR: What we need to do is read the recommendations and then go to a
DR. WARREN: Can I ask one question first? I was waiting for your committee.
As you were looking at occupations, did the standard occupational code set come
up at all?
MS. GREENBERG: Definitely.
DR. WARREN: I’m wondering about that because it takes quite a bit of effort
to get a SOC code. As we really are trying to look at SES, and occupations are
changing rapidly, especially in the whole IT world, how do we go about making
sure that we have the right codes to measure what the occupations are?
MS. GREENBERG: We did hear from NIOSH, which is responsible for the
occupation and industry coding. I had particularly wanted them there because
they have made a lot of progress in automating the system, just taking the text
both for occupation and industry and creating the codes. They have done a fair
amount of validation on it as well.
I think that’s something that is mentioned also in the letter. Maybe you
want to expand on that.
DR. GREEN: On page two, under that bullet of occupation, it’s remarkably
brief. Bruce helped us get to the sentence, I think: Automated coding of
occupation and industry survey responses should be further developed. There’s
some real progress there, to help get to the proper code.
DR. WARREN: The reason I ask for this is that the nursing community has been
looking at the SOC codes. If you just go in and what you get is a nurse, there
is a wide variety on the amount of money made, the job responsibilities, et
cetera, within the nursing profession. Right now that’s all lumped together
pretty much under one code. I think there are probably lots of other codes that
are the same way. Once upon a time they captured the intent, but they no longer
DR. CARR: We have four recommendations. So we read the recommendations, and
then a nice summary of what we heard, similar to what you gave us today. We can
go through the recommendations, if there is any discussion, and then we’ll
entertain a motion for a vote.
Do we vote today?
MS. GREENBERG: You can.
MS. MILAM: We would like a vote, Justine, if it works out. Let’s see how far
we can get.
DR. CARR: All right. Take it away.
MS. MILAM: We are really starting on the second page of the letter. That’s
where you will find our recommendations, behind tab 3. I’ll begin with the
The committee has confirmed that the following areas are the key components
that are necessary for the measurement of SES and on which standardization
efforts should focus, and has the following initial recommendation:
Education. The relationship between education and health outcomes is well
established. Education should be measured in single years completed up to five
or more years of college and should also include collection of information on
whether or not the individual obtained a high school diploma or equivalent.
Surveys should also collect information on degree attainment.
The second recommendation there is around income. Income is critical for
policy analysis and is necessary to describe the economic condition of the
population nationwide and for geographic areas and subgroups. Income is
particularly important for research on disparities in health care access and
utilization. Income should be asked for the individual survey respondent, as
well as for the respondent’s family, as well as household income. The
collection of income should include the measurement of total income, as well as
income from specific sources. We have a parenthetical: e.g., wages and
salaries, Social Security, disability income, et cetera.
The third recommendation is around occupation. Occupation and industry are
core socioeconomic variables collected in all federal health surveys. Relevant
distinctions in occupation are not detected by other indicators of SES. At a
minimum, a set of two standardized questions to collect occupation and industry
should be implemented. The set of four standardized questions used in the
American Community Survey that collect additional about work tasks and employer
should also be considered. Automated coding of occupation and industry of
survey responses should be further developed.
Family size and relationships: Family size and composition are required to
calculate poverty, and survey measures should collect information on family
size and composition, in compliance with official federal poverty guidelines as
issued and published each year.
Those are the four recommendations we distilled from the two days of
hearings. Is there discussion on the recommendations themselves? Ob.
MR. SOONTHORNSIMA: I just have a very quick question about family size and
relationships. Do you imply also where you have households with a
multi-generation dimension? How does that fit in that context, in this
MR. SCANLON: It’s in there.
MR. SOONTHORNSIMA: Is that implied?
MS. MILAM: “Composition” is a broad term. I think it could include
anything we need it to include. But you’re right, it is very broad.
MR. SCANLON: Typically, you list everybody in the household and you ask
about relationships. There is often more than one family or extended family in
a household. If you get the list to begin with and the relationship, that gives
us some flexibility.
MR. SOONTHORNSIMA: The only reason I’m bringing it up is because you hear
about hardships placed on the family head of household, whoever that is, taking
care of elderly parents or disabled family members and so forth.
DR CARR: I have Walter, Judy, Linda, everybody down to Bill. Walter?
DR. SUAREZ: This side of the table is the standards section. The comments
that I have are in line with what Ob was pointing out. I’m fully supportive of
all this, but I was trying to figure out — when you are recommending
standards, either you are very specific about age should be collected
in the following ranges, or something like that. It seems to me that the
recommendations are broad, so that someone can then define the standard. I
guess that’s the intent, right? For example, it says education measured in
single years up to five. We give them some parameters for them to actually
define the standard by which the data will be collected.
There’s only one element that I have some question about, and that is
occupation. The recommendation says, at a minimum, a set of two standardized
questions to collect occupation and industry should be implemented. I’m not
sure which are those two questions, whether it’s one question about occupation,
one question about industry or it might be helpful to clarify
DR. CARR: Maybe a little wordsmithing there to clarify.
MS. MILAM: We can fix that.
DR. SUAREZ: In that same one, a set of four standardized questions using the
American Communities Survey — is that referenced in the document or do we have
a footnote in the letter that explains or clarifies what the American
Communities Survey is?
DR. CARR: We will do that.
MS. GREENBERG: Also I think there are minutes from the meeting, a summary of
the meeting. I don’t know if those have been made available yet
DR. MAYS: They are in the book.
MS. GREENBERG: That will give us more detail about what the committee
DR. CARR: But we want to make sure the letter reflects —
DR. WARREN: I think the reference needs to be in the letter, because that’s
what people see.
The other thing is, quite honestly, I did not know there were
recommendations in this letter when I went through. I got to a conclusion —
because the recommendations were so global that I didn’t see them. And I missed
the one term that says “initial recommendations.” If there could be a
bold header for that, because if I missed them, whoever this goes to would
probably miss them as well.
MS. MILAM: Thank you, Judy.
DR. CARR: Linda.
MS. KLOSS: My question was the same as Walter’s, clarifying the two
DR. CARR: Vickie.
DR. MAYS: I want to go back to family size and relationships. A couple of
things came up, and I’m not quite sure how to change it, if possible, but how
to consider it may be something to put on the table. When we were talking about
immigrants, for example, one of the things we were aware of is the extent to
which they support family and family is not a part of the household. There are
some different things that we potentially could talk about. I’m not sure how to
do it. I was trying to decide whether to modify composition with family and
household composition, whether that helps to push it a little bit.
DR. WARREN: That would help me a lot.
DR. MAYS: It is just to modify where it says family size and composition,
which means, then, that you are asking about the family, and it’s the total
family, regardless of where they are. Then, to get to what Ob is saying, you
ask about the household, so that you get everyone, regardless of whether they
are biologically related to the person or not. That might help if we have that
MS. MILAM: Vickie, are you saying to add “household” right before
“composition”? Is that the amendment?
DR. MAYS: Where it says family size, “and household composition,”
so that it’s understood that you want to have the information about the family,
as well as the household.
DR. CARR: What was the clarification?
DR. MAYS: It should be family and household before the word
“composition.” And it should be “and,” not a slash. I think
the survey people are really going to talk about it, because to sometimes do a
whole household may be a lot. But I think that’s how you get to what Ob is
saying and then how you get to non-biological, as well as non-residents.
DR. CARR: Bill.
DR. SCANLON: It is about income. I guess I’m reading into your intent in
terms of being more comprehensive about sources. A suggested word change would
be to put in front of “income” “earned and unearned
income,” from specific sources. You might want to think about the list, to
get an example of unearned in there, as well as think about the issue of the
prevalence of different sources of income. Census talks about retirement and
disability benefits. Then there are also income maintenance benefits and
unemployment compensation. Getting at some of the bigger categories may be
helpful in that list. I would probably put retirement and disability benefits
and income maintenance and unemployment compensation and then say dividends and
interest, or something like that.
PARTICIPANT: We weren’t trying to be all-inclusive.
DR. SCANLON: The issue was that the list at this point doesn’t include
anything unearned. The disability is one of the narrower categories.
DR. GREEN: I would like to say this back to you. We need to get the words
“earned” and “unearned” in front of “income.”
Then in the parenthetical list, we need to get two or three examples of these
other broader categories. Correct?
DR. SCANLON: Right. That’s it.
DR. WARREN: That brings up another question. Just bear with me, because my
standards mind is working. I don’t know what these surveys look like. Are these
open-ended questions that the surveyors ask? Sometimes. To me, especially in
occupation and maybe even in income source, there should be a code set that’s
developed for the answers and then leave — as much as I hate the word
MR. SCANLON: There are. It’s well developed.
DR. WARREN: IS there a way to say that in this letter, to use those accepted
code sets for these responses or to work on developing them or evolving them?
DR. GREEN: I think the reason it doesn’t say that is that we did not review
them all. To say you should use these code sets would be a fairly bold
DR. COHEN: The use of standard codes is critical to understanding — we kind
of address it, but not specifically, in the recommendations.
DR. WARREN: Maybe one recommendation would be to explore all these code sets
to see if they are relevant.
MR. SCANLON: That’s the issue itself, Judy. There are some differences in
terms of the categories.
MS. MILAM: Judy, are you suggesting a fifth recommendation around the —
DR. WARREN: I am willing to defer to what the committee thinks the purpose
of the letter is. That’s why I said, when I read it, I start thinking what the
answers are, how I will deal with the data, and things like that. And can I
collapse and aggregate? I don’t see any ability to do that. If that’s something
that is next work, then maybe that goes in a conclusion or in your findings or
something. But I do think at some point it would really strengthen the letter
if the whole issue of code sets is looked at to do a comparison and a
harmonization. I think that’s one of the biggest problems of some of the
federal surveys. We use different words for the same thing, and, unfortunately,
computers can’t intuit the intent.
DR. CARR: The second paragraph maybe bears reading. The first one is about
NCVHS. The second one says, NCVHS was asked by HHS Data Council to consider the
development of standards for the collection of information on socioeconomic
status in health surveys conducted by HHS, in accordance with ACA Section 4302.
DR. WARREN: There are standards for collecting these surveys. There are also
code sets that are the acceptable values of questions in those surveys. To me,
the standards for the surveys and the code sets are two very different things.
DR. CARR: I think using the term “code sets” makes it clearer.
DR. GREEN: Let’s see if we can do it this way. In that third paragraph on
the first page, the paragraph that starts, “Policymakers, researchers, and
other data users are better served,” can we just — Susan, help me here —
in that sentence, the one that starts with, “Standardization allows more
uniform data collection and comparison across surveys,” can we insert into
that something about the harmonization as code sets, as possible, or something
DR. CARR: I have a clarifying question. We are finding it buried in there.
Actually, you have been exceedingly concise in your language, which is great.
But we want to make sure that we don’t miss an opportunity. Is there a risk
that we would to use the code sets in this or is it implicit that that would be
the only way we would do this?
DR. GREEN: Bruce and Vickie and Sallie may have a differing viewpoint. My
own viewpoint is that the reason we’re not making strong statements about code
sets is that it’s a mess. I think we need to be very careful to not overstate
— remember a while ago I was saying that I have gotten a very strong dose of
humility out of this exercise? I went into this exercise thinking, oh, yes,
we’re going to be able to make sentences like this that use this code set or
these two code sets. All we have to do is just fix these two lines and then
they will be standardized. No. And sometimes it’s because the law says, you
will do the following.
DR. CARR: Could I ask another clarifying question? Would we find a middle
ground by saying “leveraging available code sets where appropriate”?
DR. GREEN: That is the kind of language I’m looking for.
DR. WARREN: What I was thinking of — in the concluding comments —
DR. COHEN: Yes, that’s the place to put it.
DR. WARREN: — it says, we recommend that HHS undertake additional efforts
exploring gaps in data collection. Then I would say, code set composition and
analysis for the measurement of SES. Just leave it there, and that’s a future
DR. COHEN: Actually, that is one thing I wanted to discuss, whether we
should strengthen and be more explicit and specific in our concluding comments
when we recommend that HHS undertake these analyses, whether they report back
to us at some point in time about findings with respect to harmonizing and
deciding which code sets to use and how to measure some of these variables that
we are uncomfortable making recommendations on.
DR. WARREN: I would bring up again that you have a recommendation buried in
the concluding comments. Is that really a recommendation or is that a plea for
future work? It says “we recommend,” at the very bottom of whatever
the last page is, in the concluding comments.
DR. CARR: The concluding sentence: “We recommend that HHS undertake
additional efforts in exploring the gaps in data collection and analysis for
the measurement of SES on health surveys to improve the understand of health
disparities.” Then, “The committee will continue to support your
efforts to identify standards in this important area.”
Let’s just take the question that Judy raised. Should this second-to-last
sentence actually be a recommendation?
DR. COHEN: My answer is that it should be and it should be expanded and be
more explicit in terms of what it is that we’re requesting. I don’t know
whether — we sometimes put in language about reporting back to us on progress
related to the specific tasks that we request be done.
DR. CARR: There are two questions. One is, yes, it should be a
recommendation. It looks like we have two kinds of recommendations, then. What
are the components that are necessary? I think that’s what the four address.
This is a second, sort of overarching — that we undertake efforts to explore
the gaps. Is that right? You could take that back to the committee.
MS. GREENBERG: It has already been recommended, here on the second page —
this is, by the way, something that has come up in both of the letters you are
talking about today. A typical pattern of NCVHS letters in the past number of
years has been observations and findings, then recommendations. But there seems
to be a compelling need identified in both of these letters to actually start
with recommendations, at least some recommendations. I think that’s part of the
I think it was suggested that there be a subtitle saying
“Recommendations.” Then I think this recommendation in this last
concluding comment could come right after these four things. The subcommittee
can meet this afternoon and talk about how to make it more specific, the way
Bruce is suggesting.
I was just going to suggest, because often people don’t really know what you
mean by standardization — I think this discussion has helped tease that out,
to some degree — that in the third paragraph, to say, “Standardization of
questions, code sets, and mode of collection allows more uniform.” Many
times I have had people say, oh, yes, we adopted the standard, and they use
different questions and they use different categories. It’s just that they use
the same data element, and you can’t compare it. There are other ways to
standardize, too, but I think, if you at least put, since we’re talking about
surveys here, questions, code sets, and mode of collection, that would help.
DR. CARR: Thank you, Marjorie.
DR. WARREN: I was just going to address that last recommendation. I think it
would work the way that Marjorie said or keep it at the end and change the
title to “Concluding Comments and Recommendations.”
DR. CARR: But I think Marjorie raises a good point. The look and feel of our
letters, I think, makes it easy for folks to know where it look. I get it that
this one was, there isn’t enough data to make recommendations, here’s where we
have to study, and, by the way, this is what we heard. I think that’s okay. But
you brought up the fact that we ought to be able to find the recommendations in
one place. But let’s leave that to the committee.
DR. MAYS: I think now, if we are going to discuss the specifics at the
Population meeting rather than now, I’ll just hold.
Len. Could you introduce yourself?
DR. NICHOLS: Yes. I am Len Nichols. I’m a frequently tardy member of this
committee and recently absent member of the Population Subcommittee. I have no
conflicts. I’m from George Mason University.
I would preface with an apology to those of you for whom I dropped off the
planet in the last couple of months. Duty calls — day job and all that.
Therefore, I’m even more cautioned by Larry’s statement of how much humility
he has acquired in this process. I hesitate to open my mouth since I missed the
hearing. But I do have a couple of what I would consider to be kind of big
They relate to, very simply and crudely, why are we doing this? It seems
like it’s a request from the Data Council pursuant to the ACA. I’m perhaps not
the only one in the room hyper-focused on what’s going to happen roughly a week
from now. Is there a way for this not to be tied to the ACA, so that, in fact,
this is something we would recommend to HHS independent of the ACA status?
Maybe even the Data Council’s request could be couched in a less contingent
That speaks to what I take to be this concern about the whole letter. It is
kind of a bubble. It is a reaction to that question. But one who didn’t know
anything might think we were, therefore, saying we don’t know much about SES
impact on health today. I think, in fact, we do. When you look at the way it’s
framed in education, income, occupation, family size, we say the relationship
between education and health outcomes is well established. We don’t say that
for the others. It seems to me that’s an important piece of the context here. I
don’t think anybody in the country doubts that income is related. I don’t think
anybody who has read the Whitehall study thinks occupation isn’t related.
It’s not that we know nothing. I say that only because I’m worried about the
context in which this thing could be interpreted if the Court rules in a
certain way, but also in the larger policy discussion. I just think we should
be symmetric about what we think we know now.
Then we could certainly acknowledge everything we just agreed to and what, I
guess, is the real source of Larry’s acquired humility: These code sets are way
more complicated across surveys than a civilian might think. You can’t just
blatantly recommend standardization across the board.
But it’s not that we don’t know that income affects outcomes. It is that
it’s hard to be precise about precisely which subgroups, because different
types of subgroups appear based upon the surveys. Is that the fundamental
question? I would feel better if we would say that is the problem we’re trying
DR. MAYS: I just want to respond. One of the differences is that there’s a
difference between the variable of education, income, and occupation, and a
difference between when it is a socioeconomic status variable. What we have
lost that we didn’t have as much comment on is about the prestige factor. There
has been a change over time, for example, in occupations. What used to be
prestigious isn’t anymore. The issue that we struggle with in terms of saying
we know about the relationship in terms of SES is that we don’t necessarily
know it as a social status standing variable. There is a relationship, for
example, between income and health, but we are not necessarily measuring income
well enough to say where it stands relative to the concept of SES. The concept
of health, yes. But the concept of SES is some of the difficulty that we have
in making such strong statements, I think.
DR. CARR: Comments on this?
DR. COHEN: I would like to respond to Len. Thanks for raising those
questions and those issues. I don’t see that this necessarily needs to be
linked to ACA, which I think would be great. I don’t know that I totally agree
with you that we know a lot. I think our country is becoming a lot more
diverse, and as inequities increase — certainly the income inequities have
increased dramatically, and that has been clearly documented — the
relationship between SES and health — although fundamentally we know that low
SES leads to poor health the specifics about that relationship are
becoming more cloudy, in my mind, and less explicit.
My goal and interest in this is figuring out a way, given the rapid changes
in dynamics in our society, that we can standardize an approach to collecting
information so that we can confirm, reconfirm, and identify new target
populations with respect to population health related to these measures.
Historically, we think we know a lot, but I don’t know whether that
knowledge base is consistent with our current reality.
DR. CARR: Jim.
MR. SCANLON: It is a good discussion. It is actually about the meaning of
the variables themselves and the analyses and the causality. A couple of
We don’t need to do a literature review on every one of these variables and
its relationship to health. That’s not what the committee was asked to do. It
arises out of ACA, but the Secretary has the authority to issue standards for
her own surveys anytime we want. We have done it before. You can finesse that
if that’s an issue.
But I don’t think it would be helpful to go back and try to compile a — we
don’t need a systematic review of the literature on the relationship of these
variables to health. We’re assuming that we can see that. What we need now —
and this is the context in which we were asked — when we asked for public
comment on the data standards we were proposing in terms of the next step, the
next area most people thought was an area that would benefit from and was more
or less amenable to standardization, the whole SES set of variables came up. I
don’t think many of the commenters knew what exactly they meant. They knew that
they needed more than race/ethnicity in those others.
What we needed from the committee — and I think the committee has done this
— was a state-of-the-art review on how federal surveys are now collecting
these data. If there are indices that are preferable to use — it sounded like
there weren’t, really — that would be useful to know as well.
But what are the components of SES, in current thinking, in survey research
now? How is it being measured now? What’s the potential for standardization?
The reason it’s not so easy to say, let’s pick a way, is because there are
surveys that actually have legal responsibilities for income and employment.
You can’t just say we don’t like the way CPS does it. You can’t just do that in
standards, as all of you know. There are ways to look at standards and
conventions, categories, ways of asking questions that are appropriate to the
nature — who exactly you are asking it of, and in what manner you are asking
it. I think that’s really what we’re looking at.
In terms of income, for example, there are starts. We can’t just throw away
the SOC or the standard industrial classification as a federal agency. We have
to look there first. With race/ethnicity, as you know, we had to look there
first. With income, there is no overall standard. There are clearly ways that
surveys ask these questions.
So I honestly wouldn’t spend a lot of time trying to establish the need to
do this and the research on these variables relating to health. You told us
what you found in the hearings. I don’t think you need to describe them.
Certainly we know — maybe I’m assuming that everyone will know as much as we
do. We know how the data is being collected in those surveys. We know what the
potential for standardization is. You have looked at it and you said there is
no obvious standard, and it would be quite a bit of work to standardize. Maybe
standardization is not even the issue. Maybe agreeing on certain conventions
would do more to help, which is a sophisticated way of dealing with this.
I don’t know that you need a lot more. Certainly any more specific
information you could give us in terms of parameters would be helpful. But I
think we certainly know the way we would go.
DR. CARR: Go ahead, Larry, and then Len.
DR. GREEN: Len, I have a suggestion for how we can potentially be responsive
here. You had two points. One was, don’t blame it on the ACA, and don’t make it
look like we really don’t know that SES is really important and that we don’t
know a lot about it. In the second paragraph, before that “in accordance
with the ACA,” we could just say we heard no dissent to the great
importance of measuring socioeconomic status, and the Secretary and HHS has
authority to determine how this is measured. Then we can say, additionally,
we’re required to do this now by these things.
Would that be all right, Len?
DR. NICHOLS: That would be wonderful. The only thing would ask, then —
let’s go back and look at the first sentence of education on page 2. I love
that, Larry, and I’m glad Justine was wise enough to have you talk before me.
That solves my problem. But I look at education and it says, “The
relationship of education to health outcomes is well established.” We
don’t say that for the others. That’s what worries me.
DR. CARR: Let’s have the subcommittee go back and discuss that — a
parallelism. If we’re going to say it for one and not the others, there has to
be a reason.
DR. SUAREZ: I just want to go back to the occupation one and the code sets.
I actually participated in the IOM study on incorporating occupational
information in the EHR. Actually, I mentioned this during the hearing. We had
NIOSH representatives here presenting. The formal, official recommendation from
the IOM in that study was the following: Adopt the standard occupational
classification, SOC, and the North American Industry Classification System, the
NAIC, coding standards. This is for use in the EHR. It’s sort of the standard
coding for occupation and industry.
Now, I’m not suggesting that we say that in the letter, because we couldn’t
get down to that level. I don’t think the hearing would allow us to do that.
But I think what we can do is incorporate a statement on page 2, in the
occupation section, after, “At a minimum, a set of two standard
questions,” say something about “and use national classification and
coding standards for occupation and industry.”
DR. CARR: Bring that to the committee. Again, there’s a parallelism. If
we’re going to go granular for one, we should be going granular for the other.
MS. QUEEN(?): I just want to mention that most of the surveys, when they do
code it, are using those standards, like the census. When they bother to use
the variables for that purpose, where they are actually specifically coding,
they usually use those standards.
DR. CARR: Okay. I take it we are not ready for a vote. But this was great
discussion. It just demonstrates the richness of the different perspectives in
the committee. It always brings one more thing of importance.
With that, I think we will adjourn for lunch.
(Recess for lunch)
DR. CARR: Welcome back to the afternoon session of day one of NCVHS. We want
to get started right on time.
We have a little bit of rearranging of the schedule. We’re going to begin
with the quality letter. Following that will be the Privacy Subcommittee
briefing from the April 17-18 hearing. From there, we will go into the data
access and use, followed by a discussion that really came out of the Standards
Subcommittee of what all the things in front of us are, how we prioritize and
how we also recognize some of the things, like claims attachments, that have
implications for privacy, for quality, for population health as well, and get
those issues addressed.
Let’s turn it over to Paul.
DR. TANG: You have a draft in front of you. Let me give a little bit of a
background. The start of it was a hearing that we entitled “Measures that
Matter to Consumers.” You’ll see that the subject doesn’t exactly say
that. It says “Steps to Improve Support for Consumers’ Health Decision
Making.” We set out to work on quality measures.
The reason we ended up with a bit of a detour — but I think it’s actually
fortuitous and it’s a lemonade play — is because one of the first things we
had was a consumer panel where it said, I can’t even think about quality
because I can’t figure out what plan to get. So in a sense, they had on
bandwidth to even consider who to pick as a provider and what to do with every
health-care decision, because they got so wrapped up in not being able to be
well informed to make a plan choice.
So we figured we didn’t want to waste our time, and we had to fix the
problem that’s in the way of their even considering measures that matter to
their care. That’s how we got into that.
Somewhere along the line — and I think Justine may have been part of it —
we said, why don’t we take advantage of things that are already going on
instead of just saying they ought to do such-and-such? A thing that’s going on
that the federal government has control over is health exchange. Can we cause
the timing of the health exchange and — that’s one of the few times when a
consumer has choice, in this health exchange. Most employed consumers have
almost no choice. Even when they have a choice, there’s such a big financial
difference that there’s no choice. In a sense, the insurance exchange has
become a real opportunity to test the notion of whether the consume can make an
informed choice about health care.
So that’s sort of the setting for this letter. I’m just going to paraphrase
Do we need to read the actual recommendations?
DR. CARR: Yes, for the recommendations, but the background —
DR. TANG: First of all, we already live in a world where there’s much more
emphasis and there are more initiatives going on in quality-related matters.
It’s the National Quality Strategy. It’s the Million Hearts campaign. There are
a lot of things going on. It’s the ACA, a lot of things coming out of the ACA.
We are using the first page to say that’s the context. What’s not going on
is, we think, giving enough good information, consumable information,
understandable information to consumers making choices about things related to
care — plan, provider, and treatment decisions.
The hearing, which was very good, had a number of findings, which are on the
second page. One that I already pointed out is that their first decision at
open enrollment is, what plan do I choose? You try to make two decisions:
What’s a better plan? How much will it cost me? Probably the way they look at
it is, how much will it cost me, and then try to figure out what the better
That became the brick wall that we had to get across before we go anywhere.
That’s why we started there.
DR. SUAREZ: Could I interrupt and ask a quick question on that level? The
experience that a lot of people have — and I don’t know if the landscape has
changed — the experience that most people that are employed have is that
employer selects one health plan and then the individual cannot choose a health
plan anymore. They choose providers within that network. It’s more of a rare,
or not that frequent, situation where an employer offers three or four
different health plans.
DR. TANG: That’s correct, which is part of the problem. The way we looked at
it was that the exchange was the first time people had a chance at making a
choice. We’re trying to take advantage of that opportunity to fix what measures
and how they are presented.
DR. WARREN: At the hearing, though, there were some people who said that
they did have to make a choice. They were given a choice between two or three
plans, for different costs, different deductibles, and things like that. It was
exactly that — how do I even know how to choose off of this list that I have
been given? Or I have been given one plan, but different tiers of the plan that
I can subscribe to. How do I know what that will do for me and my family?
It was incredible.
DR. TANG: Most people are given almost no choice. There’s a high-cost plan
and there’s a much lower-cost plan, and that’s sort of the, quote, choice.
DR. CARR: Matt, did you want to say something?
MR. QUINN: I was just going to say, the other dimension on this, too, is
that a plan that’s good for one person might not be good for another. A plan
that is good, financially and quality-wise, for a woman of childbearing age
might not be good for someone who is not. Being able to provide that sort of
information is just not happening today.
DR. TANG: Continuing on, it wasn’t just cost. What’s the total coverage, in
a sense, for me? It really is, what’s my out-of-pocket cost? What people really
wanted to know is, for me in my stage of life, what’s the predicted cost for
over this next year? If next year you are planning to get pregnant, there may
be a different plan that’s better for you. What is the cost to me for this year
of my life? That’s what they would like to have, and that’s clearly not what
Another thing is quality information, the so-called quality measures. How
can I actually understand I would like to know, with my
conditions, what measures I should even be looking at. The problem is, even if
you know that — let’s say you are a diabetic — you don’t really know how to
interpret the measures that are being given to you. They are not measures that
matter to the consumer. Do I really want to know what my A1C or glucose is
going to be or do I want to know my chances of going on dialysis, when I choose
this plan or this provider?
It’s clearly the latter. When I’m going to make a decision about joint
replacement — well, you virtually get no information — what I would like to
know is, can I walk again? Will I be in pain? How do the other people from that
surgeon fare? These aren’t available.
If there is some information available from some survey, it’s usually not in
a standard way so that you can combine it with other information, so you can
compare the two surgeons, for example.
Finally, we heard from physical therapy. Most available ones are
proprietary. Although they can do a research project, they can’t actually
afford to do it in operations.
You can hear that they — “they,” consumers and purchasers —
really want better information that matters to them and that ideally would
matter to them on a personal level and that wouldn’t cost a fortune in license
fees. That’s sort of the finding, if I were to try to distill it.
Along with that, how would you go out and design a measure or certainly
design the presentation of this information in ways that would be useful to
consumers? Asking consumers is a good place to start. You could use
user-centered design principles so that you not only ask them in the design
phase, you test it with them.
There are thoughts that — they are straightforward and done in other
domains, but they certainly are not applied in this domain. We’re not getting
the outcome that we want.
Taking it further, moving towards an ACO, then you would like to know more
about the community so you can incorporate that kind of information, whether
you’re an employer looking for, what plan should I get — wouldn’t it matter
what communities your employees come from? You can see how this actually pulls
in a lot of our past work as well in terms of community health data. It just
becomes a very rich area to produce understandable, pertinent information that
doesn’t cost a lot of money and can be standardized and aggregated. That’s the
Our recommendations start on page 3. It says there are a lot of people who
might even be motivated and intend to either develop or use some of these
measures, but because it’s really not well done at this point, we’re suggesting
that HHS actually create or provide technical guidance to support developing
measures with these kinds of attributes, to support development and
implementation and presentation of the information that comes across. That
guidance should ensure that the measures and tools used — and here’s where we
use the federal lever — most of the insurance exchanges will be either run by
or contributed to by the federal government — partially sponsored, I’m sorry,
by the federal government — then the federal government has a say.
Could the federal government, as part of the health insurance exchanges,
ensure that the measures that are used to communicate to consumers comply with
some of these attributes: That they be usable to consumers, that the
presentation include formal usability and validation testing prior to release
and ongoing feedback. That’s one recommendation.
Another thing HHS could provide is to fund additional research on how to
either help understand what consumers need or help consumers use what they
The third piece is research on measures of patient-centered care. How do you
develop measures that pertain to individuals?
The fourth one is training materials and potentially software that help
consumers assess the information that’s coming at them to make more informed
Those are four recommendations for HHS to help the industry get towards
The second set of recommendation is to essentially start creating model ways
to communicate with consumers using its federally sponsored health insurance
exchanges as a place to have some of these model reports. The first part of
that recommendation is that the exchanges have standardized information and
tools that help patients make these choices. An example is nutrition labels,
where every food product has to have this kind of disclosure. Could every plan
have to fill out these kinds of information and it be available either in
presentation format or on a tool that says you enter in your problems and your
age and so forth, and then you get back standardized kind of information, which
could include out-of-pocket costs.
The second part is, once a consumer does get beyond the plan — as they are
assessing which plan to choose, they are obviously interested in what providers
are associated with the plan. Those things should be linked. Then we prescribe
a minimum amount of information about each provider, like where they practice,
what their office hours are, what the language is, what the quality measures
are that pertain to them, what the patient experience of care is, and other
characteristics, such as out-of-pocket costs, things that pertain to the
patient at the provider level.
The third recommendation in this set is to essentially take advantage of all
the information that the federal government already has, whether it’s PQRS,
whether it’s HCAHPS or GCCAPS(?) — take that and create model reports that
display that to consumers so that they can have that when they are looking at
any plan or any provider. Again, what we’re leveraging here is the fact that
the federal government is going to sponsor these exchanges, so the federal
government has a role in dictating what is made available through those
exchanges. The principle of the exchange is to give consumers choice, in
addition to having a set of plans to choose from with a different cost.
The third set of recommendations has to do with encouraging collaborations
among entities that both develop and endorse consumer-oriented measures of
health and health care, so that they do meet these attributes of being
relevant, well specified, publicly available, meaning nonproprietary, and
leverage consistent value sets.
Let me just read the recommendations.
Specifications for consumer-helpful, patient-centered measures should have
clear details regarding terminology, numerator and denominator construction,
value sets, metadata so that measures can be understood and implemented
consistently — well-specified measures, that is.
Secondly, the measures should be transparent. In other words, promote the
public availability and use of consumer-helpful and patient-centered measures
to the instruments.
Third, the endorsement body should incorporate these kinds of specifications
or criteria in their endorsement criteria for measures particularly that
pertain to consumers.
Fourth, additional standards development is needed to represent the
consumer-oriented quality measures.
I think some of these things probably can be collapsed.
Fifth, consumer-oriented measures should be developed and tested according
to these principles of user-centered design.
So it’s basically saying, for the people who develop the measures, for the
people who endorse them, for the people who test them, give a set of principles
that make sure that consumers’ interests are aligned all along the way.
We started out looking for quality measures for consumers to choose
providers and found out that there other things that would have to be knocked
down if you’re going to get there and also uncovered the fortuitous opportunity
we have at this moment in time when we’re going to stand up these health
insurance exchanges throughout the country that the federal government has a
big role and a say in. Could we move towards model ways of communicating to
DR. CARR: I just want to comment. I think the way the hearing was held and
then the opportunity to synthesize what was heard immediately afterward — this
is an example of how helpful it was. We came away with something actionable
that wasn’t exactly the question we asked, but informed the collective
deliberation. I thought that was really excellent.
DR. SCANLON: I think this is an incredibly laudable goal. The question is,
how are we going to get there? I don’t know how much you discussed at the
hearing the federal employees’ program. I think, in some ways, it may be at
this point the state of the art in terms of doing this. At the same time, given
what you said are the objectives, you can talk about the limitations and where
they are in terms of this undertaking.
The reason I bring it up, in part, is that I happen to be in it and I happen
to be in Blue Cross. But when you go to the information they give you in the
standardized form, this is the Blue Cross option for the country. Like
politics, health care is local. So there is the issue of what your experience
is going to be. That goes back to what you talk about in terms of providers
that you are going to be able to access.
The exchanges — we don’t know how they are going to operate. We could have
statewide plans in a lot of places or curious configurations of geographic
areas in a lot of places. I think it would be useful to emphasize in the letter
this whole issue that we have to get particularly the provider information down
to some kind of community level that is relevant to people. When you go in
there, you put in your zip code and they tell you who is within 25 miles of
you. Even putting in your city, your metropolitan area, doesn’t really help in
terms of true access to provider types.
This is incredibly laudable, and you can talk about the federal role, but I
can also tell you that there is a lot of information that currently isn’t
flowing that is going to have to be assembled to make this happen. I think we
have to underscore that this is important, but also recognize that there’s
going to have to be a lot of attention to making it happen.
DR. TANG: I think you are right. First of all, I think your idea of working
with the FEHB is a good idea. It’s the same principle, but, as you say, it has
a longer track record and applies to all the federal employees.
The second is the plug-in. We do talk about the tools, and we even describe
software tools for literally doing exactly what you said: Here’s everything
about me, here’s my zip code, what can I —
DR. SCANLON: I guess where I was going, in part, was to emphasize — I don’t
want to know what a plan’s network is, because the network could cover the
state. I need to know, where I live, who the people are in the network that I’m
going to be able to access.
MR. QUINN: The other is that often people want to choose a plan based on
whether their doctor is in it. Just like I want to choose a drug coverage
option based on the drugs that I’m on, I want to choose coverage based on
whether my doctors are in it.
DR. TANG: That’s what we intended, Bill. I see what you’re saying about
DR. CARR: And it’s about keeping care local, too.
DR. SUAREZ: I was thinking of a number of regulations and activities that
happen on insurance reform and insurance exchanges. The Office of Consumer
Information has put out a lot of directions and guidance and new ways to direct
and explain to entities that are going to be setting up these information
exchanges how to provide information to consumers and what kind of information.
There’s already a lot of that. I wasn’t in attendance at the hearing. Was that
all taken into account? There’s already a summary of benefits that are defined
now and have to be provided. There are a number of regulations that already
push in the direction of standards.
What I was trying to look at was how this can advance that level of effort
that has already been under way. Or is this sort of one step behind?
DR. TANG: I think it’s at the NPRM level now for the insurance exchange.
Those were obviously not done when we — and I don’t know whether it’s ahead or
behind. It’s a little too bad that we might be missing the window, unless we
can get this letter out tomorrow and it can affect the actual final rule.
DR. CARR: We will get this letter out tomorrow.
DR. TANG: Thank you, Madam Chair. But our intent was to influence exactly
the kinds of regs that you are talking about. If there is any kind of opening,
they could consider this and maybe create an opening that they rule on later
when they issue more regs on the exchanges.
It was just trying to capture an opportunity, if it’s still possible. And I
like what Bill said about the FEHB.
MR. SOONTHORNSIMA: I think we might be a little ahead. To his point, it’s
true, as we see the standard benefit description that you have to do, that
tries to spell out the benefits, what you are getting for the plan. But this is
different. The second recommendation, which is a really good one, is to
describe also the doctors — let’s be very specific, and to Bill’s point, this
is very local the doctors that you are either using today or
thinking about using. So this is even beyond what is being prescribed today
With that said, I think it’s a good idea and it’s a good goal, but I wonder
if we might be too prescriptive. When I read this — for example, when you say
HHS should develop model report formats for use by federally
supported is that at the state exchange? Some states are falling
back to the federal exchange. Or do you mean both?
DR. TANG: We thought wherever the federal government had a role. When the
states do relinquish and the feds come in, they can dictate.
MR. SOONTHORNSIMA: You might spell it out. “Support” means
different things to different people. To me, “support” means I get a
subsidy. That is a way of support, financially.
But maybe even more prescriptive here is point B, where you are really
saying that, even before you select a plan, regardless of what metal you are
using, you will be able to look at the provider directory. It sounds good on
the roadmap. My concern is that these health insurance exchanges are not
trivial. These are really difficult to stand up, to begin with. There’s a lot
of navigation to go through, coordination with the Social Security
Administration, so on and so forth. Then the enrollment process — new process,
So I wonder — instead of being very specific, very prescriptive, leave the
notion out there saying that if you are going to participate on the exchange,
this type of information should be provided down the line, beyond what has
already been asked for in the law.
DR. SUAREZ: And part of the reason for the point I made is that I don’t want
people who will receive this letter to say, check, this is done. A number of
these things are already done. The reason I’m saying that is that a lot of
health plans have already the information about their network. The question is
— if they are going to participate in insurance exchange, they should provide
MR. SOONTHORNSIMA: But don’t be too prescriptive about how they provide it,
because it’s hard enough. For example, in order for you to see the doctor in
your local community, a lot of these plans already provide the information
where it’s available. That’s why this is a good one, because not all plans do
this today. So this could be that glide path that we’re trying to help direct.
DR. TANG: But we are also saying that what consumers told us is that the
quality information that they do see doesn’t always pertain to them, and it’s
also not necessarily that useful. A diabetic, as I said, doesn’t necessarily
want to know what your glucoses are so much as they want to know, how many
people go on dialysis — something that’s more meaningful in their lives.
MR. SOONTHORNSIMA: Wouldn’t it make more sense for us to focus on what the
types of measures are that should be — I think that’s in here.
DR. TANG: That was the original intent of the hearing, actually. Then we
uncovered a lot of —
MR. SOONTHORNSIMA: But if we haven’t resolved that, why would we want to
prescribe it in the HIX? Maybe that’s not the intent. It says that health
insurance exchanges should provide information on the health-care provider
networks available. When you say health insurance exchanges, that’s very
specific, when, in fact, the health insurance exchanges, for the most part, are
your portals, the brokers. That’s very prescriptive. Now you are getting into
another HHS role, where they have to design how these HIXs are supposed to
DR. TANG: I guess, to that word, we could have said “contain.”
Right, the exchanges don’t have the — they are just the broker. But they can
require, in order to participate, to be listed on the exchange —
MR. SOONTHORNSIMA: That’s exactly right. Keep the spirit of what you’re
trying to say here, but don’t prescribe.
DR. SCANLON: We have a history here where we have gone from — when Medicare
Advantage began or Medicare Plus Choice began, you did go to the plan
individually to get the information. It was unbelievably difficult to get
information to compare plans. Eventually, CMS took that function on and they
started to mimic what FEHBP does in terms of standardizing information, and we
now have Plan Compare. These exchanges are not fully defined yet, and they can
have the capacity to say, here’s the kind of information, here’s the format we
want it in, and we will publish it. They are going to be self-sustaining in
terms of the fees that they charge. So I think we should hold that out as a
The issue is how far we can go. You were talking about what you want to know
about your doctor. There is the issue that if the doctor doesn’t treat enough
diabetics, all kinds of measures become “NA.” The development of
measures is still a challenge.
DR. CARR: When we had this conversation, we talked about the aspirational —
what are the kinds of measures we would like to create? But what’s there — how
many patients in your panel that have diabetes is already helpful, predictive
in some way. I think that was the theme, just to take what we have and
aggregate it, juxtapose it in a way that leads to some level of assessment.
MR. BURKE: One of the items in the major findings talked about cost. The
consumer’s principal barrier about what it will cost is a combination, as you
know, of premium, which is clear, and out-of-pocket, which is less clear. For
each plan, there is typically a series of products within that plan, some of
which have different levels of cost sharing — in fact, all of which have
different levels of cost sharing. Some of them may have different levels of
provider participation which lead to that premium.
One of the ways to — and this may be an additional recommendation, in terms
of what it will cost me — to encourage HHS to develop something like a
treatment cost estimator which would permit the consumer making this
independent decision about what product to buy let alone what plan
to purchase — to imagine, I’m going to have two visits to my endocrinologist
and I’m probably going to have an emergency room visit in there somewhere.
Based on the characteristics of the product I choose and this likely course of
experience in the coming year, what can I anticipate in the way of
co-insurance, co-pays, deductible, impact, so you can model not only which
product to select, but what’s going to come out of my pocket based on that.
DR. CARR: Actually, it can drive behavior, too. You could say, for my
condition, if I go to an urgent care center or I go to an emergency room or I
go to my PCP —
MR. BURKE: We call that skin in the game.
MR. SOONTHORNSIMA: That is done at the plan level, not at the health
exchange level. That’s the point.
DR. COHEN: I guess the point we’re trying to make — we recognize that it is
done at the plan level by some plans, not necessarily all plans, but the
exchange is the place where that information should be collected for all plans
so that individuals can compare each plan as they need to make decisions. The
insurance exchanges should have the capacity to provide this information for
all plans. We’re not saying that the exchange needs to go out and reproduce
information that already exists. Everybody needs to give to the exchange this
information in a standardized fashion so consumers can make better choices.
That’s the whole intent here.
DR. CARR: Walter is going to take us through this website.
DR. SUAREZ: The point I was trying to make is basically that this is already
in the regulation. What we want to avoid is writing a letter with
recommendations that are already in the regulation.
This is a page from CCIO, the Center for Consumer Information and Insurance
Oversight, a summary of benefits and coverage and uniform glossary. It says
basically that specifically the regulations will ensure that consumers have
access to two forms that will help them understand and evaluate their insurance
choices. One of the forms is an easy-to-understand summary of benefits and the
other one is a uniform glossary of terms. Here’s a sample of the form.
The concept is that there are a number of things already being included in
the regulation that will point to the same recommendations that we’re making.
What I was trying to say is, how can we advance one step beyond, and not be
behind, in terms of writing a letter that already covers what is in the
Here’s the sample form.
DR. TANG: Can you read out something that is duplicative with our
recommendation from your form?
DR. SUAREZ: This is the form.
DR. TANG: We can’t read it.
DR. SUAREZ: Oh, I’m sorry. This is the actual sample of the form that is in
the regulation, expected to be the one used to provide that information in the
insurance exchange. The questions are: What is the overall deductible? So this
is the payment part. Does this plan use a network of providers? Do I need a
referral? There are a number of those kinds of questions. This is one example.
Beyond that, of course, there is the question in the letter about — or
perhaps the point that can be made is that in addition to the things that are
already expected to be in the regulation or have been written — actually, I
think the regulation is already a final rule.
PARTICIPANT: For that one. There are few more implicit here.
DR. SUAREZ: The point is, in addition to the basic things that are already
in the final rule that are expected by health plans to be provided through
insurance exchanges, we recommend expanding the information to other areas that
perhaps are not covered here. They are the kinds of things that are maybe noted
in the recommendation.
DR. CARR: I think that it goes back to Larry’s comment this morning about
humility. There are a lot of great ideas going around. I think we want to be
not naïve in the fact that people are thinking things along the same
lines. On the other hand, looking at that, maybe the building blocks are there,
but the functionality, I think, is the piece that we’re talking about.
MR. QUINN: One of the things that struck me as Walter brought this up — the
website says, provide an easy-to-use summary of this. The question is, who is
saying it’s easy to use? Would my mom find that easy to use? Has it been tested
to demonstrate that it’s easy to use for the intended populations? That’s all
this talk about user-centered design in this. It’s starting with, who are the
people who are trying to make these decisions, and designing tools that work
for them, and knowing that they will work for them, and starting from there as
opposed to this.
DR. CARR: Walter, thank you for pointing that out, because we don’t want to
look naïve. But I think you’re right, we’re getting to the intersection of
what we are saying is something that is user-friendly, patient-centric that
collocates the information in an intuitive flow of thinking that would align
with how decision making occurs.
DR. SUAREZ: Just a point. I like very much the recommendations. I think it’s
probably more in the context that we understand that there are all these things
already being done, but we believe that it will be valuable to have additional
research to help understand how consumers — these kinds of things that help
determine whether this, which is supposed to be easy to use, is really easy to
use. It’s really on that line.
DR. CARR: It’s getting to the finish line. It’s kind of what we heard
yesterday when we heard about end-to-end testing. We kind of do it, but we
don’t maybe always get to the finished line. I think if the finish line is
defined in terms of the patient’s experience or the member’s experience, this
letter gets there.
DR. WARREN: I think we need to remember that we heard from some very
informed patients at the hearing, and they did not find this. This information
was not available to them. Several of the people that testified had chronic
illnesses for most of their lives. They now are patient advocates because of
their own experience of not being able to find this information. Even though
the information may be on the Web, it does not mean that people can find it or
know how to use it. I think we need to be appreciative of that.
DR. CARR: How new —
DR. SUAREZ: This is the final rules for the insurance exchanges, once they
get set up.
MS. GREENBERG: It should be. It isn’t available yet. The question is whether
this rule, if this is implemented, will then be available and what we can add
to that. You’re not saying this is now available. You’re saying —
DR. SUAREZ: No, no. I am saying this is what the regulation —
MS. GREENBERG: Requires in the future.
DR. CARR: So it’s timing. Was this available two months ago?
DR. CARR: Judy’s point is well-taken, too. There are pockets — silos, we
might call them — of information that are known to some people very well and
not to others. As we try to integrate — the whole point of this is to be
integrative, to think in terms of the end user. I think it’s an exact
playing-out of what’s actually happening.
MS. KLOSS: Two comments. One is that certainly all evolution of this won’t
stop at the end of September. I think putting forward something that is
forward-looking can help this evolve over time.
The second — and I think it’s innovative and timely, and I certainly am in
support of the letter. I had another question. I’m sure, having just labored
through this letter, you probably don’t want my second question. Isn’t there an
opportunity for follow-on and to return to the original question, which is
laying out some of the characteristics for consumer-centered criteria? Does the
subcommittee see a phase 2 of this, returning to really 3A, expanding the
specifications for consumer-helpful, patient-centered measures, and perhaps
doing more work on that as well, which may not directly relate to exchanges or
health plans, but may raise all boats as it relates to a lot of the public
information that is being put out there that’s less than helpful?
DR. TANG: I think the answer is yes. We thought about this as phases. In
this first phase, we wanted to capture the timing. We ran into this brick wall,
and so we wanted to blast through that one. We did some of the “What Would
Be Useful in the Community” data hearing, the functional status stuff. We
had it back in the very first time, when we did the quality measures. So we
have had some of this conversation before. I’m trying to remember what was in
Of course, other committees are working on this as well.
The notion that there are better measures, I think, is out there. It doesn’t
mean we can’t do that in another phase, as you suggest. We just saw a bit of a
one-time breakthrough in the way we brought things together. What would you
change about what you are even intending to do?
MS. KLOSS: It seems like there was learning that perhaps isn’t captured here
because of the focus.
DR. CARR: I have Bill, Len, and Bruce.
DR. SCANLON: Part of what I was going to say is what Linda was talking
about. This is a goal. I think if the letter doesn’t go out tomorrow, it’s not
going to influence anything, because what we’re talking about is a much
longer-term goal. The work that I was involved in at GAO that was looking at
plan comparison started in about 1995. Things were incredibly primitive then.
We’re much further along now. We’re still sort of in a very primitive state.
I think in terms of the issue of not looking naïve, acknowledgment of
this rule would be one thing. To Jack’s point about cost, there is this part of
the exchange plans which is that they have to meet certain actuarial tiers. We
have the metallic plans — the bronze, silver, gold, and platinum. Actuarial
equivalence tells you something, but leaves a lot not explained. You can have
very different experiences between two actuarially equivalent plans, and two
individuals within a single plan can have very, very different experiences.
Even though their total spending might be the same, their cost sharing might be
different. In part, it’s an issue of network. In part, it’s an issue of
So there is that thing which is — you’re talking about bringing this back
down to the personal level, and that’s not where we are in any shape or form at
DR. CARR: Len.
DR. NICHOLS: Let me first apologize to Paul for not showing up for the
meeting. At least for this one, I did help plan it.
This may sound parallel to what I started with this morning. Do we have to
link this to exchanges? Shouldn’t every American have access to this? Shouldn’t
our role be to say, as a national committee, all Americans should have it, and,
therefore, ultimately, it should be required of all plans, period? Then leave
it to HHS, although we certainly want to take advantage of this current
opportunity, to decide whether to use the exchange regs or other authorities
that they actually possess outside of the ACA. The Public Health Service Act
comes to mind. We’re debating this next week, and I won’t belabor the point.
But the federal government can regulate interstate commerce. Insurance is
interstate commerce. Whether or not the mandate is, is an interesting question.
But insurance will always be. They delegate its regulation to the states out of
convenience. The decision in 1944 led to a law in 1945 called
McCarran-Ferguson. They can take it back, and they have, with the HMO Act,
HIPAA, et cetera.
My point simply is, we should not aim so low as to link it only to
exchanges. We should say all Americans should have access to this. I think this
is the time to do that.
Then you could say, yes, you could use the exchange regulation process. To
Walter’s point, the reg he discussed does not speak to the consumer need. The
consumer wants to know exactly what Bill said: How much is it going to cost me,
being that I am a diabetic? In fact, I believe it is the case that the ACA
requires HHS to give a reg to the state exchanges on creating a cost calculator
as part of the condition of being a functional exchange. That cost calculator
is supposed to precisely answer the question: What would it cost me to be in
this plan, given what I know about me?
It seems to me that we should be a little more explicit about cost
calculators, if that’s what we’re talking about, and second, we should be more
general in our application of this to plans outside the exchange.
DR. CARR: I saw a flurry of hands go up. Bruce is up next. Bill wants to
DR. SCANLON: If we do do that, which is a potential option, we have to
change the recommendation because we have ERISA plans. They are exempt from
HHS, the regulation. We would have to recommend that the Secretary seek from
the Congress either statutory change or we write a letter to the Congress,
because there is the issue of a large group, 160 million people.
DR. NICHOLS: We should ask the Secretary to share the letter with the
Secretary of Labor. How’s that?
DR. CARR: We’ll ask the subcommittee to take these excellent insights back.
DR. NICHOLS: The federal government has power. They just choose not to use
DR. SCANLON: A major portion of that power resides in the Congress. To do
the kind of change you are talking about requires statutory change.
DR. CARR: I am going to take the chair of the subcommittee to take back
these incisive comments.
MS. MILAM: My comment is the same as Bill’s, and also that I’m not sure of,
outside of Medicaid, where HHS regulates plans.
DR. CARR: Bruce.
DR. COHEN: This is a theme that we have all been focusing on, the notion of
patient-centeredness, consumer choice, how individuals make decisions. The
context of this particular letter — the first meeting I attended, actually,
when I became a member of this committee, was talking about a scenario — I’m a
65-year-old and my PSA just went up. What do I do? What plan is best for me?
How do I choose my doctor? What are my costs? I’m the mother a juvenile who has
gained an enormous amount of weight. What do I do in terms of making decisions
about choosing plans, choosing providers, understanding the information that
providers give me in terms of my choice?
This is the context for providing general information. I think we always
felt that the next step for us was to develop more specific use cases, stories.
One of the things we heard very clearly in the testimony was that numbers don’t
work for everybody. People make decisions in a variety of ways. What was really
important to many consumers was how — I want information about how other
patients felt with their experience with providers. I want stories that I can
relate to personally.
The goal of providing quality information to be more consumer-focused
encompasses a variety of threads, and we thought this provided the context for
future work in this area.
DR. CARR: I’m looking at our time. We have about five more minutes. I have
Larry and then Matt.
DR. GREEN: A new topic at a different level. I want to draw attention to the
first paragraph in the final comments of the letter. It seems to me that this
paragraph needs a little tweaking and adjustment in two ways. One is, in the
first sentence, suddenly we’re back to talking about reporting of quality
measures that matter to consumers, when, in actual fact, the letter is more
about supporting decision making and reporting measures that help people make
decisions. It just seems to me that that doesn’t quite do justice to the
In the second one, where it says “influence the design of health
insurance plans,” I don’t think we’re writing about redesigning health
insurance plans. We’re talking about redesigning reports that health plans do.
I would just like to see that adjusted a bit.
DR. CARR: Matt?
MR. QUINN: A broader point and a specific point. Bruce brought up a really
important point about the story and the narrative and how people make
decisions. We heard over and over again about the effectiveness of sites like
Yelp and online review sites, because they tell stories. They don’t just show
numbers. They tell stories that people can read. We heard from folks at Press
Ganey who said that’s the new generation of these tools, providing that
information and then analyzing it for some quantitative metrics. That’s how
people make decisions.
The bigger picture is that these insurance exchanges are going to be the
public face of health reform for many, many people. It’s important that they be
usable and that they don’t just meet the word of their intent, but they are
actually meeting the spirit of it. Whatever we can learn from the drug benefit
rollout we can apply here. This is more complicated, in many ways. I think that
this is a real opportunity, and we shouldn’t lose it in that regard.
DR. CARR: Paul, do you want to summarize the takeaways that you heard and
make sure that we got them all?
DR. TANG: I will make a comment about whether the subcommittee was
naïve, because I think it was actually the opposite. Not only was the
subcommittee not naïve about the rules that are required to set up health
exchanges, but that’s the whole basis for the recommendation. The only thing
the subcommittee was naïve about is what the timing would be, since nobody
is allowed to know that until it comes out.
I will also comment a little bit about what Len said. In some sense, the
uniqueness of either the committee’s opinion or the letter is not that the
measures that are out there are not good. Everybody says and knows that. What
we were trying to do was find something that was, one, a unique contribution,
and two, actionable. Clearly we intend for these recommendations and the spirit
to apply to all plans, all measures, et cetera, but we found a unique
opportunity to do something actionable, if we got it in time. That’s sort of
why we left it there, in a sense, narrowed it that way.
One is, there is another opportunity where the federal government has a big
role. That’s with the federal employees’ health plan.
Two, we want to emphasize — actually, it’s wording changes to make sure
that, whether it’s the provider, what providers are in that particular plan, or
what quality measures you want to know about, it’s all very local.
We want to emphasize more — because I think it may have been lost a bit in
this conversation — that it’s the process that we are worried about. How do
you engage consumers along the process, including the testing, so that it is
understandable? That was a major point. Probably we didn’t emphasize it enough
in the letter.
I think Larry had some really great suggestions in those final comments
about what we really meant to say.
Did I miss some other things? We talked a lot about whether it’s in the rule
or not. But that was actually the point. We want it to be in rulemaking over
DR. CARR: Another fabulous discussion and deliberation. Thank you. Paul, we
look forward to the letter tomorrow.
We’re going to go out of order now and ask Linda to give us an update on
MS. KLOSS: Thank you. I am reporting also on behalf of my co-chair, Dr.
Francis, who couldn’t be at this meeting, but will be with us via telephone at
the subcommittee meeting tomorrow morning from Portugal.
The Privacy Subcommittee last November did a pretty comprehensive review of
the work of that subcommittee from 2005 to the present and came up with a short
list of focus areas that it might pursue over the next 12 to 18 months.
Discussion brought it down to a priority of extending the work of the community
health data report, specifically the recommendation under infrastructure that
called for a privacy and security framework to guide communities in using local
That led to a hearing on April 17 that was titled “Next Steps for
Community Data Use.” Like all of the experiences we have heard about
today, this was another one of those mind-expanding days of hearing, around
four topics. We had a panel on “Beyond Data Use Agreements,” calling
for testimony on emerging governance models and the range of governance models,
a panel on “Protecting Small Groups,” a panel that was entitled
“Using Results to Improve Community Health,” but when you distilled
what was presented, it was an array of partnership models or trust
relationships that were used in a variety of contexts. The fourth panel was
I think we were all struck by the range of participants at this, not even
all from health care. We had Native communities. We had refugee immigrant
communities, genetics projects, research ethics, anthropology, community-based
research practices, bioethics, an individual who specialized in community and
university partnerships. So we really took this learning from work outside of
health care, including someone whose area of specialization was relationship
I think if I could distill it down, we kind of went into the day thinking
that we were looking at a privacy framework for this context and we kind of
came out of the day realizing this was privacy plus stewardship. It was
something quite different than privacy as we understand it in the HIPAA privacy
rule context. Essentially, it reframed our thinking, I think, in four key ways:
• That we needed to look at this not from an individual view, how I
feel about my data being used in a community context, but as a societal and
community view, and that that was quite a different way of looking.
• There was a concept that was brought out over and over again during
the day that there was a chain of trust. Trust, at the end of the day, was the
most important issue that everybody who testified began with or emphasized in
• We heard about models for participatory governance. We heard concepts
of tiered release, where there was a different issue of how the confidentiality
of data was handled when it was data about the individual versus the broader
community, that the community itself had an interest and a stake, apart from or
beyond the stake of the individuals.
• We also had a key concept that was brought out with regard to needing
to look at that data, not as data for one study, but looking at more of a
When we began to mull over what we do with this, what we moved at was
preparing a letter, which is now in draft 3 form — we thank all the members of
the subcommittee and the staff who have participated and added great comments
to get it to where it is today. We’ll work on it in person tomorrow morning.
Essentially, it has come to a letter that sets out a framework and
recommendations for stewardship of health information when used by communities
to improve local health. We have a stewardship framework that has emerged from
some of those principles that we heard — it’s a six-point framework — and a
set of recommendations for NCVHS.
Bottom line is that this letter will be coming, I presume, to the committee
for action in September.
One other point to make, and this is a matter of timing, as we just
discussed with regard to the quality letter. Several of our subcommittee
members who were just at the Datapalooza realized that some of the issues that
we’re discussing here are broader than the community health data initiative
lens of our report. We looked at how to broaden it, but not blow it wide open.
I think there’s a balance there. We have acknowledged that HHS is leading
efforts to encourage innovative uses of health data and certainly are strongly
in support of that. But at the same time, the committee understands from this
analysis that it may be time to develop and promulgate guidance on stewardship
practices for use of health data outside the protections of the HIPAA privacy
rule. While the letter is framed around this community initiative, there are
implications of this letter that go beyond, and we’re walking both of those
forward in the draft of the letter.
Again, thanks to everyone who has contributed to this. We had two calls and,
I think, over the course of a week, really evolved this quite well.
With regard to format, there is so much work that the committee has done
before that certainly informs this. One of the suggestions was made to include
some of that, but do it as an appendix.
I think the other discussion that has come up is that perhaps this letter
then leads to a more formal committee revisit of the 2009 primer on health data
stewardship, because when we reflected on that and looked back on it, it’s very
individual-centered and certainly is consistent with the HIPAA privacy rule,
but it may be time to look at that and re-advance that against a new framework,
a new model.
That’s my report. We’ll look forward to continuing this discussion.
DR. CARR: Very exciting. Two things. One is that part of the charter for
the new workgroup on data includes issues of privacy, so I think that will be a
perfect kind of segue into that.
The second is, you said that the six principles that are developing — do
you want to just name them?
MS. KLOSS: I’d be happy to. Stewardship framework, participatory governance
— and we have defined that as, the rights and responsibilities with regard to
how data are collected, managed, and used should be made explicit — and then a
series of bullets under that that speak to sponsors or community leaders for
these kinds of projects.
The second is accountability. It’s there that we have woven in that concept
of not only accountability for this study, but accountability for the lifecycle
of data. The complete lifecycle of data should be managed, from initial
collection and use to dissemination of results and, if appropriate, the safe
archive or disposition of raw data. Then there are a number of bullets under
that, again trying to speak to or frame the responsibilities of sponsors of
these kinds of studies.
Third is openness and transparency. Again, the rights, responsibilities,
practices for data collection, management, protection, and use should be
available to all stakeholders, most importantly to those who are subjects of
the data, that tiered issue, and then bullets under that.
Integrity: We did get into the issue of trust in the data and the trust in
the relationships and the roles of the people who are handling the data. That’s
kind of a multilayered set of principles, both data integrity and process
integrity, if you will.
Five is privacy, confidentiality, and security, the core practices that we
would expect sponsors to have in place, or stewards.
Six is consistency. This is a concept that care be taken to ensure that
effective practices are consistently followed. Thus, stewards are encouraged to
put in place their chain-of-trust framework. There are some bullets under that.
DR. CARR: Great. That’s good.
MS. KLOSS: Special thanks to Walter, who identified out of early drafts of
the letter seven principles. I actually mashed two of them together.
DR. CARR: I just pulled out the primer. You’re right. There were four. One
was about individual rights, but the second one was responsibilities of the
health data steward, where there is alignment — I like the way you have done
it — and then security and safeguards, then accountability, enforcement, and
Have we had any intersection with Joy Pritts on this? Hearing this morning
all the work that is going on there around privacy, governance, and all that,
it would be great to share that or get some feedback.
MS. KLOSS: I did take note of her mention of that in-person meeting on July
11, and I will share that with Leslie, as she is our representative to the
DR. CARR: I think what I heard today is that they want to know what we are
working on. We need to push stuff out to them so that again, in all humility,
we all know what we’re all working on, and if there are connections there, we
ought to be making them. That’s very good.
MS. KLOSS: Any comments from others who were at the hearings or
DR. CARR: Jim, just to update you, we just heard a nice summary of the
hearings that were held in April on privacy. Although they were talking about
the community privacy initiatives, as we move into a larger space, these
principles are going to be addressed for the larger group. That’s a good
intersection with the new data workgroup.
MR. SCANLON: That will fit in very nicely with the new group, the data
DR. CARR: With that, why don’t we jump right into this?
MR. SCANLON: I think you are all familiar with what we’re asking the new
group to focus on. It’s kind of a hybrid group. It has at least a third of the
members from the full committee and then remaining members who are consultants,
following the model of the Health IT Policy Council and the Health IT Standards
Let me give you a little bit of background.
We talked this morning about some of the efforts that HHS is making to get
the data that we do have out beyond our usual research community, statistics
community, and so on, to plans, providers, consumers, patients, community-level
analysts more generally. Many of you already are familiar with the kind of data
that we have. It’s useful for many things, but it’s not a cure-all and it’s not
a panacea. You can only do certain things with it.
There is a lot of data that we have that is of the indicator type,
statistical information or estimates in tables and so on — percent of
uninsurance, for example. We have other data that is claims data or individual
record data, like Medicare or Medicaid, Ryan White, and a couple of other
programs, client-level data. But again, we don’t release that except for
research and statistical purposes or in some of the instances we talked about
this morning relating to ACOs, where there are some provisions allowing them to
get access for certain kinds of information.
The other kind of data that we have — it’s not the data that the committee
deals with normally — is operational data or administrative data that our
agency might have. It’s not really quantitative data. It’s not that you can
analyze it statistically. It’s more location-based data. In other words, it’s
where all of our community health centers are, for example. It’s where all of
our maternal and child health resource grantees are or family planning or
mental health and substance abuse, or all of our NIH grantees or all of our
other grantees. So it’s administrative data. It’s largely public data. It’s not
the kind of data you would analyze statistically, though certainly you can. But
from the point of view of geographic kinds of information systems and maps and
location-based apps, that’s the kind of data that I think folks like.
We have, for example, one app — and this is data only in the sense that
it’s location-based and it might be useful to individual — we have one app
that tells you the location of the nearest cigarette smoking quit line. You’re
walking along and you’re thinking you are going to try to stop smoking. It’s
based on your zip code, and it tells you where to call. We only have four of
them, so it’s an 800 number. Nevertheless, you think it’s location-based.
There are all sorts of things like that. Where is the nearest community
health center? What I do for mental health crisis kinds of issues? We have all
kinds of directories like that. Putting them on our websites and making them
available in brochures is nice, but that’s not really getting the data out that
When you think of the quantitative kind of data we have, the programmatic
data, and this other kind of location-based data, the idea is to get as much of
this out for people to use for choice, for decision making, for information in
What we are doing with the new group is to try to see to what extent we can
bring the developer community for the apps and the technology — folks who make
a living trying to do this — together with our data folks and all of us to see
if the combination can produce something in terms of advice for how HHS should
go. We can at HHS say we know what consumers want, we know what patients want,
and we know what communities need. But we really don’t. We have ways of getting
this, but we would rather deal with people who actually know and who make a
living in these areas.
So that’s really the purpose of the new group. We have a long charter. We
took that wording, Todd and I and others — in our classic wonkish way, we had
to develop a page-and-a-half-long charter. The main thrust is the following. We
put together folks who know the applications, know the technology, know local
public health issues or patient issues or consumer issues, provider issues,
quality issues, plan issues, and so on, in terms of the kind of data they might
want or need that we have. Part of this is a process whereby we would brief the
group on the data holdings that we have. We would take them through each of the
major data systems — all of the CMS, all of the NCHS, the CDC surveillance,
and some of the other data and so on, and these other non-quantitative data
holdings that we have. We have to do it in digestible portions. If we go
quickly to all of these odds and ends of information systems, I think it would
But we would begin to brief the new group on all of this. That gives them an
idea of what we have and what we’re likely to have. Then we would, through
their own experience on their own applications, ask them to advise us on —
well, we would tell them the way we’re getting the data out now, which you’re
all familiar with, and the we would be looking to them to give us advice on
where we could make some improvements, for example, or how we
would and this is exactly related to the community data level and
so on, but it’s individual data as well and is patient data and consumer data
— what consumers and patients, the provider community, plans, the public
health folks would really like to have or what would be useful to them. We’re
asking them, in a sense, to be our bridge to those communities and then to look
at what we have and give us some advice about, through the technology and the
data presentation itself, what — is the county the best level? Is the hospital
referral region the best level? Is the state the best level? There are
different ways to do this. All of you know that the more detailed level of
geography we go, of course, the harder it is to have enough cases and to make
the estimates. We run into privacy issues for disclosure as well.
But the folks who do apps do this all the time. It’s a zip code-based model
or it’s a census geographic code model. We’re asking them to mash these ideas
together and to give us some advice in terms of what we would do to start
We might start off with some fairly simple things. They could look at our
websites. We’ll clearly brief them on the data holdings that we have.
Obviously, any member of the community who wants to sit through all the things
we have is welcome.
We’re putting together some tools. You have the guide to HHS surveys and
major data systems. We’re also updating our directory of HHS data systems,
which is a much bigger document. It’s all of these administrative data systems
that I was talking about as well. Then we have yet another inventory of all of
those other non-quantitative data, directories and so on, that we could make
Number one is, based on how we can familiarize them with our HHS holdings —
and possibly other agencies, too. I don’t think there’s any reason — we now
have, as you know — we’ll have them look at healthdata.gov. This is where we
are posting all of the data sets and tools that we have, more or less in an
open fashion for developers to use. Again, we don’t provide any
individual-level record data. It’s wide open. It’s basically public. We do have
methods of giving data out on a restricted basis, but obviously you wouldn’t be
interested in — so you could push the geography level, but you would never
want to get to the point where your identifying numbers are small enough to
identify someone. Otherwise, that’s the end of it for all of these initiatives.
So that’s what we would — we also ask them, just generally, based on their
knowledge of the industry, where things are going in terms of social media and
Web-based services, that sort of thing. Again, with attention to the data
holdings we have, what recommendations would they give us in terms of how to
We might give them some specific questions to answer as we move along as
Again, the perspective here is, quite consciously, to try to match up this
technology sector with the information all of us deal with every day. We would
then use them, basically, as we do the committee, fully, with giving us advice
on how to proceed there.
In some cases, I think they would be just be reacting to plans or reacting
to a resource. But I do think we want to ask them to think through, from their
perspective, what would make more sense. For example, besides the
healthdata.gov, where we post, in machine-readable format, our data sets and
tools, we have the Health Indicator Warehouse at NCHS. We have about 1,100
indicators or so. I think they go down to the county level. We could ask, is
this useful? Patrick Remington, who is a member of the group, is very familiar
with that area as well. Of course, Bruce does this for a living, too, in
They are mostly health indicators. We could look at access and health-care
quality measures as well.
Again, we need a better gestalt and then we need some very specific advice
and applications. Rather than just rely on what folks at HHS know about this, I
think the idea here was to actually get the folks who spend their daily
activities doing this.
In short, that’s the charge. The first step, I think, would be that
everybody gets familiar with everyone, orientation to the full committee,
review the charge. This is more or less the charge we agreed to with Todd and
others. If there are things we’re missing, we could certainly entertain that.
But the basic thrust, I think, would be along these lines.
Then we will start, probably at the September meeting, with a presentation
on our holdings. We’ll try to do it in a way — I think many of the folks in
the group are pretty familiar, but I think there are some areas that they may
not be familiar with. So we’ll get everybody up to speed.
Then I think we would start looking at — the low-hanging fruit idea is
always nice — look at healthdata.gov, our indicator warehouse, our ways of
disseminating this information. There are privacy issues. As a matter of fact,
you can only go so far on some of these things. I think with the Medicare data,
particularly, and Medicaid data, every step forward is met with a challenge
that that’s pushing the envelope for disclosure as well.
But there are ways to do this. It might not be posting the data publicly
necessarily. That’s probably never going to happen. But there are ways to get
at it, on a restricted-access basis, without — and there are ways of modifying
the data so that you remove dates and so on, things like that that minimize the
potential for disclosure.
So that’s where we are now. It turns out that in this regard HHS is probably
ahead of some of the other federal agencies. But I can see where we can include
environmental data, probably, environmental health data. We could probably,
down the road, look at housing and transportation as well. But I think we would
kind of like to get the health data that we do have available.
A lot of the census data, depending on what level you can make it available,
is good contextual information on population, on age, on income and SES. You
can get a lot of information about your neighborhood or your county or your
city by mashing up some of these things. But you have to make it easy. You
can’t ask everybody to be a researcher.
Let me stop there and see if there — that’s the gestalt, at any rate. I
think we’re just going to have to do it step by step, Justine, as we go
DR. CARR: Actually, I thought we would go through who the new members are.
Leah Vaughn is here. Lee is from the Health Policy Group. We welcome you.
Kenyon Crowley is here. Bill Davenhall will be here this evening. Ken is
from the University of Maryland. Bill is global markets manager, Environmental
Systems Research Institute.
Many folks know Chris Gibbons, who has testified here before, from Hopkins.
He won’t be able to join us tomorrow. Similarly, Patrick Remington cannot join
us tomorrow, I think. We’re not sure. Kalahn Taylor-Clark will not.
Then Bill Davenhall will be here tonight. Mo Kaushal is coming in, and
Joshua Rosenthal just testified recently —
MS. GREENBERG: I think he’s going to be here.
DR. CARR: Yes. Then from the committee, it’s Paul, Walter, Bruce, Leslie,
I think we’ll tag onto the full committee meetings. Whether we do something
by a call or a webinar or something in the summer — we may.
MS. GREENBERG: I just wanted to say, obviously this first meeting will be
tomorrow afternoon. I was asked about this. Of course, this is an open meeting,
like any of our meetings. Any of the members who are not actually members of
this workgroup but are interested can stay for that.
Similarly, whenever we meet in conjunction with the full committee, any
members of the full committee are more than welcome to participate, and even
teleconferences, I think. The only exception would be if we were bringing the
group in for an in-person meeting that wasn’t associated with the full
committee. Our resources might restrict that. But even then, we could have a
link or by phone or something.
MR. SCANLON: If the workgroup would want to have a hearing or a public
meeting where they bring in others
MS. GREENBERG: Yes, we do have some resources for that.
MR. SCANLON: We’ll try to work as much as we can, teleconferencing and
webinar. The two workgroups it has to do with work through the
full committee. Otherwise, we can’t create — we can have workgroups,
subcommittees, as long as they work through the full committee.
It’s a bit of an experiment in the sense that it’s a workgroup. We have
extended the committee member resources through this workgroup mechanism.
I think, Paul, you have seen that operate pretty well under the Health IT
You never get the full participation of everyone who is on the group, but we
made a lot of progress that way.
DR. CARR: We’re grateful. Todd will be here tomorrow morning, I think, at
11:00 to provide further insights and blessings.
DR. GREEN: Can you just say a little more about your current thinking about
the functional relationships between the new workgroup and existing
MR. SCANLON: I think we envision them working like — we modeled it after
those others. It’s a bit of an experiment. The new working group — we’re going
to see how this works. In essence, it’s a working group under the full
committee. I don’t necessarily know how this is going to work, with all of
their advice that they necessarily have to come through. I think we may use
them, to some extent, as a lot of the agency heads use them, as a sounding
board. This is a very hardworking committee. A lot of the other agency FACAs
don’t do the level of sophistication and quality of work that this group does.
You basically come in twice a year and the director tells you something and you
say, that’s fine. I’m exaggerating. This is the other extreme. You are
definitely a different mold here.
But I think there will be some instances where we just need some feedback on
a plan or we need some thoughts on directions to go in. I don’t think it will
be in the form of formal recommendations. I think it will be a sense of the
expertise. That’s where I see some of the products coming from.
There may be other products that are policy-related where I think they will
have to come — privacy — through the full committee.
We’re just going to experiment, to some extent.
Does that make sense? We’re going to have to experiment a little bit.
MS. GREENBERG: Obviously, we have talked about this. I’m learning, too.
You’re making it up as you go alone.
Formal recommendations, I think — and that’s what the charter says — would
go through the —
MR. SCANLON: Yes, I think if it’s a deliberative recommendation, it really
needs the full committee. That would probably have to go through.
MS. GREENBERG: It would go through this committee.
MR. SCANLON: It’s more the apps —
MS. GREENBERG: Informal feedback, I guess.
MR. SCANLON: It’s the sense of expertise or knowledge from that community
about where to go. It’s almost inimical to the very idea of this kind of work
to have a lot of clearance. But anything policy-wise or —
DR. CARR: But I think you said it right. When you think about the spectrum
of advisory committees, it will be a blend of this, on things related to this
committee, and otherwise kind of a reaction group, you think?
MR. SCANLON: Yes, and they will have to find their sea legs, obviously. On
the technology side of it, for example, we have wonderful expert folks in HHS,
but we really don’t have — nobody in HHS makes their living developing apps.
We need folks there, and we have to rely on their advice. So we’re looking for
“yes, that sounds like a good idea,” or “that would make no
sense.” There’s no business case there. We’re kind of looking for that
kind of reaction.
That’s the mold of the other FACA that an agency head might have —
“we’re thinking about this. What do you think about it?” It’s not a
formally developed set of recommendations. It’s “yes, that’s a good
idea,” or “that’s not the way to go,” or something like that. If
it looks like something is more controversial or requires more deliberation,
then it has to come through the full committee.
MS. GREENBERG: I think what Larry might be pointing to — earlier today, we
heard from the Quality Subcommittee. There clearly are connections there in
what they are recommending with this new working group. Of course, Paul is on
the new working group as well.
But I think, in the spirit of never letting this committee rest, I would say
that there may be some opportunities for some collaboration between the working
group and an individual subcommittee, and where those opportunities exist, we
should try to foster them.
Justine has graciously agreed to chair the group. I would expect that she
would come to full committee meetings and report on the group, even if the
group weren’t meeting at that time, so you see how this is going.
For me, it’s very exciting subject matter-wise. But as a student of the
national committee for most of my career, before I was actually involved with
it, this was a standard way that the committee operated back in the 1970s, with
what they called technical consultant panels. That’s how we got the Uniform
Hospital Discharge Data Set and many of the really basic foundational standards
that the committee was responsible for over its history and still have
By the time I got involved working with the committee, that wasn’t possible
anymore, to convene those groups. From a policy point of view, it wasn’t
allowed. Then it became allowed, but, from a resource point of view, we weren’t
really able to do it. Now we’re back — there’s nothing new under the sun — a
very cutting-edge kind of topic, but back to a process that we have used in the
past way past, when none of us were really involved.
I think we’ll really be learning as we go along. This is a working group of
the full committee, and I want the full committee to feel engaged and to
interact with the group, because I think there will be benefits on both sides.
We invited the members to come to the full meeting if they wanted to, if they
could bear it. We thank those who were able to. But this time, obviously, we
didn’t give them much notice. But I think it’s also useful for them to see how
the committee deliberates when there will be recommendations coming before the
I encourage you all to reach out to our new cousins. Certainly I hope
everyone can stay for tomorrow.
MR. SCANLON: If I could just add a bit, I can see, in terms of what the
committee has already done and would be doing, that some of that will feed into
the workgroup in terms of the community data, for example, all the privacy
data, I think, and some of the other areas as well — quality. I can see that
certain directions that the new working group goes would feed into these groups
as well, where a more systematic, comprehensive look would be needed. Really,
it will be the challenge to us here to keep the relationship between the two
Obviously, any member of the full committee can attend any part of the
meetings of the workgroup.
Again, just as a practical matter, the original request was to create a new
advisory committee to focus on this. We really thought this didn’t make any
sense, given the NCVHS. But there was no way we could add nine members to the
NCVHS from these areas. So what we agreed to — and I actually this idea is
stronger than creating a whole new committee — is this hybrid that we kind of
developed, which I think could actually work quite well. I think it benefits
from all of the thought and the expertise on the full committee, as well as the
new perspectives that we really had to reach out for.
DR. GREEN: This is really one of the most exciting developments I have seen
since I have been on the committee. I’m very enthusiastic about it.
One of the things you learn practicing medicine is to listen really hard to
side, parenthetical comments that patients make, because they are usually very,
very rich with information. They are telling you want they think you want to
hear, but then when they make these side comments, often the truth comes out.
So I’m going to accuse Jim Scanlon — I think I wrote this down exactly —
he said, “It’s almost inimical for this type of work to have clarity”
MR. SCANLON: Clearance. But clarity is also true.
DR. CARR: If you make a PowerPoint and cover your main points, clarity is
never a problem.
DR. MAYS: I know it is just starting. One of the resources that actually is
an incredible resource — and he’s probably going to kill me for offering him;
when you’re not here, that happens — is to think about NIH, and Bob Kaplan, as
our person, is actually the best to do that. His office actually does a lot of
development of apps, particularly around health interventions, as well as
access to data. This summer — I’m going to actually participate in it — he
has a weeklong mHealth training. It really is doing this exact thing, where you
are looking at the development of apps for some of the health data.
The other thing is that NIH is very interested in this as well. My center
that just got funded, my disparities center, is all in the issue of developing
apps. We have actually done some community work — because it’s all about
disparities — of how it is that racial and ethnic minority groups want to
access their data, what can be developed. Interestingly enough, we will
actually have some competitions where the community is going to come in with
our engineering and computer science students and come up with ideas, develop
these apps. For us, it also is, how are we going to push some data out. The
whole center is based on using technology. NIH was very hot on a lot of this.
There may be some other projects around. This was one of the things that,
when we read our reviews, they were very excited about. We’re working with the
Wireless Institute and the Nanotech Institute to get a lot of this stuff out
the door. But the community is driving it, actually.
MR. SCANLON: And I might add — and again, we focus a lot on apps — some of
this is just getting the data we have — it’s probably never going to be for
your iPhone, but I could see that some of this information would be used for
local public health purposes and for quality, for access, to underserved kinds
of analyses, for workforce shortages. It would be useful for policy and
decision making, even if it’s not simply an app. In fact, that’s probably where
the payoff would be. Where we have tremendous resources, have tremendous
expense of data, and not to have its reach its full potential is almost a cost
in its own right. That’s where we’re trying to think of how to do this.
Again, there are limitations. I mentioned privacy before. The further we
push this out, the more we’ll have to think of innovative ways to do this. But
I think we can really come up with some good applications of the first kind,
and later we can expand it to other agencies. I think environmental health
would be — EPA, for example, would be a good partner as well.
But that’s kind of where we are going. Then we just need advice on
healthdata.gov and the Indicators Warehouse and so on, about how to make it —
they are definitely not consumer-friendly. That’s not what they were intended
to be. They were meant to be developer-friendly and analyst-friendly. Are they?
We’ll have to see.
DR. COHEN: Jim, you stole essentially what I was going to say. I want to
just add the focus of state health departments and community-based
organizations. The data are out there. They are just not easy to find and not
particularly user-friendly. We have spent an enormous time and many
states have — developing our own Web-based query systems. For me, this is an
opportunity to get back into pushing the feds to help states figure out how to
do this more consistently, not only focusing on the traditional HHS data, but
resource and asset data at the community level, which is really public health
data writ large that we haven’t really incorporated into our thinking.
So I applaud the notion of thinking about using other agencies’ data. I
think it’s the key to success of our local public health interventions.
We spend an enormous amount of time putting census data in our Web-based
query system, because for our users at the community level, when they are
writing applications to get grants, the first thing they need is easy access to
that information. So all this contextual information and less traditional
public health information really, I think, is going to be the key for providing
access to useful information for communities and state and local governments
for making decisions.
DR. CARR: Excellent.
Jim, thanks. We look forward to it and we look forward to working with our
new colleagues. Terrific.
Now we have a chance to have a bit more deliberation where we are headed,
where we are going. But before that, I alluded this morning to some additional
comments I wanted to make.
I think you have heard, and we don’t have to say, how outstanding the
Standards Subcommittee was yesterday, because I think we have heard it from
many sources. It truly was exceptional. In the morning we covered lessons
learned from 5010, preparation for the implementation of the operating rules
for eligibility and claim status, a DMSO update, and the ICD-10 transition.
This afternoon we would like further discussion as a full committee around
the administrative simplification and the ACA assignments. As a way of teeing
up the discussion, I would like to review challenges that came up since our
last face-to-face meeting related to the naming of the authoring entity for the
remaining operating rules.
As you know, we had a teleconference on May 4 and recommended CAQH CORE as
the authoring entity for the remainder of the HIPAA transactions. We also
recommended that CORE and WEDI collaborate to ensure industry participation in
the process and, finally, recommended that the SDOs also take an active role.
The development of these recommendations came about through the
ever-diligent efforts of the Standards Subcommittee, working tirelessly between
March and June meetings in order to help the industry and the Department meet
the very challenging deadlines presented in the Affordable Care Act to adopt
and use operating rules, identifiers, and other mechanisms to quickly achieve
the benefits promised by administrative simplification.
An unintended consequence, however, was an industry perception of lesser
transparency than that to which they have become accustomed through NCVHS when
we convene, as we did yesterday, face-to-face with full industry
representation. While extraordinary efforts were undertaken by the Standards
Subcommittee to reach out individually to key stakeholders, incorporate their
input, the committee recognizes the concern about transparency that emerged,
and we want to ensure the industry that NCVHS is and continues to be an
advisory committee that encourages participation by all industry segments and
provides a level playing field for organizations and stakeholders.
To ensure that misconceptions are dispelled, the NCVHS will continue to
adhere to its process of soliciting comment, discussion, and discourse within
the public hearing venue and other established vehicles, such as the submission
of written comments to NCVHS for its consideration, so that total transparency
continues to be maintained.
I’ll be even more explicit and say that we will take a leadership role in
fostering the transparency and industry participation that’s required in the
statute for the adoption of operating rules — is adhered to and fulfilled by
any designated authoring entity. We may from time to time make additional
requests for information and/or documentation from any authoring entity to
ensure that every good-faith effort is being made to achieve broad
NCVHS will work with authoring entities to develop a process for revisions
and/or appeals made by industry for changes to operating rules, similar to that
undertaken by standard-development organizations. We wanted in particular to
take this moment to reflect that we tremendously value the feedback. We value
not only the importance of being collaborative and transparent ourselves, but
also the importance of being a model of collaboration and transparency that
others will aspire to and follow.
With that said, we would like to turn the attention of the full committee to
the challenges we have in front of us and the industry, including multiple
important and time-sensitive assignments from the ACA, as well as aggressive
implementation timelines in an industry that is experiencing a convergence of
requests from all sectors that are costly in time and resources.
I feel like at every meeting I’m commending the unbelievable efforts of the
Standards Committee, because they are extraordinary. As we have been working
through these various things, it has led us to kind of reflect on we’re
running a marathon like a sprint, in many ways. The end is not yet in sight. A
lot of the discussion that came out in our calls was about whether we should be
thinking about a couple of things.
One is, when there’s competition for our time and energy, should we be
Then also in some of these things, although they fall to the Standards
Committee — and I’ll mention claims attachments — it’s not just about
standards that go into the content of the claims attachments, but it’s really
us thinking more broadly. Claims attachments were first put forward, I think,
in 2005. The world has changed in the last seven years. Now that we have 48
percent of hospitals and one in five providers on EHRs, there is a lot more
digital data that’s available that may call into question the traditional
concept of a claims attachment. Similarly, as we have been discussing today,
the whole issue of privacy what is the granularity of the
information there? Where is it going? Who sees it?
Finally, although I agree that we will know later this month a little bit
more about what the Supreme Court says about the Affordable Care Act, in many
ways — certainly in some states — accountable care is a
reality approaching a reality. It’s my reality, anyway. As we move
forward and we try to be our deliberative and looking-at-where-the-puck-will-be
selves, should this committee be thinking about new models of care, and even
whether you need a claims attachment in the new health-care world.
So I wanted to just throw out those ideas and turn it over to the committee.
Let me just say that over the next 45 minutes, we might want to tee up the
kinds of things we ought to be thinking about. We would like to have an
Executive Subcommittee retreat for a day this summer to further refine that in
preparation for a larger full committee retreat, as new leadership comes in,
So today’s work, in the next 45 minutes, is to put out some ideas about how
we might go about our work. What is the space we should be in? We heard a lot
this morning from ONC and CMS and the role that we have played successfully and
the role we might play — not the three-times-a-week tiger team approach, but
taking it up to the 15,000-foot level.
Those are just some ideas. Let’s use the next half hour to throw some ideas
out and the last 15 minutes to kind of tee up some themes that we will carry
through in the executive retreat and planning for the year ahead.
With that, I invite Walter and Ob — they led the hearings yesterday on
what’s on the horizon. I also invite Bill to give his perspectives. Then we’ll
open it up to the rest of the committee.
DR. SUAREZ: In a sense, it is hard to frame this because the topic of
Affordable Care Act Section 10109 is very specific, very concrete. It asks us
to look at five things and see if there is any opportunity for standardization.
Those five things are very small and very concrete.
One of them is provider enrollment, the process of enrolling providers in
health plans. Is there an opportunity for standardization?
Another one is claim edits, the edits that health plans do as they receive
health-care transactions, and the decisions that they make based on those
edits. In many cases it’s a rejected transaction because this particular edit
stopped the transaction. There was an error. But there is inconsistency in how
those edits are done and which edits are done. So there is a question as to
whether there is an opportunity for standardization there as well.
Then there is the balanced side: If we standardize the edits, what happens
to program integrity, as Bill will point out, and what happens to fraud and
abuse and other things that may be easier because everybody knows how to
process and try to game the system or something like that.
Before I get into that, I think what we started to see — and I think it has
been emphasized by what we heard yesterday in the hearing, but we are hearing
from the reports from the other subcommittees — is that we have an opportunity
to step back and step up, if you will, and look at things in a more holistic
way and realize that, while there are some immediate priorities that we have to
address claims are still going to happen for the next X number of
years, five years, three years, maybe ten years between now and the
next few years, we will still have claims and enrollments and premium payments
and all these processes. How long the transformation of the health-care system
will take us away from the traditional way of payment to a brand-new mechanism
in which there are no claims and there is no enrollment or there’s enrollment
in a different way — yes, that’s an important step.
I like the point that was made earlier by Robert from CMS. I think CMS is
looking at that that way, too. They are stepping back and stepping up and
saying, what are we looking at here? Are we looking at 2017, with the current
version of how we do business, or are we looking at 2017 with the new way of
thinking of health-care delivery and payment?
In many respects, we have to look at both at the same time. Are there
opportunities for current improvement in standardization and the benefit that
standardization brings? How, at the same time, can we think about the new way
of doing business and what kind of standardization might be needed?
In my mind, there are certainly a number of big themes. There are several
themes around continuous collaboration, openness and transparency, enhanced
communication and outreach. That will continue regardless of what direction we
go. But there’s one theme that I think is coming up quite a bit around
standardization, and that is, in my mind, the convergence of the administrative
world with the clinical world, the fact that we are seeing more and more the
same types of standards that have been used in exchanging information for
administrative purposes played against the clinical data exchange standards.
The best example, of course, is claim attachments. In the old, ten-years-ago
world, claim attachment was the attachment of some medical information that was
sent on paper, fax, FedEx, or whatever other mechanism. Nowadays, we’re talking
about the claim attachment as a clinical message that should use the same
standard that providers use to communicate between themselves, between two
EHRs. If we have a different standard to submit the attachment to a claim for
payment purpose than the one that we use to exchange data for clinical care —
if it’s the same source, the same data, but now sent in two different formats,
that doesn’t seem to make sense. It’s coming in many cases from the same
source, which is the EHR. Not all claim attachment data comes from EHRs, but a
significant amount of it.
So that’s one megatrend, if you will. It’s really this progressive
convergence between the administrative world and the clinical world, and the
fact that there are many transactions where we’re looking at the nine HIPAA
transactions, we’re looking at four or five more transactions that are beyond
HIPAA, and all these administrative transactions need to be played along with
the standards that are being used for meaningful use, for electronic health
records, for health IT, for HIEs.
So that’s number one, I think.
Number two is this point about having to balance between what we have as
current issues — today we have 5010 and ICD-9, and coming up, ICD-10. We have
the standards, the transactions. There are some opportunities to improve the
standardization of those, making sure that we benefit from the standardization
of those. As we said in our HIPAA report, we have achieved some, but we have
still a long ways to go to achieve the full benefits of administrative
simplification. So today and in the next few years, we still have the
opportunity for some of that, and at the same time, trying to think of these
new forms of insurance exchanges and health reform. What are the implications
for the standard transactions? There are still going to be exchanges of data,
of course. Whether we go to transactions or simple exchanges of data, or
however we call it, there is still going to be data being moved from one place
to another via some mechanism. So we have to think strategically about how we
divide the time.
The last point I want to make is that clearly we, based on that, need to
identify priorities and decide to go along with those and move along with
those. Part of the issue that we have struggled with in the Standards
Subcommittee is prioritizing the things that we were called to look into from
the Affordable Care Act. Certainly that was one of the hopes that we had in
bringing to you here some of this discussion. We have five major areas or
domains, five topics, that the Affordable Care Act suggested that we should
look into for potential standardization. Those were provider enrollment and
standardization of claim edits and standardization of audits being done by
health plans and the possibility of bringing the property and casualty
insurance, which includes workers comp, into the same type of standard that
HIPAA requires, so that we don’t see things like, when I send a claim to a
payer, I have to use 5010, but if I’m sending it to workers comp, they are
still on 4010, so I have to dumb down the transaction to 4010. In the future
I’m going to do ICD-10 with the standards, but workers comp might not go to
ICD-10, so we still have to do ICD-9.
Those kinds of inconsistencies are the kinds of issues that we need to look
We developed a couple of strategies and discussed and have debated them. We
wanted to bring them back to the full committee for consideration.
By the way, in the process of looking at these five areas, we did have a
hearing in November and heard in each of the areas from different perspectives
about what some of the issues are, some of the opportunities for
standardization, the relative importance and significance of some of these
areas. Out of that, in March, we submitted a letter to the Secretary
recommending that we need to take some time to look into the details of each of
these topics and develop a strategy — we noted that we would be developing a
strategy by June, basically, to bring back to this full committee meeting — on
how HHS can move forward with the identification of standardization
opportunities in each of these areas.
One of the options of the strategies was that maybe we can organize smaller
groups. We have created a workgroup now, the first workgroup in NCVHS
activities. We could organize something like tiger teams, like HIT standards
and HIT policy committees are organized and are known to do. Each of those
tiger teams will involve people from different — not just committee members,
but we’ll bring in subject-matter experts, stakeholders from the industry that
are knowledgeable and represent different perspectives on each of these topics.
For provider enrollment, we’ll bring representatives for the providers and the
payers and other groups. Similarly we would bring people from different
perspectives in each of these other topics.
We would have those groups do the research and come back to the committee
with a report — we found this. There are clearly some opportunities. There are
some benefits. These are the concrete benefits, and here is the status of the
standards, or lack of standards. This is what we recommend.
So that’s one possibility. Certainly the idea would not be to launch five
different tiger teams all at the same time and give a year to all of them, but
progressively identify one or two topics and start moving along with those. In
some of the topics there is already a lot of research that has been done in the
industry. So they might take three or four or five months to look at, develop,
and come back to the committee —
DR. CARR: I do want to leave time for discussion. We can get into what the
details would be. We could do tiger teams. We could do —
DR. SUAREZ: So that’s option one. The idea was to organize some tiger teams,
That was the end of my talk. Ob?
MR. SOONTHORNSIMA: I’ll stay at a little bit higher level. Yesterday was
very, very rich in terms of content, and they all flow very well. In a
nutshell, personally what I took away from the discussions and the
presentations was the following.
I think we’re being asked to really do some reality check in terms of what’s
going on in the industry. A lot is being asked, and a lot of mandates and
initiatives. The industry is overall, regardless of what aspect of health care
— everybody is trying to improve quality, access, and cost of care — but lots
of mandates, lots of competing priorities.
At the same time, we also heard of different ways, different solutions,
different ways to tackle all these different things, such as how we do
By the way, testing end to end — all day long, all we heard was end-to-end.
We have been talking about end-to-end testing forever and ever.
But there isn’t a silver bullet. Everybody has different perspectives.
While we are trying to complete all these different mandates and
initiatives, we also have different means of testing. We all recognize that
testing is not effective today. That’s what everybody is saying.
Thirdly, because testing is not effective, what we heard also is that the
value of these initiatives isn’t being realized, in a nutshell, let alone that
we have yet to complete all these mandates on time, on budget. Also ensuring
that the quality of work, when it’s done, whatever those mandates are — how do
we ensure that whatever the work is does not impact the industry negatively? We
heard a lot around provider cash flow issues as a result of 5010.
That gets back to the issue of quality assurance testing. This sounds like a
broad, broad concept, a very simple concept. But they are very difficult to
implement. How do you do end-to-end testing?
That’s kind of the backdrop, what I heard. What Justine and Walter tried to
tee up a little bit maybe I can summarize a little bit better — in a different
way, not better, excuse me.
Maybe three things we can consider. Yes, we do have to take care of ACA,
what ACA is asking us to do. But the second point is, how do we also consider
all these changes — and “convergence” is a good word, because they
are converging. If you think about the major issues and opportunities around
delivery models and payment innovation, payment reform — we heard about ACO,
we heard about bundled payment, and so forth. That’s an opportunity, as well as
If you think about quality measures — we heard what Paul and the team
talked about — if you think about the transparency initiative — how do
consumers consume health care, and how do they select quality providers and
plans and so forth?
So delivery model, quality measures, transparency – these are bigger,
bigger goals of the industry. I didn’t mean to be exhaustive, listing
everything, but those in a nutshell.
How we then marry, if you will, the work that we have to do for ACA with
some of these things that take longer to do, that are much more transformative,
to use Jack’s word — and perhaps we can better prioritize these things as a
I’m going to steal the word from several of our presenters yesterday,
“roadmap.” Maybe there’s a better way to create a roadmap that
marries all these objectives together.
I’m going to pause right there and hope that we can summarize the
conversation we had yesterday. I think it’s now the task and the value that we
as a committee can bring to the table to help elevate these activities to much
more transformative, much more strategic things that are going to add value to
DR. CARR: Bill?
DR. SCANLON: To sort of follow up on what Ob is saying in terms of
reconciling, our charge from the ACA is with this bigger picture. I think
there’s also the third element, which is our bandwidth or our resources. That
actually triggered our initial discussions — and there has been a lot of
evolution in the discussions of the subcommittee — the issue of bandwidth and
the Section 10109 and these five very specific charges that led us to starting
to talk about priorities. I think what has happened in the evolution of the
discussion is that the issue of priorities is becoming much more global, and
this recognition that there are more parts about the ACA with respect to HIPAA
besides this Section 10109, which we have dealt with in plan identifiers and
operating rules, et cetera. Just to realize the potential of HIPAA as passed in
1996 in today’s environment is an overwhelming challenge.
That’s putting aside all of the transformation or reform that’s going on and
it already going on. Private plans are paying providers for some bundles.
Private plans are paying for ACOs. Medicare has an ACO survey. Medicare is
thinking about bundles. All of these things have information needs that we’re
not meeting today and that we need to be thinking about meeting in the future.
So there is this issue that simultaneously there are huge demands for thinking
about how we change the information flows.
To get back to the specifics of Section 10109 and to meet this sort of
obligation or respond to the charge from the ACA, we did have a hearing. We
have had a lot of discussions. The alternative proposal to the tiger teams that
Walter was talking about and trying to set priorities for them was to say we
have assessed this issue and we think that, among all the things that need to
be done with respect to HIPAA and with respect to health-care reform, these are
things that, even where they have merit, can be postponed. That is our judgment
at this point in time.
For three of them, there’s a question of what the role of standards might
be: audits, edits, and payment rules standards. Particularly in a world where
everybody is trying to transform payment rules, how do you set meaningful
standards that are going to be effective in terms of simplification at this
point time? Maybe there will be a future point in time when that’s possible,
but it’s very clear, at least to me, that it would be almost impossible to do
it today, to have a standard that says we really accomplished something in
terms of simplification, because we also had to leave open all this latitude
for plans and providers to come to arrangements that are more effective in
terms of producing quality, efficiently delivered care.
I think those are potentially something for the future.
The issue of provider enrollment — again, it may be a legitimate issue.
There may be a pain point in terms of providing information multiple times. But
relative to the bigger problem of providing information multiple times by a
factor of 2 to 5 or 10, which is what claims information involves, it’s not as
big of an issue.
Yesterday’s hearing reinforced so much about — we have so many problems
with claims information that these need to be sorted out. At yesterday’s
hearing, we had 39 witnesses. Think about how much we digested of that
testimony by this morning. It doesn’t happen that fast. So there is this issue
that we really need to be thinking about how we build off of what we heard
yesterday to try to inform this bigger discussion.
Property and casualty again is an area where there may be some very
legitimate concerns. The point about why they are stuck in the 4010 world is a
very legitimate one. But they are small relative to the bigger problem. We need
to be thinking about focusing on, I think, the bigger problems and, at least
for the short term, give our full attention to those.
Justine brought up the issue of attachments. It’s going to be, I think, a
key to both standardization and simplification, as well as to reform. How do we
develop the standards and the information flows that are done in the most
efficient way possible, which is probably going to involve some kind of linkage
to the EHR, and yet we’re going to support reform? If we don’t improve the
information through reform, we’re not going to do the quality assurances, we’re
not going to do the risk adjustments that are necessary, and these good ideas
that we are trying out now for their potential value are going to not work.
Then we’re going to discard them. And we don’t have another arsenal full of
good ideas there in the background that we haven’t tried yet.
One of the problems we have had in policy is that we sometimes implement
things before we’re ready, and the entire idea gets discredited. We need to be
careful about that in the process.
The two alternatives are to do the tiger team approach for a subset of these
charges or to say the Standards Subcommittee needs to step back and focus on
the big issue. But it’s not just the Standards Subcommittee that needs to focus
on the big issue. We think that the Standards Subcommittee needs to keep coming
to the full committee to talk about this broader context ad get your input.
Quality is probably the prime example of the intersection that we have heard
about today. There’s a huge quality dimension to all the reform ideas that are
being discussed. If that is not adequately addressed, those ideas have a very
great risk of failure.
DR. CARR: Let me open it up. Len.
DR. NICHOLS: Well, that was a tour de force. I would just say it this way:
Walter, Ob, and Bill, I hear you. But it seems to me that, at a minimum, the
claims adjudication, what you call edit, is so important to getting clinician
buy-in to the entire reform apparatus that I just hate to see it postponed out
of legitimate reform and transition fatigue. I need to see way more
cost/benefit on why that thing is — Bill, I’ll just tell you what the IOM has
told us, which I believe you probably know better than I do. Hospitals spend 20
cents on a dollar getting paid, physicians spend 30 cents on a dollar getting
paid. This is money that we could actually carve out of the system that
arguably is not adding to social value. Tell me again why we can’t —
DR. SCANLON: On that one, it is not sort of reform fatigue that is the
issue, at least in my mind. The question is, what is an edit? In my mind, an
edit is, a plan receives information from a provider, looks at it, either can
reject the claim or can go back to the provider and ask for additional
information. The additional cost that we can eliminate is if we can reduce the
requests for additional information. I’m not sure that as a provider I should
care what a plan does with my information when they get it. As long as we have
some transparency in terms of what payment rules are, then I turn my
information over, and if a plan is being consistent with the rules that have
been promulgated, then I’m okay with it.
But I don’t like the fact that they keep coming back to me and saying, give
me more information. This is the attachment issue. This is where we’re talking
about I can’t remember the number, but I think we’re in the
billions of attachments where a provider is asked, give me more than what you
gave me on the claim. We have a standard claim, but now I want to know more.
That, I think, is what the IOM is getting at in terms of the unnecessary
administrative costs, because we’re making specific requests that —
DR. NICHOLS: That are plan-specific. I think what the IOM was saying is that
we need one set of rules across —
DR. SCANLON: Those, I think, are the rules for attachments, not necessarily
the rules for edits.
DR. SUAREZ: I think there are different types of edits in the administrative
world. There are technical edits. When a claim is received, there’s processing
that is done to ensure that technically the transaction is correct. There are
differences in those and how they get applied. When I send this claim to these
people, they reject it, and these people don’t. So there is that possibility.
To Bill’s point, what is the size of that issue? Is it a $1-billion, is it a
$10-billion, or is it a $100-billion issue? That’s the question. But that’s one
level of claim edits.
Then there are other levels of claim edits, more into the adjudication
Then there is — I wouldn’t necessarily call it an edit — more a policy
decision of the health plan to request additional information to support a
claim. That’s not necessarily a claim edit, but it is connected to the fact
that the claim was stopped because there is additional information needed.
So we have different levels and conceptually different things about edits.
There are potential opportunities for standardization in some areas. Then there
is the concern that if we try to standardize too much, as Bill points out, we
are now trying to standardize policy or payment, which is something
inappropriate, probably, for this.
MS. KLOSS: I heard two additional things yesterday. One was how helpful it
would be to this whole dialogue to have the pieces connected, to have some
clear understanding of how doing this improves this part, and looking to the
committee to work at that level, to lay that out. It seems to me, if we did
that, it might give us an opportunity to say where the greatest impact in
dollars is. I do think that we have to drive cost out of these processes and
that has to be really one of the main jobs here.
The second thing I heard is that we’re at a point of real vulnerability in
getting to the ICD-10 successful launch and that the committee should give some
thought to what we can do to kind of provide a platform or a convening role or
something that helps move all the pieces and parts of this together. As so
often happens, our industry is siloed in terms of who is doing what piece of
that. Of all the complex tasks, this is really where an end-to-end view is
I heard those two things. Understanding — that may help inform how we
prioritize, but a special focus on getting to successful ICD-10.
DR. SCANLON: I would like to apologize if I gave the impression that it was
health reform fatigue sort of driving this. Actually what’s driving it, at
least in my mind, is the idea that we have to have the biggest impact we
possibly can on necessary administrative costs.
DR. NICHOLS: Agreed.
DR. SCANLON: The issue of mapping out where they are and trying to attack
the problem that’s most amenable.
DR. NICHOLS: I thought we had postponed ICD-10. I thought the Secretary said
she was considering it —
PARTICIPANT: One year.
DR. CARR: Judy?
DR. WARREN: The ICD-10 has a special spot in my heart. In 2008, this
committee held hearings on whether or not we should go to ICD-10. We decided we
should. We brought industry in. We had them help us decide what the realistic
timeline was for implementing it. They also gave us lots of really rich ideas
on how to do that. AHIMA stepped up to the bat, HIMSS stepped up to the bat in
educating people, in developing roadmaps, timelines. I can’t think of anything
we teed up that was better than that strategy. Yet, as soon as meaningful use
came in, before there were even any regulations, people quit looking at 5010
and ICD-10, which were already in regs, and started looking at meaningful use
and then said, well, we don’t have time to get ready for 5010 and ICD-10 —
certainly not ICD-10 — and then started lobbying to have the date pushed back.
One of the things industry told us in 2008 was that one of the problems with
HIPAA was that every time there was a deadline, industry could lobby and the
date would be pushed back. So we were to ensure with the Secretary that we
would not push back 2013.
Now we’re here, with 2013 having been pushed back to 2014. Industry is
telling us, you need to set a realistic deadline. You need not to push back the
I’m at a loss as to how to proceed past that.
DR. CARR: I think the thing that was very compelling yesterday was not the
vision, it was the execution. I think there are things that you can articulate
in policy, but what we heard were very powerful things, like the delays with
5010. They have caused at least some practices to go belly-up. The cost of 5010
was estimated at $16,000 for practice, in a one-person practice, with the cost
of the implementation and the delay of the implementation.
I think those are the things that get lost in the detail.
I was impressed today with the speed of uptake of EHR, to the point that
those same practices — one in five now has an EHR. But what’s the
differentiator? The regional extension centers. That took this lofty idea that
actually also had financial alignment and then out went out and showed, here’s
where you plug it in, here’s the on button, here’s what you do. That’s what we
are talking about in the local providers.
What I came away with yesterday is that this repeated request for a roadmap
was, in many ways, an alignment of incentives. As we sit here and say how many
workers comp cases we have versus how many whatever else we have, if we can
begin to see that it’s this many events and this many dollars and this many
resources, we can, first of all, understand the 80/20 rule of where an impact
could make a difference, but also get alignment of the embrace. Right now,
today, as we have heard over and over — in fact, at our last face-to-face
meeting — for the practitioner in an office that does an encounter and sees a
patient for a sore throat and writes down sore throat and this is what I did —
they have not been instructed to now add the fact that this person also has
hypertension, diabetes, and all that, because that’s not what you are treating.
In practice, it is an encounter with one diagnosis.
Trying to help them understand why it’s important to have a more granular
diagnosis — left tonsil, right tonsil, whatever it is — it doesn’t tie out
with what they are doing every day.
That’s not to say that in the totality it makes a difference. We talked
about where the risk adjustment happens. There are two things. One is to point
out risk adjustment, population health, and how this all ties in with the new
health-care reform. But we’re still left with a 15-minute visit for a solo
practitioner who has just spent $16,000 on 5010 and has another $16,000 on the
EHR, and now has to shrink that visit down to get all the other diagnoses.
Obviously, I’m being extreme in this, but I think the point is that we have
not reconciled the perspectives. As we heard yesterday, the implementation,
with all the things that we had heard but we thought wouldn’t be an issue —
and it turned out they were an issue — or the voices we didn’t hear about what
the obstacles are — I think that the regional extension centers’ construct is
the local reality to the vision. I think that’s something for us to think
DR. GREEN: I am really appreciative of this report from yesterday. I think I
have an advantage in having not been there yesterday. Listening to this
conversation reminds me of two or three things.
One of them is, we are more than a decade past the IOM’s start of the Chasm
series. The initial book’s fundamental conclusion was what? The US health-care
delivery system as it is designed is so screwed up, it cannot possibly be
It’s my personal opinion that we will now spend a decade trying to fix it
and that this conversation has confirmed that they got it right.
A second thing — you guys are going to hate me for this; I’m going to hate
myself for this — it reminds me of Robert Frost and a poem he wrote called
“Directive” that some of you may know. It opens up with the phrase
“Back out of all this now too much for us. Back in a time made simple by
the loss of … there is a house that is no more a house … that you can find
by taking a road that is no more a road … if your intent is only to get
That’s where we are.
DR. CARR: Is this a recommendation? If you’re going to raise an issue, I’m
expecting that you have an alternative health system in mind.
DR. GREEN: The old one is burning down.
DR. CARR: We get that.
DR. GREEN: The comments about the visit — the visit is now no longer the
prudent, useful unit of analysis. We know this.
T.R. Reid’s book, when he offered his shoulder to 14 countries and he
tracked down how those systems worked when he did it in Canada and he
got to reputably the largest hospital in Canada and he asked to go see their
claims processing operation and their billing operation, where did they take
him? To a room. What was in the room? One desk. How many people worked in it?
One human being. You know what that means? They don’t do claims adjudication.
The requirements of the ACA make no sense to a Canadian. The edits and all
this stuff that are part of our life don’t even exist — Bill’s comments about
the ACOs, the restructuring.
I’ll hush with this. There’s change that’s normal and there’s change that’s
abnormal. Laws are written for normal change. That’s not where we are. We’re in
abnormal change. My recommendation is that the committee follow Linda’s advice
and Ob’s advice: Look for where the opportunities are for convergence.
I thought Bill had a very good — look at some of these things that have
been prioritized first to do and say those can be deferred, and avoid going
down those roads. We should actually plan for the future rather than the past.
DR. CARR: I think it is going to be hard to top those comments.
We have gone a little bit beyond our time. This is why we need a retreat.
I’ll ask you to take these thoughts away — and those who were here
yesterday — as you go back to the subcommittee groups and think about how this
But I think this isn’t going to be solved in a meeting summary tomorrow.
We’re trying to put some dates together for the Executive Subcommittee.
Marjorie, are we thinking of a one-day —
MS. GREENBERG: Probably. Certainly a full day, in person.
DR. CARR: What we need to do is get a checklist out to each of the members
to see who is available, a Doodle. If you have a computer and you can do it
tonight, let’s just do it so we can leave the meeting tomorrow with a date and
Part of the work today is to think about, what are the kinds of questions,
given the discussion we had just now — what’s the focus? The things that I
heard today were just the roadmap of how things tie together and opportunities
for convergence. I think those were two important themes.
MS. GREENBERG: Can I just add one thing, which goes with that? It was very
rich yesterday. Along those lines, there was a recommendation for some type of
summit to bring together the different stakeholders and try to do these things,
to at least do an outline of what a roadmap would be like and identify
opportunities for convergence and look at different opportunities to make some
sense out of this broken system.
I know the CMS folks were very taken by that. They saw that as something
that maybe they would benefit from.
I don’t exactly remember — I think it was around EHRs, which were then
called computer-based patient records or something — the committee has
partnered with the Department in the past on a summit type of meeting. I think
that is something that we might just think about.
But I would have to agree that there are a lot of things that have to get
done. If the committee could help move this along, this, I think, would be a
major contribution. People have been talking about needing some way to bring
these things together for a very long time, and particularly those issues of
convergence between the clinical and the administrative. I will certainly
devote my time over the next year to help you with this, if I can.
DR. CARR: I think, just to elaborate, what that summit idea came out of was
the idea that — one of the payers said, I understand my world completely. I
don’t understand the provider world, and the provider said, I don’t understand
— no, we didn’t have a provider, but the representatives of the providers said
the same is true.
We can have these conversations, but until we understand the problem we’re
trying to solve —
DR. WARREN: We did this when — we were going to have a Standards
Subcommittee hearing when Jeff and I co-chaired it, and at the time meaningful
use came out, and we took our meeting time and we really devoted it to starting
off the meaningful use kickoff. There were hundreds of people at that meeting.
It was huge. I think it really set the tone. So we might want to think about
something like that again and bring in ONC and CMS and HHS and all the people
that have all these deadlines and all these requirements and all of the regs
and all of that, and really help each other learn about what it’s doing to
impact others that may not know about.
DR. CARR: With that, we will adjourn.
Thank you, everybody, for yet another wonderful deliberative day.
(Whereupon, at 3:30 p.m., the meeting was adjourned.)