[This Transcript is Unedited]

DEPARTMENT OF HEALTH AND HUMAN SERVICES

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS

November 4, 2004

Room 800
Hubert H. Humphrey Building
200 Independence Avenue, SW
Washington, D.C. 20201

Proceedings By:
CASET Associates, Ltd.
10201 Lee Highway, Suite 160
Fairfax, Virginia 22030
(703) 352-0091

TABLE OF CONTENTS

P R O C E E D I N G S [9:20 a.m.]

Agenda Item: Call to Order, Welcome and Introductions, Review of Agenda – Dr. Cohn

DR. COHN: Welcome to the first day of two days of meetings of
the National Committee on Vital and Health Statistics. The Committee, as you
know, is the main public advisory committee to the U.S. Department of Health
and Human Services on national health information policy.

I’m Simon Cohn, a member of the Committee, and Chair of the
Subcommittee on Standards & Security. I’m obviously not the normal chair;
I’m sitting in for John Lumpkin, who is detained in New Jersey today. He will
be coming in and be joining the meeting tomorrow, so never fear, he will be
here, just a day late.

I actually do want to apologize to those on the Internet that
we are starting a little late. There is rain in Washington today, and as you
know, rain in Washington is like snow in some other climates, hence the clogged
freeways and, as I understand, also the subways today. But at least we have a
quorum at this point.

Obviously, I want to welcome fellow Committee members, HHS staff and others
here in person. Also, want to welcome those listening in on the Internet and of
course as always remind everyone to speak clearly and into the microphone so
that those listening in on the Internet can understand and appreciate our
deliberations.

We start with an update from the federal government, from data council,
from Jim Scanlon, Maria Friedman, Susan McAndrew. Then we move into a
discussion from the NCHAS Board of Scientific Counselors, and then we get an
update from David Brailer on national health information technology. We follow
that with, I think, a discussion around a population report that is being
developed.

And then this afternoon we move into reports – or a presentation on
the American Community Survey, and then a letter that we’ll be reviewing for
action at this meeting. After that, we’ll be moving into subcommittee meetings
and workgroup meetings for the remainder of the day. Hopefully, there’s not
anything else that we’re missing on all of that. Please let me know during the
introductions if there’s additional agenda items that we need to be adding to
the agenda today.

Now with that, let’s have introductions around the table, and then around
the room. For those on the national committee, I would ask if you have any
conflicts of interest related to any issues coming before us today, would you
see – please so publicly indicate during your introduction.

[Introductions]

DR. COHN: Thank you for joining us. Well, now, with that, we’ll start with
updates from the Department. I think Jim Scanlon, you’re on first.

Agenda Item: Update from HHS – Mr. Scanlon

MR. SCANLON: Let me start. Good morning, everyone. Let’s see, we last met, I
think, in September, and a couple of additional developments have occurred in
data policy and health information technology issues, and let me just spend a
couple minutes this morning bringing you all up to date.

I’m going to talk about budgets for fiscal year ’05 and ’06 and then I’m
going to talk about some — what are largely population statistics initiatives
here in HHS that we started under the Data Council and then a privacy notice or
two as well.

The – and obviously the elephant in the living room – someone said
there was an election yesterday!

[Laughter.]

MR. SCANLON: And you all know – the day before yesterday. Well, I voted
in it. [Laughs.] So you all know the results of the election and we’re – I
think all federal agencies are now kind of regrouping and planning their
budgets and other initiatives as the smoke is clearing.

So let me first turn to budgets. As you all

know, our fiscal year begins on October 1st, so we’re a

month well into the fiscal year ’05. We and most other federal agencies are
under a continuing resolution pretty much until the end of this month,
November, which really means that we’re authorized to spend at the level more
or less of what we did last year for the same period.

It also means there are, you know, for the most part, no new starts, no
major new initiatives, anything like that. The continuing resolution is more or
less aimed at continuing the service level of the previous year.

So presumably when Congress reconvenes later this month, we’ll either have
another continuing resolution or we’ll have an actual appropriation, and then I
think we’ll all have a better sense of where we are for fiscal year ’05.

The President’s budget for fiscal year ’05 actually had a number of very
positive developments in health information technology and in data policy. I’ve
told you about them previously; just a quick recap.

The budget includes a $10 million initiative between my office, AHRQ and —
ASPE and AHRQ for standards development and maintenance and so on, and that
money was put to very good use this past year. It really was, allowed us to
move forward on our Xnorm daily med and a number of very critical other
developmental work in standards, and it also supported some of the essential

terminology mapping priorities that the NCVHS recommended, and I think Betsy
Humphries will be describing those to you as well. So that really helped us a
great deal.

That 10 million is included in the President’s ’05 budget and at least
there seems to be every indication that that would be continued in ’05, and
Mike Fitzmaurice and I are kind of co-managing that. We’re starting to think
about what the priorities might be for ’05 on the assumption that it would be
forthcoming as well.

You’ll remember as well in the AHRQ budget there was $50 million initiative
for planning and demonstration grants to improve patient safety through health
information technology in hospitals and ambulatory practices and so on. Those
grant awards have already been awarded for the previous fiscal year. That is
included in the President’s ’05 budget as well, so we’re hoping that that will
ultimately be enacted into the budget for ’05.

In addition, you’ll remember that in the Departmental level budget, the
Office of the Secretary, the President’s budget proposed a $50 million
initiative for grants involving actual data exchange networks. This is again
for ’05. We don’t have that funding yet.

The focus here would be on data exchange and

interoperability. It looks like Congress is – I think

it’s been reduced to about 25 million at least in some of

the deliberations in Congress, so that may be somewhat

less, but hopefully we’re getting – we’re hoping that we’ll at least
have some funding there as well. But again, we won’t really know what the full
budget picture looks like for fiscal year ’05 until for at least another month
and possibly longer.

On the population statistics side, as I’ve indicated in the past, the
President’s budget contains some very good news and some very good initiatives,
and primary of which was a $25 million increase for the National Center for
Health Statistics. The purpose here was to help maintain and transform the core
statistical systems at NCHS. They’re very familiar to you all, everything from
vital statistics to the health interview survey, to NHAIS on the health care
surveys.

And there was some money as well slotted for methodological research. This
was our highest priority in the data area for that budget, and again, we’re
hoping that that will be acted upon favorably within the months ahead.

Not knowing exactly what the budget picture will be, I think we’re
basically continuing at the same level, but just postponing major developments
until – and major decisions until we get a better picture. Obviously, if
that doesn’t materialize, we’ll be in a more complicated situation with respect
to the surveys.

For planning for the ’06 budget, the – we’re really in a very
preliminary stage. The Data Council has reviewed all of the – sort of the
current portfolio of statistics and data activities and the proposals for
enhancements that agencies identified in their very early preliminary budget
thinking, but it really has not progressed beyond that stage, and I think now
the focus on the ’06 budget will take on more energy and move along much
quicker.

So that’s kind of where we are with the budget situation.

Let me turn a couple – to a couple of statistical and population-based
data initiatives that the Department through the Data Council has been focusing
on in – over the past few months, and here we’re hoping to make some
enhancements and some improvements.

In the first area, we began in September to look at what was our current
capacity for prescription drug data in our major surveys and administrative
data systems. This was associated with the Medicare Modernization Act
prescription drug benefit program but more broadly as well, and we at a
September 8th Data Council meeting brought together all of the
agencies across HHS as well as some of our policy research agencies on the Hill
and in other Departments and we looked at what the current capacity was.

There are only a few surveys that actually get much detail on prescription
drug information, prices and so on, but we looked at ways that we could work
together. We had some follow-up discussions about how – where the major
gaps were and how we could work together to address some of those gaps in our
major surveys, and I think in the months ahead we’re putting together a set of
proposals for improving that information and we’ll be working on that in the
months ahead.

In September, under the Data Council auspices, we convened HHS and other
federal agencies to look at national health insurance data and data on the
uninsured and how we measure it and gaps in that area and ways we could work
together.

You all know that every season there are at least three estimates for how
many uninsured there are in the U.S. In a way, their – the surveys all
answer somewhat different questions, but it is somewhat perplexing sometimes to
sort it all out, and we’re looking at how we can make better sense of that, how
we can some improvements, and how – there’s a fair amount of imputation
for the researchers around the table done in these areas and how this all works
together.

So we will be, in the months ahead – again, we have some follow-up
activities. We’re looking at how we can address some of the gaps there, how we
can have a better framework for what these various estimates mean, and how we
can move forward maybe on modeling and some analytic and methodological work as
well.

In a corollary, in October, in – Dr. Cohn has been participating in our
Data Council meetings, so he’s aware of these developments as well – on
October 13th, I think, we looked at – we followed that
discussion to look at health insurance information and access and related
health care data at the state and local level.

And here we looked again at what the current surveys and other data sources
were for providing that kind of information, what were the limitations, what
were the strong points, what were the gaps, and is there a way to proceed, and
I have to say there we – there are clearly more challenges in the data
area for states and local areas than on the national area. And there was a
point reached at which even the federal role is questioned about how far down
in terms of estimates do you go here. And we do have some surveys, the current
population survey and others, that actually provide some state data, but that’s
clearly more resource intensive. But we’ll be looking at ways we can address
some of those gaps as well.

And then finally, in a couple of related areas, we began a project late in
the fiscal year to look at the extent to which our surveys and other
statistical systems are using data standards such as the ones we’ve talked
about here, not so much message format standards as much as terminology
standards, as well as statistical standards for those researchers around the
table. There’s a whole set of standards for how you do things –
statistical practice, classifications and so on. It’s the extent to which our
surveys are using those and what are the opportunities if they’re not using
them for moving forward.

And then, finally, on the privacy front, the – we are – we have
open at the moment a recruitment for John Fanning, who retired a couple months
ago, so we’re hoping to get some good candidates and we’re hoping to hire
someone to cover the privacy policy area fairly soon. The position is open and
it’s described on our hhs.gov/careers website, and I know some of you have
already shared that announcement with some candidates and we’re very
appreciative.

DR. COHN: Any questions? Vickie?

DR. MAYS: A couple questions. One, there was some discussion about a
geocoding project that I think was underway; if you could update us on that.
And the other, in terms – I don’t know whether or not this is something to

ask of you or whether or not this is something to go directly to NCHS, but
in terms of, if they get the budget augmentation, there is this discussion
about them – some of the money for methodological work, and I’m wondering
if there – if it was – if in the request, whether or not you have
specific things they were asked to do or if that’s something that’s up for
discussion.

MR. SCANLON: Let me start with the first, Vickie. The geocoding project,
we’re running that out of – under the Data Council auspices. Dale
Hitchcock is our project officer.

And I think Dale told me yesterday that it was pretty much almost finished,
and we were proposing to brief the Data Council, and I think we’d be happy to
brief the NCVHS or the populations subcommittee, however you’d like to do that,
at any time in the couple months ahead.

That project actually worked very well. It’s a way of kind of integrating
through – for geographic kind of information all of the holdings –
this is primarily in the environmental area where there are physical –
where literally you have a – you have geographic coordinates, you can
literally have GPS coordinates, other ways of describing where something is
located. But there’s – that’s sort of the most detail level. Then there
are zip codes, there are block groups, there are counties, cities

and so on, so you can aggregate a – but it’s actually a nice way of
approaching standards for collecting and displaying geographic level data.

On the methodology program, Jack, I’m remembering that our request was
fairly broad; we didn’t describe any particular methodology projects. I think
we’d clearly be open to – I know the full committee is in the process of
pulling together research and methodology ideas, so I think – I won’t
speak for Jack, but I think we’d probably be interested in your ideas.

MR. ANDERSON: Yes, Jim is correct. The request in general was to shore up
the Center’s programs, not relate to starting new activities. If we don’t get a
sizable increase, we’ll probably actually have to actually cut out one of the
major data systems.

So we’ve been shoring those up over the past several years, trying to keep
them going. And one of the things that has suffered has been the methodological
research program as we’ve taken money – have not put money into that,
‘cause we put every dime we’ve had into actually trying to keep the data
collections going.

But Jim’s actually absolutely right. We would hope to use some of that money
for methodological research if we get it, and we were not specific in the
request as to specifically where it would go, so we are quite open as to input
from this group and others as to what might be the best use of that.

DR. COHN: Okay. Jeff, then we’ll all move onto the next presentations. Jeff?

MR. BLAIR: Jim, you referred to the funding that’s available for data
exchange networks –

MR. SCANLON: Right.

MR. BLAIR: — and could you help me sort some things out?

MR. SCANLON: Sure.

MR. BLAIR: As I started to look at a number of the demonstration projects
and funding projects and initiatives in the federal government, I was trying to
pull them together, and I kind of started with the idea of the local health
information infrastructures and now where I’m not sure it’s the same thing, but
you were referring to the regional health information organizations now.

MR. SCANLON: Right.

MR. BLAIR: There’s the data exchange networks, there’s telemedicine, but
then there’s a whole number of other initiatives that seem to be considered
within the domain of regional health information organizations, including
e-prescribing, possibly the docket program. Could you help me out? Are data
exchange networks the same as RHIOs? Are they subsets of it? How does this all
fit together?

MR. SCANLON: You’re asking for a unified theory of national health
information technology –

[Laughter.]

MR. SCANLON: — and we actually hired someone to – [laughs] –
David Brailer will be with us later today, and I think David’s working on a
strategy. He’s working with the agencies on a strategy for how to sort this all
out.

Yes, there are some – all of the things you’ve mentioned, Jeff, they’re
very – those are pieces of the overall framework. I’m not sure the
terminology has changed or is changing, but I think – I will defer to
David when he gets here later. He’ll bring you up to date on where we are with
moving forward on the whole health information technology including the data
exchange networks and the infrastructure, the backbone, and things like that.

The e-prescribing is a somewhat – I mean, obviously it’s all related,
but e-prescribing, as you’ll learn from Karen and others, that’s obviously on a
faster track and a more direct track in support of the Medicare Modernization
Act prescription drug benefit which begins in January ’06, but I think –
so that’s somewhat – that’s moving ahead, and – but it’s being
envisioned in a way that would fit in with all of the other –

MR. BLAIR: That would work – yes, exactly.

MR. SCANLON: — and it could be —

MR. BLAIR: In other words, someone that has a regional e-prescribing network
might be – might qualify as an RHIO funding –

MR. SCANLON: Well, we’re not even – the Medicare part is being
considered without regard to data, so –

MR. BLAIR: Yes.

DR. COHN: Okay. Well, with that – and Jim, let us know when you have
the relative – the unified theory together – I would like to hear
that.

[Laughter.]

DR. COHN: Okay. I think we’ll move on, in the interests of time. Karen
Trudel, you’re presenting for CMS on HIPAA and a variety of other issues.

Agenda item: CHI Update – Ms.
Trudel

MS. TRUDEL: Right. Good morning. Just to provide a little bit of context,
I’ll go over some of the implementation dates, compliance dates that are coming
up shortly; I’ll talk a little bit about regulations that are in process, the
Medicare compliance rates for claims, et cetera; talk a little bit about
complaints and also on e-prescribing and CHI.

With respect to the compliance dates that are coming up within the next few
years based on regulations that are already published, we have the security
compliance date coming up in April of 2005 which is for most plans and
providers, 2006 for small health plans, and the national provider identifier
compliance date in May, 2007, and again, 2008 for small health plans.

We have the claims attachment standards proposed rule, which is admittedly
moving somewhat slowly, scheduled for an early 2005 publication; the national
health plan identifier around spring, and modifications to the transactions and
code sets, the second round of modifications this summer. And I’ll talk a
little bit about the e-prescribing rule in a moment.

In terms of Medicare’s own compliance experience, we’re now receiving more
than 98 percent of our claims in a totally compliant format, and at this point,
62 percent of the receivers are in production with HIPAA compliant 835
Remittance Advice, which is a great deal of progress. We’re very pleased with
that.

In terms of complaints, we have received a total of – and I know Sue
is going to scoff when I say this; she’ll definitely trump me by orders by
magnitude – 247 in total, and of those, 110 are open and under
investigation. Seventy-five percent of the complaints that we’re receiving are
against private sector entities as opposed to governmental organizations that
qualify as covered entities under HIPAA.

The complaints that we’re receiving generally allege that a compliant
transaction was rejected, a noncompliant transaction was sent, that there was a
companion guide that violated the implementation guides – in other words,
they did not agree; that appropriate code sets were being rejected, or that
there were disputes between trading partners, and for the most part, those had
to do with excessive telecommunications fees.

We’re currently monitoring for corrective action plans, but we’re finding
that our parties resolved most of their complaints before we reached the point
of corrective action plans.

Just in terms of outreach and the kinds of inquiries that we’re receiving,
it’s kind of interesting to note that our hotline is now receiving about 1200
calls a month. Contrast that to about 5,000 in the August to October period
last year and 15,000 or so in the October period the year before because of all
the requests for exceptions from the folks from ASCA.

Let me talk a little bit about electronic prescribing because I know the
Subcommittee on Standards & Security members and the full Committee have
labored long and hard over the summer to come up with the recommendations for
the Secretary.

We are indeed pursuing, as both the Secretary and Dr. McClellan have
publicly stated, a proposed rule that would propose the set of foundation
standards in keeping with the NCVHS recommendations. There would be a hope of
implementing those standards in January of 2006 as the Part D plan, actually
going to – into operation, and the regulation is moving with a
considerable amount of speed. You know, I do believe we will be seeing it in
the relatively near future.

We also have a contract where we have someone gathering information on the
characteristics of current e-prescribing implementation and we’ll be using that
language – that information to construct a strategy for the pilots that
will take place in 2006.

Just a short word on CHI. As I think I’ve mentioned before, it actually has
two — has three focus areas. We’re working on implementing the standards that
we’ve already adopted, maintaining those standards, and focusing on needs for
new standards in areas where there are critical business needs in the federal
sector. We’re working with Federal Health Architecture to coordinate our
activities and our resources.

And I think that’s about all I have to report, so I’d be happy to take
questions.

DR. COHN: Questions? Michael?

DR. FITZMAURICE: Karen, in your commentary,

you said that 98 percent of claims were totally compliant. I think that’s a
remarkable achievement. It means they’re compliant with the HIPAA standard,
right?

MS. TRUDEL: Yes.

DR. FITZMAURICE: Remarkable. Any complaints from payors that – I’m
sorry; from – I guess, from payors that providers are having a hard time
getting the standards up to the payors? I mean, what I’m hearing sometimes is
that the health plans and the payors are ready to go and that they’re having a
hard time getting the providers to get to that place and that in between there
are people who are just turning it around and putting in what’s necessary to
get the claim paid. I haven’t heard that with the Medicare program, but I’ve
heard that with other programs. It seems like the Medicare program has made
remarkable progress and that we can expect similar kinds of progress in the
private sector. Is that your opinion, too?

MS. TRUDEL: What I have heard, just anecdotally, is that the primary focus
in many cases in terms of testing and implementation was placed on Medicare and
that once Medicare implementation was accomplished, especially in terms of
large vendors, clearinghouses et cetera, then the focus went to Medicaid, Blue
Cross Blue Shield et cetera.

So I would say that what I’m hearing, and I’m

sure Harry can attest to this as well, is that other plans,

Blue Cross Blue Shield plans, et cetera, are not experiencing levels that
are as high as Medicare’s but that their numbers are steadily going in the
right direction.

MR. REYNOLDS: Yes, I would concur with that. We’re running right in –
right now at 66 percent, but we have all of our trading partners that actually
would send them to us are in production. It’s just that they’re having to go
around provider at a time and now that they’ve got Medicare done, and then
they’re looking at Medicaid, then they start looking at the commercial
business.

So it’s – you got a little bit of a pecking order, obviously. Medicare
was the big dog in this fight –

[Laughter.]

MR. REYNOLDS: — and so I think it’s key, but it also – I’m also seeing
similar increases in the other transactions. Our membership inquiries have
grown dramatically. Our claims inquiries are growing.

So it’s starting to – as vendors really click in, they’re bringing
large numbers of providers to the table, so you may be waiting for providers,
but as that vendor actually gets those providers ready, you’ll get a chunk of
‘em.

So one – we have one clearinghouse that sent us 20,000 membership
inquiries. So as you pick each of these

pieces up, your volumes start jumping dramatically. So

we’re comfortable we’ll be approaching Karen’s number because we actually
have 93 percent of all the providers signed up to be in production; it’s just
how fast they come on now that they’ve gotten Medicare out of the way.

PARTICIPANT: I would concur.

DR. FITZMAURICE: I have heard, too, that the thing that people like most is
the remittance advice, that they’re getting more accurate remittance advices
and it’s broken down more finely. And that seems to be a benefit of HIPAA
standards that they wanted, they expected, and they’re beginning to get. They
do like that.

MS. TRUDEL: That’s excellent, and I would also point out that prior to HIPAA
there was no requirement that the remittance advice balance, and obviously if
it doesn’t balance, you can’t use it to post very effectively. So I think
that’s another big plus.

DR. COHN: Good point, Karen. Thank you. Hear one final question and then
we’ll – Russ, did you have a – and then we’ll move on to the last
–

MR. LOCALIO: This is Russ Localio. When we met in September, there was some
discussion, and you’re going to see Steve – oh, there’s Steve; okay,
Steve’s here, okay – good. Back me up, Steve!]

[Laughter.]

MR. LOCALIO: There was some discussion, and I

think I voiced some concern, that although we have pilot objects in
prescribing coming up in now a matter of 14 months, I was not aware that there
was anywhere in the federal government yet available a list of all of the drugs
that might be prescribed.

PARTICIPANT: All of the – drugs, did you say? Drugs?

MR. LOCALIO: Yes, prescription drugs that might be prescribed. It would have
the drug in NDC code and the drug class.

Now, the FDA list, to the best of my knowledge, is incomplete, and those of
us who work in research buy databases from the pharmaceutical industry
‘cause it’s the only place where you can get complete lists. Where the
– my – am I totally out of touch, or do we have something in place,
or is there going to be a rule coming out that says, As of today, we have the
database ready and this is what — these are the possible drugs that can
prescribed? How is it going to be maintained, on a daily basis? Or are my
questions out of turn?

DR. COHN: Well, maybe I’ll start out, and then Karen? Yes, Russ, actually I
think you actually passed a letter in September which I know you voted in favor
of that addressed some of these issues, and I think it’s actually been a full
committee observation that the NDC files are in a – somewhat of a state of
disrepair, which is what I think you’re observing.

MR. LOCALIO: I just wanted to know –

DR. COHN: What’s happening.

MR. LOCALIO: — do we have any progress as the clock ticks?

DR. COHN: Yes, Mike, did you want to answer?

MR. LOCALIO: I’m very concerned.

DR. COHN: Yes. Mike, did you want to answer that one?

DR. FITZMAURICE: Well, as Jim mentioned earlier, he and I are in charge of a
$10 million investment in data standards. It’s the first time the government’s
ever had this kind of initiative. And a good part of that initiative has gone
to reduce adverse drug events, and in order to do that, you need to know what
the drugs are.

Seventy percent of that money went to FDA and the National Library of
Medicine. And one of the products to come out of that is to be an electronic
drug listing and then followed by a listing of all the products that FDA
approves. That has been started. I met with FDA yesterday, and that is ongoing.
By ongoing I mean they are starting the system and then throughout the course
of ’05 they will begin filling in the system. We think some of these things are
critical to – for e-prescribing, and so it was imparting a sense of
urgency — and to find out that they were making good progress that we met
yesterday.

MR. LOCALIO: Thank you.

DR. COHN: Great. That’s a more reasonable answer in my – yes.

MR. LOCALIO: Thank you.

DR. COHN: Well, actually, Russ, you should keep asking that question
probably every quarter just to see what happens, ‘cause we’ve been buggin’
‘em about this one for years.

MR. LOCALIO: What I want to do is when it’s online, you should get a demo
and then have a party.

[Laughter.]

PARTICIPANT: It will be a big development –

DR. COHN: Yes.

DR. FITZMAURICE: Well, part of it is that they are to take that information
and give it to the National Library of Medicine. It’s also posted or referred
to on the daily med website, so there’s coordination involved as well as
producing it.

DR. COHN: Yes. Steve, do you have a comment? And then we’ll move on to the
next one.

DR. STEINDEL: Yes, just as a quick comment in support of what my other
federal colleagues have said.

There’s been progress being made; a lot of it is behind the

scenes, but it is moving forward.

But I think a good public example of the progress is at the Med Info meeting
in September, the National Library of Medicine announced their first version of
a browser for RX-Norm which does – it doesn’t integrate the NDC codes yet
but it does integrate the prescribed – prescribable drugs, and they know
clearly that this is a very preliminary version.

DR. COHN: Okay?

DR. STEINDEL: At least they can demonstrate it; that’s good.

DR. COHN: That’s good. Okay. Thank you. Karen, thank you very much for the
update.

Our last Department presentation is from Sue McAndrew. Thank you for joining
us.

Agenda item: Privacy Rule Compliance
Update – Ms. McAndrew

MS. McANDREW: Sorry I was late, but I will keep this brief.

The complaint status is a little preliminary at this point for the
complaints through October; usually we allow the regions up to six days at the
end of the month to firm up their numbers. So when – these are a little
preliminary.

But as of now, we’re showing a total received of 9,277 complaints. We have a
– we are maintaining a closure rate of 60 percent. There’s been no major
shift in the nature of the complaints; it is still predominantly impermissible
uses and disclosures followed by concerns about safeguards, access, minimum
necessary, and authorization.

There’s been no shift in terms of the nature of the entities against human
complaints. Our main – it still is predominantly private practice
physician offices, general hospitals, outpatient services, pharmacies and group
health plans.

We are – we were able to compare the volume compared to October of last
year, and we are running actually a little below the total number of cases that
were received in October, so whether this is just a matter of preliminary
numbers or at least to stabilization, which we would gratefully accept, it is
not continuing to trend upward, which is hopeful ‘cause we may get a
chance to catch up and close the more cases.

The closure rate was significantly up from last year, and the – among
the closures, we are typically running – about 25 percent of the closures
are for jurisdictional reasons. The rest are either nonjurisdictional or other
closure bases.

In terms of policy emphasis, we are continuing to work on a number of
materials for the website. We have been working with NIH to get out another
major component of their research fact sheets and FAQs in terms of how this one
will address the privacy rule in relation to human services research.

We are also continuing to work hard and are getting close to technical
assistance materials directed toward small providers and are working on FAQs to
address a lot of questions that have been accumulating in the litigation area
– how covered entities can use and disclose information when they’re
involved in litigation or have been subpoenaed for their records in relation to
somebody else’s litigation process.

Other than that, we are continuing to, as – both because of the large
volume of complaints that we get in the safeguard areas and with the moving
toward the security rule coming on line and the security complaints that CMS
will be handling, we have stepped up our efforts to coordinate with Karen’s
staff to work out how we and CMS will be moving forward on complaints that
overlap privacy and security and how to sort out those that are strictly
security complaints.

And we are also getting – continuing to push through clearances on the
enforcement rule which will take the interim procedural enforcement regulations
and expand them to encompass all aspects of the security – I’m sorry;

the enforcement process, and this will be a regulation – a set of
enforcement rules that will be common both to OCR and to CMS.

And that’s – and in the meantime, we’re doing whatever needs to be done
on the – [laughs] – preparing for e-prescribing and the other
onslaught of technology initiatives.

DR. COHN: A good summation. Let me ask one question, and I know Mike has
another question, which is: You’ve mentioned the enforcement role –

MS. McANDREW: Yes.

DR. COHN: Do you have a view of when the proposed rule might come out on
that?

MS. McANDREW: I have ended prognostications on – [laughs] – when
that is out. It is getting close; we are hopeful. Hopefully, by the end of the
year.

DR. COHN: Oh, okay. Thank you. Mike?

DR. FITZMAURICE: I will say that recently I was able to review – of a
health services research and the privacy rule. What I’m seeing in the things
coming out is there is a consistency across the guidances and a further
delineation of where the lines might be drawn or where lines can’t be drawn,
and so I think the public cannot wait to get, with good bated breath, the
– these guidances, because they are helpful and they help in making
additional decisions beyond just what you would read in the privacy rule. They
don’t change the law, but they give you guidance in areas where you really have
questions.

DR. COHN: So, thank you, OCR.

MS. McANDREW: Well, thank you.

DR. COHN: And then Jeff?

MR. HOUSTON: Well, I guess the last time, or two times ago, when we met, we
talked about getting some type of detailed statistics. I know I’ve given you
some of my thoughts on that. How close are we to getting something together or
having more detailed statistics available?

MS. McANDREW: We have done some preliminary runs off of our database to try
to refine and correlate some of the statistics. I think we are still –
we’re having two sets of problems. One, there are other time demands on our
system’s people, so we get a small slice of their attention.

And the material, the preliminary runs, have not been very user friendly.
And I – so the first couple of times we looked at these, we kind of
scratched our head and need to do some further investigation to either do some
clean-up of the data which may be making it look like – or work – get
more of the time of the technicians that are running the system, to say, Can’t
you refine these correlations because they’re not adding to any information
–

MR. HOUSTON: Now this is – sort of as a member of the privacy
subcommittee, I mean, I’m personally very

interested in knowing where – what the trends are because I think
that’s going to help shape where we go with some of our hearings and things we
want to try to focus on, I think, so, you know, the sooner the better from my
perspective, and I understand the challenges, but I think it’s really important
to try to get to some understanding where those debate issues are still.

DR. COHN: Okay. Jeff, and then Richard Harding.

MR. BLAIR: The privacy regs were put in place in order to build public
confidence and trust, that we move health care into the information age, the
public can trust that their personal health care information will be kept
private, confidential and secure.

While your briefings to us are relevant in terms of the administrative
accounting of people who are complaining within the system, I am frustrated and
concerned that we’ve lost sight of the basic goal and we’re not keeping track
of whether that goal is being achieved and the consequences of not achieving
that goal could be extremely severe.

Has there been any survey of public understanding, awareness, of the privacy
regs and whether

or not the privacy regs have moved the ball forward to having the public
trust health care information in electronic form. If not, it is a wonderful
vehicle to build that trust, but so far, I don’t see us using this regulation
to do the job it was intended to do, and that is, to build trust in the public.
Has there been any attempt to find out if the public understands this and what
their attitudes are?

MS. McANDREW: I think to the extent the – I think both the complaint
system itself taps into many of the attitudes of the public to the rules, we
also have and continue to get and handle a lot of correspondence from the
public, although I must say I don’t – I think the volume of that is
significantly down from when the rule initially went into effect. But we don’t
– and we have just recently put out some consumer oriented fact sheets and
representations – summaries of the individual rights that are given under
the rule so that there will be some – I mean, we have clearly heard that
the notice process was too – became too technical and the documents were
far too lengthy to be meaningful to anyone other than someone who was already
privacy oriented.

So it is – we are hopeful that these consumer oriented pieces will go
far to explain the rule in terms that the average consumer can better
understand. In terms of actual formal surveys on consumer attitudes, no, we
have not engaged in any of that.

MR. BLAIR: Well, let – you know, let me just indicate that I feel like
if you don’t have that information, we could be winning the battle but losing
the war. So I would just encourage you.

DR. COHN: Okay – Jeff, you have a – I would suggest that you might
reconsider joining the privacy subcommittee, based on that question.

We have time for one more question, and I think, Richard, you’re on for
that. And then we need to move actually – I see David Brailer here and we
need to move into his session. We’ll do the Scientific Counselors session after
the break at that point and just — so, Richard?

DR. HARDING: A typical question – Richard Harding, a member of the
Committee — that is asked of me all the time, one is, what percentage of the
complaints have merited a referral to the Inspector General at this point? And
the other, have there been any referrals for omissions as opposed to
commissions in the complaints? That is, someone who didn’t do all the things
they were supposed to do as opposed to voluntarily not doing things or
something like that.

MS. McANDREW: Hmm –

DR. HARDING: Those are questions that come from general doctors a great deal
to me when they find out that I’m blessed to be on this Committee, but –

[Laughter.]

MS. McANDREW: One – no, we don’t make referrals to the Inspector
General.

DR. HARDING: Pardon me, the –

MS. McANDREW: Did you mean DOJ?

DR. HARDING: Yes, yes, pardon me.

MS. McANDREW: We have referred I think it’s a little over now 140 cases to
the Department of Justice, so that’s whatever percentage of 9,000 –

DR. HARDING: So a little over 1 percent of the complaints, okay.

MS. McANDREW: — and in terms of – I mean, it is possible within the
body of the complaints that if the – if an entity has failed to do
something that they were obligated to do, that can give rise to a complaint as
much as someone who has actively violated, but by and large, I would imagine
that if it is simply a failure to have done something – if you’ve
forgotten to put your policy in writing – that those are fairly easily
addressed through voluntary compliance, and so in all of those cases, we would
look to voluntary compliance before taking any kind of –

DR. HARDING: And then finally, when will the health research guidance be
available that was mentioned?

MS. McANDREW: The health research guidance? Once – it is still in the
– going through Departmental clearances, but once it is done, it will be
on the NIH privacy website which is – and there will be a link to that on
the privacy website.

DR. HARDING: So that’s the next few months or –

MS. McANDREW: I would expect –

DR. HARDING: — a year from now?

MS. McANDREW: — it should clear this month.

DR. HARDING: This month? Wonderful. Thank you.

DR. COHN: Okay, thank you. Okay. Well, thank you all very much. I think on
the basis of this, maybe next session we’ll make sure we have a little more
time for the Department updates, so we’ll talk about that with John.

With that, let’s change, and David, welcome. For those on the Internet, our
next session is with David Brailer, who is the National Health Information
Technology coordinator. David, thank you very much for joining us. I’m looking
forward to hearing about updates and new initiatives that you have underway.

Okay?

Agenda Item: Current Initiatives in National Health
Information Technology – Dr. Brailer

DR. BRAILER: Thanks, Simon; thank you all for taking time to hear about what
we’re doing on this rainy day, for those of you on the Internet.

We have been busy and have made good progress, and I thought what I would do
would be tell you just some of the activities that are underway and some of the
things that have, or are, coming to fruition, and I’ll be brief and I’ll be
happy to address comments or questions.

As you know, we long ago now, in July, released a strategic framework and
laid out four goals, and one of the first things that we did was organize the
activities of both our group and the agencies that we’re working with along the
lines of those strategic lines so that we could all take a task basis and make
sure that we were able to keep focused on specific goals. And let me just
review some of these, and I’ll talk about some more cross-cutting issues.

The first is on informing clinical practice. And as you know, then, and when
I gave the update here before, we were concerned particularly about the risks
that were inherent in the HER purchasing decision on the part of small
clinician offices in particular and endemically throughout the industry and
felt like there were three tasks that needed to be thought through and
implemented, not necessarily by government or only by government.

The first was developing product certification that allowed a minimum floor
to be placed on what kinds of product features were deemed adequate for an EHR
to be used. Secondly was mechanisms that reduced purchasing finance risk. And
third were mechanisms that reduced implementation risk.

And I’m happy to say that since that time, the Certification Commission for
Health Information Technology, CCHIT, has been formed. Mark Leavitt, who is a
name you know as a well-known entrepreneur in electronic health records, now an
officer of HMISS, is acting as the Chair of the Commission that has 14
Commissioners that are drawn from physician groups – purchasers, payors,
technology companies, other industry associations. They’ve now had their second
public meeting; the third is scheduled for just within a month.

And the – this group, I think, has made remarkable progress towards
forming four work groups to begin developing both the criteria and the
mechanisms for application. Those work groups are Functionality, who’ll be
looking at what features and user functions should be present, how would they
be assessed; Interoperability, which is not just standards but the other
components for patient controlled information exchange; Security and
Reliability, which is the whole life cycle of protection; and then an
Inspection Process – there’s a work group only on the process, about how
this would work, what the issues would be, et cetera.

So there is a public nomination process underway for these work groups, and
I think they’ve publicly announced this, and we’ve been supportive of that, and
I’m encouraged that the Certification Commission is moving forward. I think
it’s going to do something important. We are still in the mode of watching it.
We haven’t formally related to this group, although my office has designed two
ex officio government employees to sit with the group not only for guidance but
for input for us.

But we’re also exploring mechanisms by which, when this group crosses the
line of delivering some of the key outputs that it will, how we’ll formally
relate to it.

We’re still making progress, although it’s a little bit behind the scenes,
on purchasing risk reduction. As you know, there’s a group that’s been formed
called the “PEHRC,” Physicians HER Collaborative. It’s still
exploring mechanisms by which it will relate to – it will develop some
risk reduction strategies.

There’s also an entity that’s the National Purchasers Alliance that the
Office of Personnel Management and CMS are involved in that are looking at
financing schemes, particularly regional financing, the idea being if a
Tennessee, for example, since I’m going there tomorrow, will develop a regional
project, how can the regional payors synchronize their support of HIT adoption
on the part of providers? This seems to be an issue in terms of how the regions
operate.

And then in the implementation risk reduction area, I think this is a topic
where many of you know there’s been a lot of discussion about how to bring
resources to the table for that. And this is very linked in with what we’re
doing with the regional organizations, and I’ll come to that in just a second.

So we’re making good progress on informing clinical practice and I think
that there’ll be a variety of things happening over the course of the next
several months.

Goal Two is interconnecting clinicians, and central to this is formation of
the regional health information organization. I know many of you know already
that AHRQ released a series of contracts to state entities and to other
regional entities that are, if you’ll pardon the twisted phrase, proto-RHIOs.
They’re regional health information organizations in formation. I say that
because we’re still working through a process of what it is that the regional
organizations need to do or be, but we obviously don’t want to hold up the
march of progress.

So one of the great urgencies now is: What are the criteria that would
establish a regional health information organization? And what benefits does
being a regional health information organization convey upon the participants
that they individually or loosely collaboratively couldn’t gain themselves?

And so we formed a variety of internal work groups that are looking at how
the regional organization relates to Stark, to antitrust issues, to tax issues,
to a long variety of issues, to determine if there’s any standing that can be
given to them and if so, what criteria need to be – conditions precedent
in the RHIO to be able to accomplish that. We have nothing that’s been
announced. It’s still a very detailed exploration of simply what the statute
says and what other interpretations have said, but it’s our goal to be able to
lay out very quickly at least a blueprint or what the governing rules around
RHIO should be in time for the next round of grants which – and contracts
which will be announced in March or April, I would presume, and, oh, sometime
before the end of the FY ’05 fiscal year.

So that is an urgent activity. There’ve been a lot of discussions on this
topic and there’s still more to be done, but we have been impressed by the
level of activity here in the private sector. The number of organizations that
are forming regional organizations are in excess of any federal monies that are
flowing, and we want to make sure that we give guidance so that there’s not
risk posed to these organizations by some future set of criteria or other
guidelines that could cause reengineering and this slows down this movement. On
the other hand, we don’t want to see the movement kind of spin off and become a
series of semi-unrelated regional entities.

The other part of this is the National Health Information Network, which is
the key components of – that allow interoperability to occur. And as you
know, we – our first step in this is the forthcoming RFI, and this is
imminent; I think this has cleared all the organs and is waiting for the cue to
go out into the Federal Register.

And the RFI is intended to do two things. First, ask questions so it can
have facts on the record about what are some of the technical needs, what are
some of the technical gaps, what are some of the economic, cultural,
professional, legal issues that would be encountered in the development of a
set of tools that can be available to allow information to be shared widely.

And this RFI, I think, is the first of a multi-step process of understanding
how this can be billed, what role government will play, if any, in this, and
how it is that we can make this a treasure and public asset for achieving the
goals that have been laid out here with health information.

Clearly, this is near and dear to the hearts of NCVHS, and so I hope that as
the RFI comes out, you will get on the record various testimony or comments
that can supplement what we get on a one-up basis and we’ll be able to share
some of that discussion because we’re looking at a

variety of scenarios that could open this, these facts, or inform, and I
think one of the real judgment calls can be what are the next steps towards
moving this agenda forward, and I think it will be a complex set of activities.

So that’s interconnecting clinicians.

In terms of the personalizing care goal, we have continued our efforts to
evaluate mechanisms for expanding and accelerating the release of personal
health records to consumers, and I’ve been impressed by the number of different
organizations and entities who have interests, support, capacity for this, and
very much see this as a goal that is – in many ways is moving forward, and
we’re – the role we’re playing is frankly more of an observer.

As this was happening, we have not developed a specific strategy of
formulation in that to advance what’s happening although I think we’ve learned
a lot about what opportunities there are for making sure that personal health
records are themselves interoperable, that they are truly person centric, that
they need a standard of being usable by consumers, and that they can insure
that they gain public trust as well, and we’re looking at that.

One of the elements of that is the issue that this strategic framework
raises, which is privacy and confidentiality. And I speak about this not in the
HIPAA sense but in the sense of the public trust and concern and

the technological issues around authentication and authorization, and so we
formed an internal working group that is discussing kind of where we are with
this, but we very much welcome the input that I know will be coming from the
privacy work group at NCVHS because it’s our sense that there’s a social agenda
contained in this that needs to be well defined as we move forward with
interoperability, and so we want to make sure that we understand the issues
that that raises.

We’re still involved with efforts to streamline quality metrics. As you
know, there’s been a burgeoning discussion about how to unify these, that our
strategic framework was one of many comments on that question, and that’s
gained some traction and we’re still supporting it and watching that along with
CMS and some of the other agencies.

In terms of population health, we are at work developing white papers in
collaboration with AHRQ, AMIA and HEMS around how structured clinical decision
support will work, how knowledge bases can be created, parsed out, made
available, integrated in with electronic health records et cetera.

This is a highly complex problem not just technically but in terms of
knowledge engineering but also one that’s necessary if we really want to have
the goal of – that has been laid out both by us and NIH of closing that
adoption gap of what is know versus what is in practice or the – you know,
the gaps that you know well from Beth McGlynn and others about what share of
physicians follow available evidence. Ultimately, this becomes an exercise of
how to make the right thing to do the easy thing to do, and that means, how do
we engineer knowledge distribution into both the networks that we’re building
and electronic health records? The starting point of that is even envisioning
how this could work, with scenarios with the – even easily.

So this is underway, and I expect something to come from that in the early
spring as we understand what is even possible here, what other industries have
done, and what some of the early demonstration projects that are happening in
the private sector teach us about the broader question.

We’re also involved with a variety of efforts with other agencies, NIH and
the National Cancer Institute, looking at population health improvement topics
with FDA, with the Department of Homeland Security, CDC, et cetera. There are a
variety of discussions underway about how it is we can take the appropriate
next steps along – according to the population health goal.

So those are a quick update on where we are on a goal basis.

In a cross-cutting sense, there have been a couple of things I’ll just
discuss. And one of those is that since we met last, Federal Health
Architecture has been fully integrated into our office, and we see enormous
potential in terms of federal alignment from this process. Federal Health
Architecture is part of the eGov health line of business of management
techniques in the federal government and CHI, as you know, has become an
integral part, according to eGov and Federal Health Architecture, and we really
want to take these to the next level and build on their very promising paths.
So that is underway.

The Federal Health Architecture has the following work groups:
Interoperability, Electronic Health Record, Public Health Surveillance, and
also, showing the cross-cutting nature of it, Food Safety. There are other work
groups that are being contemplated, including some that are components of our
strategic framework as well as others that are, again, part of the broader
health line of business initiatives in the government.

Also, some of you might have seen a very brief announcement that the OMB
scorecards that get passed back to the President’s management agenda to all of
the agencies in the federal government including language that asked for
reporting on agency plans to conform to the strategic

framework and for an analysis of where there are variances or gaps with
respect to that framework. I think that, plus some of the pass back language
that is being discussed, is very promising in terms of operationalizing this
framework into a real budget instrument, a mechanism for causing discussion
about where we see duplication, where we see opportunities, where there are
synergies that can be exploited that haven’t come to rest – or haven’t
been identified to us yet. So that’s moving forward.

Also, we are moving forward with the Health Information Technology
Leadership Panel. Those of you who have followed this know that this is a group
of executives who are not of the health care industry; they are mainline
industry or financial companies who are – we’re asking to give us their
expertise on whether or not health care as an industry deploying widespread
information use would be expected to follow the same experience of substantial
productivity improvements that their industries did. And this group is coming
together and will be giving us their insight as has happened in some other
countries.

And we’re still involved in our support of the President’s 2006 budget
analysis that’s still underway in the government.

So that’s a quick update on things that we’re doing. I’d be happy to answer
your questions or respond to comments, and again, I just want to say thank you
all for your leadership and partnership. We have a long way to go and we
couldn’t get there without our close working relationship with you.

DR. COHN: Well, David, I think I’ll first of all start out really not so
much with a question but as you’re talking, as I’m taking pages of notes here
just noting all of your activities, I just wanted to – I think we ought to
think how impressed we all are and congratulate you. I have a hard time
reminding myself that you only came on mid-year this year, and I think the
progress you’ve been making has been nothing short of phenomenal, so –

DR. BRAILER: Oh, thanks.

DR. COHN: — we want to congratulate you.

DR. BRAILER: Actually three years in the past six months.

DR. COHN: Yes, exactly! I think that’s certainly the observation one would
make, so – I mean, great work. Questions from the full Committee? Mark?

MR. ROTHSTEIN: Thank you, David, for that presentation. I want to assure you
that the Privacy and Confidentiality Subcommittee is planning hearings this
winter on the issue of patient controlled limits on the contents of AHRs.

I – the issue, as you know and have expressed, is a very fundamental
one, and as I see it is something along these lines. Patients are, or many
patients, would be reluctant to participate in a system with a longitudinal
electronic health record if they didn’t have some control over the contents of
that health record.

On the other hand, many physicians and other providers are reluctant to
provide services to individuals who have expunged their record, making it very
difficult to provide quality health care, exposing them to liability, requiring
them to take additional information, and so forth.

So, we need to figure out how you can accommodate these physicians in a way
that won’t totally destroy the viability of the system, and that’s quite a
challenge for us as well as you, and we hope to get to work on that this
winter.

DR. BRAILER: Thanks. I welcome that because I, as you, when I look out and
see the real fundamental bridges that need to be crossed to get where we’re
going with interoperable, widespread use of electronic health records. That is
one.

And, you know, the public confidence in this, which I’m sure when
scrutinized will be acknowledged, that this is – these technologies have
the capacity for preserving privacy in a way that paper cannot raises the

relationship issues, as you described, between doctor and patient, but many
other issues, and we want to make sure that we understand the consequences of
these and then we can ask for the public confidence in a way that is based on
good fact and good data. So we look forward to working with you on that.

And the work group that we’ve formed internally is trying to develop a set
of questions that we think we need the answers to, but I think by no means can
we answer all of those questions internally, and so we’ll certainly want to
interact with you on that.

MR. ROTHSTEIN: Thanks.

DR. COHN: Michael?

DR. FITZMAURICE: Earlier this morning, we had an update on the Departmental
budget situation and Congress is not in session so there’s – all we can do
is debate about it. But we’ve heard that in the President’s request for $100
million that – which would be $50 million extra for the Department, a
continuing $50 million that has already been spent – that it had been
reduced on the House side and then reduced on the Senate side, and yet they
still have to get together and compromise on their thoughts.

In talking with people in the private sector, I hear a lot of great support
and some kind of astonishment that such a large goal is – has so little
funds. Sometimes that happens because people don’t know exactly what the
framework is and what the goals are. So I guess my question is: Has there been
ample opportunity for the Department to inform various members of Congress and
the staff about the goals that you described in the strategic plan and to get
some feedback from them?

DR. BRAILER: Well, first, I – you know, I certainly support the
President’s budget request and, you know, I think we’ve tried to use the
opportunities that we have to educate Congress about why that additional $50
million is important and necessary, and I think we’ve made good headway in at
least explaining what it is that that’s for.

On the other hand, I think we have a long way to go, for two reasons. First,
you know, this agenda is one that is – you know, it’s not IT; it’s about
the fundamental function of the health care industry. And it is something that
requires a lot of education. And there are key decisions that need to be made.

And so I think there’s a set of sequence that we need to be able to spend
some of that $50 million to be able to be in the position of having more
thoughtful issues addressed downstream.

But secondly, there is – unlike many things that happen, I think, in
health care, this is one of the few

areas where the discussion isn’t about how do we spend the money to achieve
a social goal. It is: How do we achieve – how do we spend the money to
achieve a social goal and put ourselves in the position of saving money? And
that is so unusual that we’ve gotten caught in a lot of very thoughtful
discussion around pro formas and the analysis of what the expected value
realizations will be, and clearly that $50 million is small compared to what a
long investment might need to be to get those benefits. But it’s changed its
posture.

So I think we – it’s not been adequate. We’ve done a lot, and I’m
hoping that in the end that either the Senate itself or the conferees will come
together and will restore that funding and that we’ll certainly be talking
about where we go from here next year.

DR. COHN: I guess I just follow up with that comment ‘cause I think we
– I know you’ve been having many thoughtful conversations about cost
benefit and I think it’s a really critical question in all of this. There’s
obviously going to be many ways to put all of this together where it’ll be of
great value but it may cost more, and it’s going to be a difficult issue.

How do you see moving that agenda forward in a way that actually moves to
implementation? Do you have thoughts about that?

DR. BRAILER: Well, let me tell you what we’re doing, and then where it does
from here, I think, to be candid, is probably not – it’s obviously not my
decision so –

We are looking – we wanted to step back and ask ourselves: What is the
economic basis for health information technology? And we looked at it from two
perspectives, in the way the health services market works in the private
sector, in the way federal health spending works internally.

And we have since now developed 15 value propositions across those two large
divides, and there’s actually some interface that sits between them obviously
in terms of procurement and in services, in and out. And we’re now in the
process of validating those conceptually and then with whatever data that we
can find about whether those value propositions should be deemed offsets
against spending.

And some of them are pretty straightforward. For example, it’s no surprise
that in an uncoordinated environment with about $5 billion in spending health
information technology, which is what the federal government is spending, there
are silos, there are duplicative efforts that have existed within agencies or
within Cabinet departments that have not been aware of each other, that if they
were, they might have different enabling assumptions but turn out to be
achieving the same goal. And we’re now in the process directly through that
scorecard finding those efforts and understanding whether or not there’s an
opportunity to streamline those. Those are straightforward.

Then we have the more conceptual ones; I think they’re the more famous ones
in terms of reducing duplicative testing and, you know, that Medicare pays for,
or unnecessary hospitalizations. And those are further out in terms of how
immediately visible they are to the budget.

So we’re trying to get a sense of how confident we are that each of those
value realizations can offset investment and what kinds of investments can
bring those back.

So we’re doing that now. It’s a pretty – if you can imagine –
detailed and data intensive process, and I’ll be candid: It’s probably the most
time consuming thing that my office is involved in, but it’s the least visible
because there’s so much just in the – just in collecting data and trying
to find models and do this.

And so I – where that goes from here, I think, is a separate decision.
You know, we’re trying to create opportunities for decisions to be made about
this topic, but clearly those decisions have to be made by people in the
Executive Branch that feel like this issue has been adequately evaluated, has
been analyzed, and has broad enough consensus about what, where, how and when
that it’s worth kind of taking the next steps.

So that is our goal, is to do that evaluation and at some point, I think,
you’ll see hopefully more of it.

DR. STEINWACHS: Don Steinwachs. Talking about value, I guess – when
you think about the longitudinal record and the idea that eventually that can
extend from birth on, I guess one of the basic sort of value propositions is
how useful is that information as you think of it over time to the patient care
process and to the person, him or herself, and also we recognize that probably
that information will have errors in it and other things, and so that also by
retaining all that information over time, you’re in a sense retaining –
are there groups working on – I haven’t been in looking at research
recently in those kind of areas about – how do you begin to assess what is
the value of information as it moves along, and so

I as a patient or consumer can say, hey, that part of the record I don’t
want to share, but these parts are critical that I share with the next provider
because indeed that could influence important clinical decisions versus just
being information that probably is no longer used?

DR. BRAILER: Right. There is work done that’s underway on the
social-economic benefit. Blackford Middleton’s group at Harvard did this study
of interoperability and they looked at a variety of scenarios of how
interoperability could offset existing costs, and that’s where they found the
$87 billion in savings estimate that’s been discussed a lot.

To your question, though, that is only economic, which means it accrues to
those who have, you know, financial skin, which are largely payors. I don’t
know that there have been a lot of studies that look at this more from a
social, uneconomic perspective, you know; from a quality of life benefit
perspective there.

There clearly have been a variety of studies that look at health status
benefits in electronic health records or of data availability, but it’s not
been comprehensively looked at. In fact, one of the challenges we have across
the board is trying to make appropriate generalizations from little dots of
these studies that exist kind of out in space, trying to understand, you know,
how to connect lines.

So there’s some that are suggestive. There’s not a body of evidence from
which the kinds of policy decisions that normally get made was stressed upon.
And so it requires a lot of inference. And one of the reasons that I think the
AHRQ studies that are currently underway are so valuable is that they will
dramatically expand that body of evidence.

But I think that even putting those aside, there are the nonsocial,
economic and non-economic benefits. They’re more what I’ve called the
accounting benefits, which are how do we spend money on things today, either
NIT or not, that because of the way we manage them, we can get more mileage out
of them, the effectiveness of federal programs, the degree of leverage and
efficiency of spending?

And those things, I think, ironically have more visibility in terms of the
short-term savings than the real kind of savings that are out there because of
the certainty of the estimates.

So we’re working on it, but it’s – there’s a real evidence gap that
needs to be closed before I think this becomes just obvious to people.

DR. COHN: And Vickie, one last question, and then we’ll –

DR. MAYS: Anybody thinking about – in terms of the e-health record,
about the mental health utilization data, whether or not, for example – I
mean, some people actually go out of their way to pay out of pocket, insure
that those things are somewhat hidden, but at the same time, as we start to
think about health, we need to have people’s mental health data combine with
the health data.

And so there’s mental health data, there’s, you know, drug and substance
abuse data. Is there any discussion? I mean, these are going to be particularly
thorny issues.

DR. BRAILER: Sure, and it’s not even just a mental health or drug. Clearly,
HIV would fall under that category as would a variety of disabilities where
there’s some sense of potential stigma or a bias in treatment that’s been shown
in other studies that would make us have reason to think that there’s –
there could be special status for those.

There’s two issues with that. One is, how do we regard those? And the second
one is, should we regard those? And I mean the latter this way. There’s so much
that’s happened in states about a variety of those types of data. There’s one
separate discussion that is – if the continuation of that is a state
effort, what is the role of federal activity over and above what’s already been
done?

And secondly, if there is, how should they be regarded? And one of the
topics in our own privacy discussion group is exactly that, and I hope that
you’ll continue that in your work as well because this is an area where, you
know, there isn’t enough guidance, and there’s clearly a lot at stake.

So it’s a topic on the agenda. I don’t have an easy answer. In fact, I’m
afraid there isn’t an easy answer to those. But we at least need to think about
how to conceptualize the problem for a variety of different outputs, whether
they’re technology or policy, federal, state, et cetera.

DR. COHN: Okay. And I didn’t see Jeff’s hand up in the corner. Jeff, I’m
sorry; you’re in my far left. Do you have a – why don’t we let you have
the final question, and then we’ll take a break.

MR. BLAIR: Oh – well, and maybe you’ve – you answered this in
part. Let me phrase what I think I heard you say, is that in terms of the
initiatives being done by ARHQ and CMS and ASPE and all of the others that
relate to data exchange networks, relate to standards, relate to quality
improvement, I gather that we’re still a little bit way’s away before we can
fold in how these all fit together into our HIOs and NHIT. Is that correct?

DR. BRAILER: I think yes, although how far we are varies. I think we have
great urgency to understand what I would call the first pass implications of
those topics in the regional organizations. Said another way, we know that
whatever criteria we promulgate this spring for regional organizations will not
be the final work, but we want to be close enough to give good guidance and
start helping so many different efforts along so they can start having real
standing. I think that process will unfold in a relatively linear basis.

I think the broader questions, though, Jeff – I don’t know where the
receptor site is. I understand how they fit into RHIOs, but into broader IT
policy – and I think there’s a question here about what is the framework
going to be on – our strategic framework, what’s the policy framework
overall? And I think that’s one of the questions we’re dealing with a variety
of the organizations you mentioned; there are fiscal components, there are
privacy components, there are – with technology development components,
and I don’t mean that in a software sense but I mean it in the industry sense.
There are manpower components. There are organization and oversight components.

You know, one thing that I think the – happened quite well – I’m
living through this in the past six months – is the – you know, when
you create a canary in the cave with a national coordinator that cross-cuts a
topic, you discover many things that either would tell you that there was no
need for coordination because things are pretty optimized, or in fact that
there are many issues that need to be addressed and uncovered. And we found a
lack – this topic is big, it’s significant, it’s – it has huge
impact, but almost by definition it reveals a question about what is the whole
framework, how we view this, in a policy sense. And we’re struggling with that
right now, and you know, again, my view is our role is to create opportunities
and options, options for the elected leadership to decide how they want to move
forward and opportunities for society to move forward in their own way, because
this will never be solely a federal issue.

So, you know, I think we take that view, but I think it does beg this
broader question that ultimately we want to wrestle with in the name of health
care overall.

MR. BLAIR: Thank you.

DR. COHN: Okay. Well, David, again, congratulations on, I think, what’s been
some exceptional work over the last six months, and obviously we’re happy to
help you in any way we can, and I guess we’ll just be a continuation of that
ongoing effort.

I think with that we will take a 15-minute break.

(Break)

DR. COHN: Okay. Over the next hour and a couple of minutes – we’re
going to first start with an update on the NCHS Board of Scientific Counselors.
Excuse me; we’re starting. Thank you.

First is the update on the NCHS Board of Scientific Counselors. Then we
follow that with a discussion around the Population Report. So, Aldona, would
you like to take a lead?

Agenda Item: NCHS Board of Scientific Counselors
Update – Dr. Mays and Dr. Robbins

DR. ROBBINS: Yes, thanks, Simon. I’m going to give you a brief update of
what the Board of Scientific Counselors has been doing.

There was something that was supposed to be in the briefing books which is
just an updated roster of members, and that’s sort of coming around to you in
two different directions.

The Board began its second year with a meeting last September, and we
welcomed five new members. You can see them on the roster. They are the ones
whose terms will start – begun in May 1st of this year, 2004.

We are still in a something – we are still to some extent in an
organizational mode, but we did make some progress. There are two major topics
discussed. One was the CDC reorganization and the second was how the BSC could
best assist the NCHS going forward. And I’m slipping into acronyms which is, I
guess, inevitable when you’re working with government, but –

On the CDC reorganization, Dr. Sondik, who’s the NCHS Director, outlined the
CDC futures initiative for us and its new organizational chart, and if you
remember, because you all had a briefing last June, I believe, NCHS is going to
be put in – there are going to be new coordinating centers at CDC and NCHS
is slated to be put in the center along with Health Information in Service and
a National Center for Health Marketing, and also a National Center for Public
Health Informatics.

Now, the Board members had much of the same reaction that this Committee
did when it heard about the reorganization and specifically concern was
expressed about the placement of NCHS in the CDC reorganizational chart is
essentially – it appears the agency’s been moved down a level and whether
the extent to which NCHS will have – continue to have the control it does
now over its budget, its research agenda, data releases and personnel
decisions.

And after the discussion, the Board ended up taking its first formal
action, and that’s in Tab 3 of your book. It was a letter to Dr. Gerberding,
who’s the head of the CDC, and it extends – the letter extends her to a
meeting at our next meeting in January and invites her to come and discuss the
reorganization and its “effect” on NCHS’s ability to function in its
role as the nation’s principal health statistic agency.

Our second area that was – we spent a lot of discussion on is related
to what we’re going to be doing over the next year and specifically how we
could best help the National Center going forward.

Perhaps the most immediate thing will be – the CDC has a requirement
that its agencies conduct peer reviews of all their programs, and I take it
this is probably in many cases the first time this sort of thing’s ever been
done within the agency.

So for NCHS, it’s going to include their data surveys and their research
activities. So a subgroup of Board members who have had prior experience in
– with program reviews will be discussing coming up with – yes, the
thought was to come up with some kind of a structure of a generic review, and
we’ll talk about that more at the January meeting and then perhaps that is the
basis on which the various data surveys and agencies within NCHS, or programs
within NCHS, can go about their peer reviews.

Vickie Mays and I – Vickie’s your liaison to the Board – also
relayed the proposals that were made last August from the NCVHS Executive
Committee that concerns some joint activities over the next year or so between
this Committee and the Board, and Dr. Sondik welcomes the suggestion of these
joint meetings; they were going to talk it more in January.

And the last thing I have is that I was invited to a meeting of the
Populations Subcommittee about a week ago and I’m going to be taking back an
idea from them, and in particular I want to thank Vickie and Jean on the
Committee to use the NCVHS report health statistics, The Vision for a
21st Century, that I think many of you probably worked on, and
perhaps that that might be a good framework to use in terms of these –
developing these program reviews ‘cause it’s sort of – there is sort
of a natural link here.

And so I think I will stop here and in fact I’ll – you may want to say
more about that, Vickie.

DR. MAYS: In terms of (?) from this side, it – again, I think I agree
that things are still in the organizational mode and so most of the work was
actually done to get this letter out and a couple of other planning activities.

One push that we’ll probably do is to try and see if we’re going to have the
joint meetings to try and get them on a calendar, so – I think we talked
about when we thought it – you know, to do this, which – of course,
it is in 2005, but if – as we sit here, the extent to which the meeting
times are really getting taken up by some of the subcommittees that we might
want to try at the meeting and actually see if we can nail down a time.

The Subcommittee on Populations met on October 26 and one of the things that
we – you know, as we go through what some of our responsibilities are, one
of our responsibilities is actually to be monitoring the report from the
21st century – Health Statistics for the 21st
Century. We’ve tried here to make sure that we integrate, you know, population
health issues into our activities, so realizing that they’re about to do a
major evaluation and kind of look towards the future, we put it on our agenda
to figure out how to kind of hand that over to them so that that would be kind
of an ongoing part of their process.

So at the meeting we did talk about that, and I think this is probably a
very good time for the BSC to begin to think about how you want to really try
– since it was a joint project, how you want to think about population
health and integrating it as you go forward in terms of whatever the evaluation
is going to reveal about new directions. So that – so we kind of, you
know, wanted to talk about that, and Aldona was at the meeting, so I think that
is going to go forth.

DR. ROBBINS: Yes, I’m certainly going to make

that strong suggestion, anyway.

DR. MAYS: And then the other thing is that I’m

probably going to ask this group is when we do the research agenda to figure
out if there are specific things in the research agenda that are NCHS related
that we make sure that that’s something then that we can kind of pull out and
report on for the BSC meeting. So, you know, that’s about it in terms of, from
this side, in terms of the report. Marjorie, was there anything else that came
out –

MS. GREENBERG: No, there was nothing else.

DR. COHN: Okay. Any questions? Don?

DR. STEINWACHS: Don Steinwachs. In the program reviews, is the effort to
look at both current activities and current surveys, things they used to do,
things they might do? I was just wondering, is the review sort of where they’re
at right now or does it try and sort of project where they think they ought to
be or how to use the framework of the Health Statistics for the 21st
Century, where maybe they ought to be pointing?

DR. ROBBINS: Right. My sense is that is really a review of what is, largely
what is, but also then we want to work in – I think the Center would like
to work in some idea of where do we want to go from here, and I know that the
Board, I think, is – that’s where we’re most interested as well and so,
you know, things like what are the extent – are there questions or –
that may have outlived usefulness? Is there – are there questions that
should be added and what are we trying to accomplish with these surveys, too?
Where are they going to – how is the information – what’s the best
way for this information to be used and is it – who are the clientele and
so forth and so on?

DR. STEINWACHS: Yes.

DR. MAYS: I know there was a discussion of the scope of the review and I
think that’s what they will decide upon next, but, I mean, the scope was all
the way to do we have the right people that we need to be able to do some of
the things that we want to do? So it won’t be, as I understand it, just to look
at here’s what we’re doing but also in terms of directions they want to go in
and what it will take to do that.

So, I mean, that’s why I think this is like, you know, just a really good
time to make sure that if, you know, there are ways in which we can be helpful,
it’s a good time for us to be helpful.

DR. ROBBINS: Yes. And I also – both Vickie and I are also bringing back
a lot of what we’re hearing here about the information technology ‘cause
that’s what this Committee is spending so much time on. And, you know, that we
have surveys that go back 30, 40, 50 years, and all of this is happening very
quickly and in some sense it is to what extent do these surveys fit in with
this potential new world? And so, you know, that was discussed to some extent
as well, and how do we fit in with that?

DR. STEINWACHS: Could I have a second? As you look at the organization — I
know, once you meet with Dr. Gerberding, you’ll probably have a much clearer
idea of how these pieces fit together, but is there a sense now that the
Director Ed Sondik shared with you about how the different organizational units
in this new division that really covers all of sort of public health
information or data needs and surveillance, NCHS, and there was another one,
how they’re supposed to work together? Or is it just sort of an idea that
they’re all in the same place and the work together is – could either be a
plus or a minus?

DR. ROBBINS: Yes. I think it’s still – they – I didn’t have a
sense that they’d move much further than we were with – you know, when you
saw the charts last June —

DR. STEINWACHS: Okay – yes.

DR. ROBBINS: — and I’m not even sure whether they’re – Marjorie, I
don’t even know if they’ve staffed any of these other –

MS. GREENBERG: Yes, I think Jack wants to –

DR. ROBBINS: Jack – okay. Sorry.

MR. ANDERSON: Let me back up a little bit and answer your first question
also, which is to give a little – how this came about, the peer review.

Dr. Gerberding is a strong reviewer, peer review, and when she came to CDC,
as CDC has moved into more research over the recent years, a lot of the
research grants actually have not received peer review; they’ve what we call
‘em, objective review.

And so one of the first things she set in place was that grant programs had
to receive peer review. And once that was in place, she said, well, we also
need to do this for our intramural programs, and so it was actually to focus on
ongoing projects rather than future.

However, that said, Ed Sondik has a great desire to get a lot of input on
where we ought to be going, and so whether we do it through that peer review
process or we have to build something else, we’ll be looking very much for that
input on the future.

As far as the organization goes, the idea of the coordinating centers was
to bring together, closer, those centers that would seem to have more synergy
between them. So the idea is not that we would just pop together these three
centers and the coordinating centers and operate independently.

That said, most of our effort today has been in trying to actually set up
these organizations. Within our

coordinating center, the new Center for Marketing and the Center for
Informatics don’t exist, and so we need to bring those together and actually
set them up. And so even though that the idea is that there’d be a great deal
of synergy between the three, the other two aren’t there yet. We actually
informally stood up the Marketing Center October 1, but the Informatics Center
is not even informally set up yet.

So, clearly the idea is for synergy between the three centers, but quite
frankly, right now we’re moving to get the organization in place.

DR. COHN: Gene?

DR. LENGERICH: Thanks. I’m Gene Lengerich. I wanted to follow up on
something, Jim, that you had said earlier this morning, and that is related to
the coordination or standards set up for across health surveys – is that
what I remember you’re giving us? Was that part of what you had said? And I’m
just wondering if the Board of Scientific Counselors is going to be involved in
that action as well then.

MR. SCANLON: No, well, we’re – what I referred to this morning is
– remember, the Board of Scientific Counselors is advising NCHS on the
scientific and programmatic content of its programs. These reviews that I
mentioned earlier, those are really federal agency reviews of data needs. Some
of those surveys are conducted by other programs than NCHS though NCHS is
clearly a big part of that.

So, I certainly don’t have any problem with having someone from the Board,
you know, participate in the meetings as long as we could do it without getting
into trouble, but this is a broader – across the federal agencies and even
the Congressional Policy Research offices. It’s looking at the data that’s
needed, the adequacy of the data, in a few areas, and then what are the ways to
proceed, so it’s a little broader than your framework.

DR. COHN: Okay. Well, maybe we should move on to the Population Report? Is
that a –

Agenda Item: Populations Report – DR.
MAYS

DR. MAYS: Okay.

DR. COHN: Okay, please.

DR. MAYS: Okay. What I wanted to – oh, did you – tell them about
the new members?

DR. COHN: Yes.

DR. ROBBINS: Yes, I did. I just mentioned that they were ones whose term
starts last May.

DR. MAYS: Okay. Okay, there’s that handout at your place which is Proposed
Structure for Population Report.

As many of you know, we have this – you know, it seems like it’s never
ending but needs to come to a close. I feel like I’m – you know, this is a
– I’ve lived with it or something too long. But anyway, what we thought
would make sense is rather than plop down a large report for you next time with
no familiarity and no background, what we wanted to do is actually go through
the structure and just talk about it conceptually as to where we’re going and
also to kind of hear your input so that, you know, as we make decisions about,
you know, how much we’re expanding or not expanding something that we have some
idea of where you are relative to some of these issues.

So what I’m going to do is actually to walk through it. I have some
graphics; I’m trying to decide whether to use them or not because I don’t want
them to pull us off of what is the actual outline. The graphics actually are
the ones that have come from previous presentations on population health that I
just want to kind of bring us back to, ‘cause what we’re trying to do in
this report is to integrate the kinds of perspectives that we’ve been taking
that are very specific to population health survey.

Starting with the proposed structure, we’re not going to start with the
summary, but what I thought we would do is talk about it, and let me go through
a section and then we’ll stop and have discussion, answer questions, and kind
of do it in big sections as opposed to trying to walk through the whole thing.
Does that –

DR. COHN: Yes, I think that sounds fine. I think maybe you might want to
focus on the purpose of the report first just to remind us all and make sure
that we’re all grounded in that –

DR. MAYS: Well, there’s actually a place there –

DR. COHN: — which is Section 1.C., I think.

DR. MAYS: Yes, I was going to say there’s actually a place in here where I’m
going to do that, but let me kind of do it in terms of, well, what’s important,
and then why we’re doing the report, so we’re going to kind of start out with
the background. Oh, I think that went on the Internet. So, let me – I
guess I’ll read some of these.

We’re starting with articulating the main issue that’s addressed by the
report and what the importance is of these issues. And there are five points
that are there.

One is the federal efforts to eliminate health disparities and improve the
health of all Americans depend on the ability to really be able to target those
at high risk and with great need. And as you know, this actually dovetails
quite well with Healthy People 2010 which, you know, HHS conducts under the
leadership of the Surgeon General, who has major responsibility for Healthy
People.

Any of you who’ve been involved the Healthy People initiative know that one
of the things that’s necessary in Healthy People is to set targets, and those
targets are to achieve kind of better health for all Americans. And the process
of how specific those targets are I think really will depend on whether it’s,
you know, Healthy People 2000, Healthy People 2010 and, you know, as we begin
to look to Healthy People 2020, but, you know, it varies in terms of specific
targets for racial and ethnic minority groups.

The targets – say, for example, if we were to talk about the reduction
of diabetes or some disease, often what you will see is that in order to set
these targets, they will use NCHS data sets, they will AHQR’s data sets.
Typically, the targets are set using population-based data.

So what you run into is that sometimes targets are not available for
subpopulations because of the lack of that subpopulation data. So this is just
to kind of give you a sense of how important the need is for having data on
subpopulations.

Two is the target depends on adequate – this targeting depends on
adequate data about the health status and health needs of specific population
groups and subgroups.

Three, the groups at particular risk of ill health and poor health care are
those about whom health statistics and data are often most deficient.

Point three is an interesting one in that one of the worst groups in terms
of their health are actually African-Americans, and we have a significant
amount of data on their health, and we’ve had that for the longest. On the
other hand, we don’t have comprehensive data on American Indians, we don’t have
comprehensive data on Mungs or a lot of other subpopulations.

But what we know from Census 2000 is that these populations are, you know,
putting American Indians aside, the fact these other populations are increasing
in the U.S., and again it’s like part of the data need that we have is to be
able to target what the health needs for these groups are as well as for health
planning.

Four is improving the collection and use of national data on racial and
ethnic groups can make substantial contributions to efforts to ameliorate
health disparities and improve the health of all Americans.

So again, part of where this leads us to is it’s important to improve the
data collection of data on race and ethnicity so that we can provide the kind
of data that I was just talking about as well as to be able to determine the
health of these populations so that we can improve the health – their
health status and reduce morbidity and premature mortality.

Five talks about data gaps first identified in the Heckler Report. We’ll be
talking about, you know, acknowledging that. You know, that was where that was
done, what improvements have been made in the 1990s and the need for the
collection of race and ethnic data increasingly because it’s becoming a norm in
research in terms of that data.

Section B really talks about briefly explain the role of NCVHS and the
Subcommittee on Populations and why they’re concerned with these issues. The
question of why is always a valid question and for us, part of where this began
was in 1997, OMB guys came along and it was determined that race and ethnicity
would be collected differently than it had in the past.

There was a lot of preparation, for example, by the Census Bureau in terms
of getting geared up for Census 2000 to determine how to best collect this
data, what are the questions, what order should these questions be in – I
mean, there’s a lot of methodological work.

And in addition, what we knew is that OMB guys had recommended that the
federal data collection systems be ready and in place by January, 2003, to
collect this kind of new perspective on data on race and ethnicity.

So, the Committee at the time that it started really wanted to see where
things were in terms of whether agencies were ready. They wanted to get a sense
of whether agencies were going to be asking for variances in terms of how the
data was collected. We wanted to see what the development was like in terms of
mechanisms in place to bridge from the old data to the new data. And we also
wanted to know, in terms of these large scale surveys, what other variables
that they thought would be important to determine health disparities and some
insights from them in terms of what kind of data is needed to be able to put
into place planning for the reduction and elimination of disparities.

So that was really the background of why it is that we even started having
hearings and beginning to concern ourselves with these issues.

Next part is – you know, in terms of describe the purpose of the
report, and really, the purpose of the report is to begin to examine the nature
and limitations of national health data on racial and ethnic – it should
probably say “subpopulations” – as currently collected and used.

And two is to recommend strategies and actions for federal agencies to take
to remedy these problems and fill the gaps.

Next section really has more to do with what kind of sources did we have
for this report. ‘Course, we had the hearing back in 2002, and in this
hearing what we did was we brought in some of the major surveys from NCHS. For
example, we had Jackie Lucas, who’s on the Subcommittee on Populations, come in
and talk about the National Health

Interview Survey. We had NHANES in. We had the National Survey of Family
Growth. CDC, also for example, was there to talk about the Behavioral Risk
Factor Survey. We had the Agency for Health Care Quality Research talk about
MEPS. So a lot of the major surveys were there to answer some of these
questions.

And we asked them – we gave them a set of questions and we asked them,
for example, to tell us a little bit about how they were collecting race and
ethnicity. We asked them to identify contextual variables that might be
important in determining health disparity.

And of course, the variables that — you know, and we could predict this a
little bit from research – the variables that often were identified would
be to collect data on social/economic status, to collect data on education and
I think, depending upon what the particular survey was, we might have other
contextual variables that they thought were important.

But yet an issue that came up was that one of the things we have to
remember is that different ethnic groups have been in this country for
different amounts of time, and in terms of thinking about when we just look at
data by race, we don’t really then understand a lot about within group
differences. And so it becomes important for us to be able to have numbers that
give us a large enough population to be able to look at within group
differences.

So for example, we can think about looking at the health status of
African-Americans, and it may look horrible, but then again when you break it
out by education, when you break it out by where people live, when you break it
out by income, you might see a different picture.

And one of the recognitions of why we need to do this is that we’re at a
place with racial and ethnic minority groups that there is enough within group
diversity that it’s time that – I mean, it’s kind of the state of the art
in many other places – that it’s time that we try and generate data that
will allow us to do that.

The other place that we have information from that we’re putting into the
report, and this actually ate up a lot of time but I think was worth it, was
that – when I first began on the Committee, I was trying to understand the
history of this particular Subcommittee because it’s had a lot of names,
different names and iterations, and so I wanted to kind of see historically
what it had done particularly in the area of minority health statistics.

And so in looking through the archives, what I actually found was that
there was actually a lot of recommendations that have been made in the past but
they haven’t necessarily been enacted. And I think some of it is a lesson that
I kind of learned, which is some of them are mom and apple pie, some of them
are things that, you know, oh, of course we all want to do it, but budget! Some
of them are very general and very broad.

And so again I think that in the past what that has done is it’s led to the
recommendation – but I don’t know that it’s actually led to enough
specific strategies necessarily that were affordable and could be done, you
know, without major budget implications, so I’ve kind of learned some lessons
about where we need to try and go this time in terms of some of these
recommendations and also that it’s not just the issue of thinking about the
data itself but the mechanisms by which these things happen. Are there ways in
which there are particular stakeholders like health plans, are there ways that
there are stakeholders outside of HHS that we need to start thinking about in
terms of our recommendations so that we can broaden that recommendation to not
just dependent upon the Department to do it?

So, I think that, you know, as we push along in this process and discuss the
recommendations, my hope is that we will kind of hear from some of you places
in which we might need to change them in order to engage stakeholders that I
think are more appropriate than just depending upon the Department to do this.

DR. COHN: Vickie, are you done with Section 1 now or are you into Section 2
or where are you?

DR. MAYS: No. Let me just get the – I’m actually doing D. I just did
Part 1 in D and I’ve done 2 and then I’m about to move to 3.

DR. COHN: Yes, okay. Then we’ll have a conversation at that point, okay.

DR. MAYS: Yes, and then we have a conversation.

And three, in terms of reports and recommendations of other groups, one of
the reasons we’re also pushing forward with this is that there have been a lot
of reports that have emerged from the Institute of Medicine, and those reports
I think have been very good at making the business case as to why it is we need
to collect data on race and ethnicity. So I think that relieves the burden from
us to do that but at the same time what we have is that, as these reports are
being released, we don’t want to be the last to offer some comment.

So when we brought to you the commentary, for example, the last Institute of
Medicine report on the collection of data on race and ethnicity had just come
out, so what we did was we made sure to reference that, and that’s something
that we will also try and do in this report is to comment when we have
recommendations that are overlapping with other major agencies like the
Institute of Medicine so that it makes the business case.

E is just providing a roadmap to the report, which is just an outline, so I
don’t think we need to discuss that.

So what I’d like to do is actually to stop here and to take comment,
questions, concerns.

DR. COHN: Okay. I guess I just had one question and then – obviously I
don’t know the frequency of your hearings or exactly how you do it, obviously.
I’m at usually the other subcommittees.

DR. MAYS: Not quite as many as yours but –

DR. COHN: Right! Nobody does quite as many as ours [Laughter.]

DR. MAYS: But anyway – go on.

DR. COHN: But I guess I was wondering as the source material – I was
looking at the – it appears to me that this whole report is substantially
being generated out of the 2002 hearing, and I may be mistaken; this may be
misrepresenting that. But I was wondering – I was just sort of counting
that, well, gee, we’re almost at the end of 2004 now and wondering whether or
not some of that data needs to be refreshed or revalidated, given that we’ve
had two years of history since. Am I off on this or do you feel that that’s
still the pertinent up to date information that should guide this work?

DR. MAYS: The primary hearing that we had on looking just at the
population-based surveys was in 2002. We followed it up with several other
hearings. We’ve had a hearing on state data, we’ve had a hearing on – in
our breakout session we had information on vital statistics, and then we’ve had
specific population hearings after that, so that probably took us through 2003
and maybe early 2004.

So that’s helping to inform us. And, the population-based data sets, most of
them belong to NCHS, and so we’ve been working closely with Jackie Lucas and
Jennifer Madans to insure that if things that we thought were the case in 2002
have progressed along, that those things are getting removed or deleted or
enhanced as, you know, as appropriate.

DR. COHN: Okay. So I just misunderstood then. This is really 2003-2004
hearings as well –

DR. MAYS: Yes.

DR. COHN: — we’re all talking about.

MS. GREENBERG: Yes, I was just going to mention when I looked at the outline
that although I remember that – the hearing in I guess it was February
2002 –

DR. MAYS: Yes.

MS. GREENBERG: — it was extremely interesting, but I know you had several
other hearings in 2002 as well as in 2003, so I assume that all of those were
going to contribute to this?

DR. MAYS: They will inform just the specific area of population health
reports. Each of the other hearings were a little bit broader or not quite on
the same topic. Here where we’re talking specifically about the federal
government’s national population-based survey, the other hearings, as topics
come – came up in that hearing that were relevant to those surveys, it
will inform it, but the – we’re not trying to take and put all of the
hearings into this one report.

MS. GREENBERG: So this is primarily about the national surveys?

DR. MAYS: Yes.

DR. COHN: Okay. Bob Hungate, and then Mark.

MR. HUNGATE: I wanted to pick up on the comment on the concentration on both
federal efforts and national surveys and make the observation that surveys are
fairly lengthy between data collection and data dissemination, that the data’s
pretty old by the time you see it.

And I’m of the quality improvement school, believing that what we’re trying
to do is reduce disparities, and that if you’re going to do that, you got to
know what’s going on today.

And so it seems to me that it’s important to do the linkage between what’s
done federally and what’s done in the private sector. And I think that moves
you squarely into the administrative data. what happens in the collection of
race and ethnicity in the administrative data, which we haven’t fully grappled
with.

But yes, it’s an action item, and where most of the work will be done to
make it occur. Yes, so it seems to me that that’s something we ought to deal
with in order

to strengthen the impact of the report.

DR. MAYS: I agree there is –

MR. HUNGATE: Makes sense?

DR. MAYS: — oh, no, it makes a lot of sense.

One of the discussions that we did have was about how quickly data can move
from the collection to its dissemination and even then by some people’s
standard it’s old. But I think the question is, really: What is the outcome of
what you’re needing to look at? There’s some think that that – it’s fine
and then there’s more immediate things that you might want, you know, the data
to be, you

know, just straight out of the field.

One of the issues has been, particularly for NCHS, is resources to turn it
over quickly. They have actually decreased – and I’m sure Jackie would
comment on this – the time from collection to availability for the user to
be able to work on it. So I think that is continuing to decrease. And
technology is allowing us to decrease back.

But the other issue that you bring up is one that later in the report we
talk about and that is linkages. Linkages aren’t quite as easy as it seems.
There’s all kind of privacy, confidentiality. There’s issues with, you know,
data that’s collected at NCHS and, you know, how would one link that with, you
know, private? I mean, we can – there’s talk about linking – having
better linkages with other surveys and other data that’s both, you know, survey
and administrative.

DR. COHN: Okay. Mark?

MR. ROTHSTEIN: Vickie, as you know very well, any documents, studies,
surveys et cetera that call for increased use of race – let’s leave aside
ethnicity and other measures for a minute – are often responded to by the
assertion that because race is not a scientific concept, it’s only a social
concept, and mentioning or collecting more “race specific data” just
reifies this pseudo-science et cetera, et cetera.

And is it your plan anywhere in this document to address the issue of, okay,
even though it’s not a scientific distinction, it’s a very important social
distinction in our country, and why race specific data with ethnic specific
data et cetera is an important link in trying to eliminate disparities?

DR. MAYS: Yes. Let me just take the cup off the chair.[Laughter.]

PARTICIPANT: Oh, oh – you asked the wrong question!

MR. ROTHSTEIN: Can I withdraw it?

[Laughter.]

PARTICIPANT: Too late, Mark, too late!

DR. MAYS: He was a plant – no, kidding!

I actually have a slide – maybe I’ll try and use some of them, but this
is actually from a presentation that I did.

What is we will talk about is kind of the fact that how race and ethnicity
classification is used, but then we will also address this issue, that it’s a
social construct. In part what we advocate for is that we have variables beyond
just race. I mean, I think, the opponents will say that we’re reifying this
construct, it’s not a good construct, but at the same time one has to look at,
you know, this is the way in which, you know, we conduct our business in terms
of health. And we’re not trying to change all of HHS but I think we’re trying
to convince people that there are some valid approaches that can be used, and
what one has to remember is that race probably alone is not the only variable
you want to look at. So that’s why we want to do some of the contextual data.

MR. ROTHSTEIN: So would that be in the introduction?

DR. MAYS: Actually, we haven’t thought about putting that high up, but I
have a feeling that, you know, I can – as you talk, I can see the groups
that will actually, if it isn’t, you know, sooner in the report, will then just
attack the report and maybe not even read later where it’s buried in the
report, so I think that’s a good suggestion, so I think we will – yes,
that’s a very good suggestion. Okay.

Shall we go to the next section, so we could – that’s why I wasn’t
going to do all the slides ‘cause I thought we’ll never get through this
– let me try, but I like Mark’s question about that.

Okay, in Section 2, which is the overview of the uses of race and ethnic
data and relevant data systems that collect these, again, part of what we do in
starting this is to say all the different places that this data is used, but
– and that we need it, so it’s not just in terms of surveys but that, you
know, it’s in clinical research and health care utilization and outcomes
research, and where we have federal programs – Medicaid managed care, you
know, et cetera, these are little notes to us. If you see little question
marks, this is like internal working notes to us.

In Section A what we did was we made the case for the different uses of
data on race and ethnicity to demonstrate that it’s important to most of the
mandates that are within HHS. And the hope here in this section is to
understand that investing in better collection, that there really is a big
payoff, that we’re not talking about just changing a survey, we’re not just
talking about the importance of this to a couple of things, but that if this is
enacted throughout that the quality of data collection throughout HHS is
increased, that that just helps all of us.

For example, one of things later you’ll hear us talk about is that we’re
probably going to have a recommendation about the investing in developing
guidance or some kind of training so that individuals who actually collect this
data, like, for example, funeral directors – funeral directors fill out
your death certificate, and there’s very little training. I mean, having kind
of gone through some incidents within the last couple of years, I now
understand and I was there trying to correct him and help him and give him
little lessons about, you know, what to do in terms of this variable.

You know, I think that, you know, having some, you know, quick and easy
things for these individuals to think about as they fill out death certificates
– or even in a hospital as mothers fill out their birth certificate.

Again, these are places where a little bit of training will be helpful.

When the discussion in the quality work group was going on and people are
trying to figure out, well, who actually if — you know, the health plans are
collecting data, who’s going to be the person?

Well, if it’s a clerk at the desk and you come up and say, well, I’m pre-O,
she’s kind of look and she says, I have these categories; now what does that
mean? Or if somebody comes up and says, I’m a black Puerto Rican, so she’s
like, what does that mean, now? Did I do one group for you, or two groups for
you, what do I –

You know, so, in reality, you know, for the people who are on the front
lines, you know, we’re thinking about it in these great national surveys, but
what you have to think about is when people walk up and it – you know,

particularly in terms of in an environment like a place of care and they’re
there for something else and they – and you say, what are you? It’s like,
I’m African-American, or I’m – you know, it’s kind of quick and to the
point.

Now, it might be, and this is one of the other issues that we will address,
it might be quite fluid. You go in to see the provider and you’re worried
about, well, why are they asking me that? Is there some implications about my,
you know, genetic history? So then they start telling you about the great-great
grandmother who’s German and then the great-great grandfather who’s German and
Irish and Scottish and Jewish and – you know, it goes on and on and on
with all these different – and you stop and you say, okay – if you’re
at the front desk, how do – what do you do?

So, in reality there really – we really do need some guidance to help
people to be able to figure out how to capture the data so that it is good
quality data.

In Section 2, there’s also the relevant data systems, and again, as you’ll
see here, that national population-based surveys is the major focus of the
report, but we will, because we actually heard other things, talked a bit about
state surveys, vital statistics of the disease registries, some of the targeted
surveys, this is – the targeted surveys, we’ve brought this up before,
about the need for – you know, you can’t get all the data that you need
and have a large enough sample size to be able to do these multi-varied
analysis sometimes at the national level so that maybe you need for targeted
surveys.

We will discuss a big about administrative data and also talk about this
notion of the sources of contextual data.

Let’s see – here it’s like we – this comes from our own report,
Health Statistics for the 21st Century, and so we say, this is what
we want to do, this is what we think is important. But what I would challenge
you with is if you go look at a survey, it is very hard to be able to get this
kind of information.

But we know that many of these contextual variables are important, so what
we have to do is start finding out ways in which either the surveys are going
to be linked, ways in which, for example, that some of this may be added to
various surveys, so we’re a long ways from what we know both from theory as
well as research is the way to look at health, but again it’s like what the
purpose of this report is, is to get us to seek more and more about how it is
that we are able to get information on place and time and context and community
attributes.

Okay, let me stop here for this section then.

DR. COHN: I’m not seeing any comments.

DR. MAYS: Mom and apple pie!

[Laughter.]

DR. MAYS: So I conclude the – okay.

The third section, called Challenges in Collecting and Using Race and
Ethnicity Data, here’s where we really do get into what complicated issues are.
For example, multiple subgroups, privacy/confidentiality concerns.

In privacy and confidentiality concerns, that’s a very major one for us. And
the reason is, when we talk about national population-based surveys, often what
we end up with are small numbers. When you have small numbers then, the result
is that quite often we can’t get to that data unless we go to a data center.

And typically what this has boiled down to is that the data centers are
– there’s like maybe one or two, depending on which agency you’re talking
about; I mean, Census has several – but in terms of some of the national
population-based surveys, usually it means coming to Rockville. If not, you
also have to worry about whether or not you can even get subpopulation data on
– if you go onto public data, in the public data you’re just not going to
be able to get a breakdown so you might end up with a category that says Asian
as opposed to being able to have the Asian subpopulations because you just
don’t have the numbers.

And so, again, that’s one of those challenges. I mean, after collecting, the
challenge is the ability to be able to use it.

The other thing that we heard consistently is people’s belief that it was
against the law to collect data on race and ethnicity. And so, again, we’re
suggesting to the Department that it might be helpful, for example, to put out
some information that indeed it is appropriate to be able to collect that data.

Now, this goes to something that Mark brought up, and that’s the whole
notion about misunderstandings about the meaning of race, some of the debates
in the field about race, whether or not it is an actual entity that we should
be measuring – shall we even be talking about people’s race?

There’s going to be also, I think, some challenges that we have to think
about, and that is, when we talk about individuals who have one or more races
that they tend to select, when we talk about a person who is biracial or has
multiple racial categories, what does that mean in terms of health? Should
realize the science still isn’t there because we haven’t been trying to look at
– well, would we see a different pattern in terms of disease if we knew
that you were part Vietnamese and part Japanese? I mean, we haven’t in the past
had such fine breakdowns, so

now that we’ve had them, some people are asking the question, well, what
does it mean? You now know that I’m half-this and half-that; what – I
mean, is it that you treat me as this half in terms of the data that you know
about African-Americans? Are you treating me at the other half in terms of what
you know?

So there’s really some, you know, research issues and science questions
there that need to be addressed.

The cultural validity of the questionnaires, we’re just putting that on the
table to insure that as we move along that that’s thought about in terms of how
questions are asked and how – make sure that there’s relevance across the
populations.

It also comes up within the context of language. A lot of the surveys, data
is collected in Spanish, but not in a lot of other languages. Having sat
through now and having the privilege to analyze some of the data for the
California Health Interview Survey which was collected in seven languages, you
realize that you start to get – you start to be able to do some
methodological work by being able to have people whose language is other than
just Spanish in the surveys and to learn things about their health as well as
some of the methodological challenges to do it. So that’s coming up.

I kind of talked about there’s problems in data

collection; it’s almost like, okay, what’s – if one has more than one
race, what do you do in terms of, if you’re going to have a primary race
variable – I mean, I think that’s an issue.

Subgroup data, sample design, again, these were all things that I think,
you know, are typical survey issues but just very specific when you have such
small groups.

Problems in tabulating, analyzing, interpreting data on race and ethnicity,
again, that has to do with size, it has to do with some of the bridging issues
also. If you change how data has been collected, the bridging process from
going to the way it was collected to the new way that it’s collected. And I
really have to say NCHS has done a lot of work on this. It has done – if
you look in the field, I think it and NORC and a couple of other people have
really been doing work. But it’s done the lion’s share of trying to insure that
we’re going to be able to bridge data.

The race question in terms of Hispanics, this is still there. I’m, for
example, out in the field right now with a 4,000-person survey that’s a
follow-back on the California Health Interview Survey, and when you ask
Hispanics about their race, they’re like, I already told you – which we
asked them about their ethnicity. And then when you want them to think about
– well, no, but now, what’s your race, even sometimes when you give them
categories, there’s kind of this “Hamm” that – a little bit of a
struggle in terms of understanding that. So there still needs to be work on
that.

And then there’s the issue which if Peggy were here, she always makes sure
that we think about some of the state issues, and that is the burden for states
in collecting race and ethnicity data. I mean, ‘cause for them, there’s a
desire to collect the data in very fine detail because they have to service
their population, but at the same time there is a burden in terms of doing
that.

So should we stop at this section?

MR. SCANLON: Just one question, Vickie. Isn’t – there’s a similar
burden on anyone whose main job isn’t data collection – medical providers,
health plans, everyone else.

DR. MAYS: Oh, I know.

MR. SCANLON: Isn’t it more general? It’s a burden value kind of an equation.
It’s not just states.

DR. MAYS: Yes.

MR. SCANLON: So I just wonder if you –

DR. MAYS: No, I think that’s a good point, and it’s probably one where if
we’re going to broaden this out that it’s useful for us to maybe just make some
comment about the burden case and to give some examples, because I think that
that’s what comes up quite frequently as a potential barrier.

MR. SCANLON: Health plans and providers?

DR. MAYS: Yes, yes – no, I think that’s a really good point.

DR. COHN: Very good. Yes, Judy, and then Don.

MS. BEREK: Also, not just a burden issue but an accuracy issue. When I
worked in the Department of Social Services and we would try to add things to
the welfare enrollment form because we wanted to learn something, the front
line clerks would basically say, if it’s not needed to calculate a grant level,
don’t ever expect it to be correct. Yes, you can put 9,000 things on, but we’re
not going to pay attention to it.

And so I think that’s sort of an important thing to think about as we add
things on.

DR. MAYS: That’s also an issue where there is the fluidity, too, in the
sense of you may do well at collecting, but if the person is changing, you also
don’t – you know, you go back and you say, well, is – was it a
collection problem or was it a, you know, person identified problem, so I think
that’s something we’re going to see over time what’s going to happen in terms
of that, but I agree.

I think somebody else –

DR. COHN: Don?

DR. STEINWACHS: The distinction between Hispanic and Latino, are those
supposed to mean two different groups, and if you sort of remove the European
part of that, or is it supposedly interchangeable?

DR. MAYS: Now, I live in California and it’s like — I have pushed
“Hispanic” for the report, but it’s like in California I usually
don’t say “Hispanic.”

DR. COHN: Yes.

DR. STEINWACHS: And I thought maybe people were trying to drive a
distinction.

DR. COHN: Yes, and maybe this – we’re moving down to a depth where we
can –

DR. STEINWACHS: Gee, there, Simon, you won’t want me to learn after all
–

[Laughter.]

DR. COHN: I’m sorry, Don. The one thing that you didn’t know about –

DR. MAYS: No, “Hispanic” is a federal government term.

DR. COHN: Okay.

DR. STEINWACHS: It’s a standard in both parties.

DR. COHN: Okay.

DR. STEINWACHS: Both party standards.

DR. MAYS: Yes, it’s one of those standards that you have to deal with. Okay.
Moving on.

DR. COHN: Oh, but I think Bob had a question, quickly.

DR. MAYS: Oh, I’m sorry.

MR. HUNGATE: Just a comment regarding this immediate discussion, you know,
and the language back and forth. It seems to me that there needs to be almost a
rational statement that says that the accuracy that you can achieve is very
limited. You know, you can’t say that you’ll ever get this one right because
one’s perception is what you put down. The difference between self-reported and
observer-reported.

DR. COHN: Yes. Seems like a challenge.

MR. HUNGATE: And if you’re dealing with disparities, then it’s the
observer-reported that’s critical.

DR. MAYS: This is one of those federal government standard issues where
people in the field don’t like the term. I mean, we haven’t even gotten to just
– if it says “Latino/non-Latino” but just even the issue of, you
know, “Hispanic,” so it – believe me, it’s a challenge.

MR. HUNGATE: I’m sympathetic with the problem, but –

DR. MAYS: It’s a challenge, and I think it’s one of the ones that the people
in the field collecting the data – we’ve resolved what to do at a data
level. It says, you know, you just give the person – you fill it in and it
says “Hispanic” or “Latino.” But the person in the field,
when they’re given the complexity of it, I think that’s who has the greatest
difficulty and which gets to the issue of accuracy.

So this is why we actually need training. Even though the health plans and
others are interested in doing it, we really have to give some guidance
‘cause it’s not – it really isn’t as simple as it seems.

MR. HUNGATE: I hear you, but I just don’t think it’ll ever get there. That’s
what worries me is the –

DR. MAYS: Yes.

MR. HUNGATE: You know, the very complexity. This becomes baseline data,
which is useful on a national basis, and I’m not sure whether it works in
Lowell, Massachusetts.

DR. COHN: Okay. And I just wanted to do a time check for a moment here
– it’s 12:05. We’re happy to go into lunch as far as everybody wants to go
on all of this stuff but – and obviously we will go beyond 12:15 because
we needed to deal with this one, but I really want to make sure that everybody
focuses on – remember, this is a report; we’re critiquing what is intended
for this report, as opposed to necessarily discussing all of the issues.

DR. MAYS: I think it’s important they learn issues, too, but anyway, go on.

DR. COHN: No, I just wanted to say –

DR. MAYS: See, that’s why I didn’t do the slides, ‘cause I really
thought we’d get into a lot of

discussion about this.

DR. COHN: Yes, well, I’m sure we will.

DR. MAYS: If I need them –

DR. COHN: I just wanted to frame the issue, ‘cause I mean we’ll
obviously go through.

DR. MAYS: Yes.

DR. COHN: We needed to be comfortable with the recommendations and the rest
of it. I just wanted to frame the level of the conversation.

DR. MAYS: Okay.

DR. COHN: Okay.

DR. MAYS: Let’s do Section 4, ‘cause I can tell we need to get to 5 so
that you can get lunch, but in 4, the gaps in existing data on race and
ethnicity – and I think this – you know, I’ve kind of walked through.

First is subgroups, the ability to be able to know the data on
subpopulation.

Contextual factors – I showed you a broad list of those that, for
example, we would want to be able to collect. And that also applies in terms of
surveys and claims data.

Geocoding and linkages — you know, I said earlier that we would kind of
visit this issue, and that is, you know, place makes a difference. Place,
starting from all the way into maybe what state you’re in to what neighborhood
that you’re in, and, I mean, I can do, you know, long lectures about that but,
you know, for the sake of time, I’ll move on.

Socioeconomic position – we had – Dr. Virginia came here who did
the presentation and really kind of laid out for you kind of the implications
of – and so did Dr. Heurtin-Roberts – about what socioeconomic
position looks like when you begin to break data down by race and ethnicity and
include that and how important that is.

Health care expenditures – you know, it’s the other thing that in some
surveys we don’t have very much information on. In some surveys, that’s all
they do, but in others, they don’t.

And then we put on here, you know, needing to have data from targeted
surveys, needing to look at some of the state vital statistics, behavioral
risk, disease registries. You know, not all disease registries have, you know,
some of this data in terms of race, ethnicity, socioeconomic positions, so they
vary in the extent to which they collect that.

So those are the gaps. Simon?

DR. COHN: Okay. Don’t mean to stifle conversation here but –

DR. MAYS: No, that’s okay; stomachs are growling so I understand. We’ll
focus with, you know, the bang for the buck is, which is in recommendations.

Here what we’re going to try and do is really to kind of talk about what
some of the cross-cutting weaknesses are relative to infrastructure capacity,
and that’s where we think we might be helpful to really point out some of the
resources that are needed and some of the kinds of training and contributions
that would help.

First and foremost is to really talk about the guidance in terms of the OMB
structure for the collection of data on race and ethnicity. As I said before,
it really would be helpful to have training and to give people – you know,
you go on a website and find out, oh, here are ways to ask, here are ways to
customize it, I think would helpful.

Coordination – it shouldn’t surprise anybody that, you know, again, I
think that HHS has a committee where the surveys are coordinated – there
are studies in coordination among the surveys, but what we find is that each
survey may be asking very similar questions in different ways and there may be
reasons, but at the same time what we’re hoping is that by bringing this to
their attention that they can look and see whether or not there can be similar
questions across surveys allowing for better linkages.

Funding – a little talk about that.

Monitoring these issues and support for methodologic and other research
which hopefully there will be resources do that.

When you got the commentary – we’ve given you a set of overarching
recommendations and we’re kind of going back to that and kind of walking
through these. HHS should extend and intensify its current leadership role to
promote and support research on some of these issues. We want them to
accelerate implementation of multiple strategy approaches to developing data
and racial and ethnic minority populations.

In particular what we’re saying is that that probably involves also doing
some of these targeted surveys.

We will talk about specific strategies that we think will allow HHS to
engage some of these overarching recommendations, and that would be the
coordination in terms of race and ethnic data collection, the capacity of the
health infrastructure to – not only just to collect but to also
disseminate. I mean, this gets to what Bob was talking about.

And again, in consultation with OMB, to talk about supporting research and
provide guidance in key technical and methodological areas such as improvements
in the quality of race and ethnic data specifically and vital.

Statistics records – we will talk about what we’re sampling in national
surveys, where it’s feasible.

Targeted studies for data development on subpopulations when oversampling is
not feasible or cost effective.

Methodology to improve the quantity, quality and participation rates of
racial and ethnic subgroup data collection. NIH has, you know, has put a fair
amount into it trying to do that.

The use of Census denominators for rate calculation – I mean, we have
to kind of worry a bit about that, particularly in terms of like some of the
bridging.

Mechanisms to augment existing samples to target specific race or ethnic
groups nationally or in subnational areas with adequate privacy protection. And
the cultural validity of data collection tools. Linking and matching across
surveys and administrative data sources.

So that’s kind of where we want to go with these recommendations. Again,
they – I don’t want to say they’re mom and apple pie, but some of them
will actually have very specific – remember, we were the ones that came
with the 69, so we’ve pared down to 69, so when you now ask me in terms of like
pumping up again that these seem very vague, it’s like we will have, you know,
not 69 but we have specific strategies that we want to propose to be able to do
this, but these are the overarching categories that things will fall into.
Okay?

DR. COHN: Okay. Justine?

DR. CARR: Nice job.

DR. MAYS: Thank you.

DR. CARR: I just I’m thinking a little bit about what Bob just want which
is: What information do we have in terms of the achievability of the vision and
where we are today with 50 million people collecting data from patient –
individuals who may or may not want to give the information? That’s a huge,
formidable challenge.

And so – and we want to get to a place where all of those data
collectors are educated and all of the data givers are knowledgeable. And so I
wondered if we have information on what’s the magnitude of the gap. We talked a
little bit about the meaning last week and how far off the mark are we today
and are there – you know, as we go more complex, it becomes more
challenging, but if we had less complex that were more reliable, would we be
farther ahead?

DR. MAYS: We can’t really comment on where we are in terms of the private
sector. I mean – in terms of the hearing for where we are and what is the
magnitude that we’re off relative to some of the population-based surveys,
there’s been a lot of work methodologically to look at these issues, like maybe
you could go to any of the NCHS people and they could, you know, rattle off,
you know, what – 10 years of work.

But when we move outside of that, we began to talk about administrative data
sets and we began to talk about some of the private entities. That’s where we
don’t have that information yet and that’s where I do think, you know, we need
to move.

And I think that the question of once we moved out of a – I don’t want
to call it a more controlled environment, a population – when you’re doing
a population-based survey, you do have very specific things that you do and you
keep the person either in person or on the phone until you can kind of, you
know, get what you need, and there’s a lot of training for that data collector.

But when we move out into the everyday world, I agree with you. The
question will be: What’s the best, and what should we strive for? I mean,
‘cause – now, we can ask for data that would be perfect, but I don’t
think that the burden will allow the data to be perfect. But in terms of it
being consistent, as long as it’s not person- related but consistent and
reliable and accurate, those are probably the driving things and the question
to ask ourself is, in these environments, how do we achieve that?

And I think that’s where, you know, in terms of working with quality, those
are some of the questions that we need to understand is – as soon as we
say where it’s collected, then I think we have a better sense of what’s
achievable. But I think the question is still up in the air; in things like
health plans, it’s like: At what point are you going to ask this question, and
who is going to be the person asking that question?

DR. COHN: Other comments or questions? Well, let me ask a question here as I
look at all of your stuff ‘cause I’m hearing that there’s a lot of actual
specific recommendations underneath these bullets that obviously we can’t see
and I’m always reminded, first of all, that the value of any of these reports,
to my view, relate to the recommendations, and I’ve also observed that the
higher level the recommendation is, the less actionable and likelier they are
ever to occur ‘cause nobody can figure out what it means. Becomes a mom
and apple pie stuff. And obviously I don’t see any of these here, so I mean, at
the mom and apple pie level – of course I can agree with all of these and
at the specific level I may fight to my death about some of them, it’s hard to
know –

[Laughter.]

DR. MAYS: So we have to forget to just move ahead to the report – no,
I’m sorry!

DR. COHN: Well, so I guess I’m wondering – it seems to me if there’s a
– we need to have a way for us to begin to review the more specific
recommendations.

DR. MAYS: Right.

DR. COHN: Now, I guess I’ve sat looking at all this when I am sort of
transfixed with Item 8 at the very end here of 3 which is “Linking and
matching across surveys and administrative data sources,” and maybe you
could just explain to me very briefly what you mean by that.

DR. MAYS: I’m sorry – I didn’t hear. Where are you?

DR. COHN: It’s the very last – it’s the very last one.

DR. MAYS: Oh, “linking and matching across surveys and administrative
data sources?”

DR. COHN: Yes, and maybe you could just explain to me what you really mean
by that.

DR. MAYS: When you look at many of the population-based surveys, what you’ll
find is that one might have great data, for example, on family constellation,
and then another one may have other data that you need, but you don’t have a
way, for example, to be able to link them together. Some surveys you can, some
surveys you can’t.

So what we’re trying to recommend is that because what we’re interested in
is also getting contextual data that not all surveys are collecting that but
that the ability to be able to link, say, NHIS to some other data set, would be
very important.

Yes, like – well, Census data, we’re probably able to do some of that,
but they’re – you know, there’s – I mean, the list is very long of
–

DR. COHN: A lot of fair data.

DR. MAYS: — data sets that we have.

MR. SCANLON: And Medicare administrative claims data.

DR. MAYS: Oh, okay.

MR. SCANLON: There’s a lot of linkage with the administrative and the
population data.

DR. MAYS: Yes. So we want to be able to insure that we can do as much of
that as possible, but always what gets raised are things like it may be privacy
and confidentiality, it may be that you don’t have permission, you know, the
IRB didn’t indicate that that was going to be done, so – the more that we
put this on people’s radar, the more that as surveys – and I think it’s
about every – I may be wrong, but I think it’s about every either a
five-year or 10-year cycle for some surveys – that they, you know,
reexamine and in the reexamining and adding and changing, then they go back in
and they get, you know, a new IRB, the whole bit. But that’s the point at
which, for some people, some of these changes might take place.

DR. COHN: Jim?

MR. SCANLON: Two suggestions.

DR. MAYS: Sure.

MR. SCANLON: In the report itself, it would be good if the recommendations
were included in the executive summary —

DR. MAYS: Sure.

MR. SCANLON: — ‘cause that may be all –

DR. MAYS: — be overarching ones, okay.

MR. SCANLON: And secondly, if there will be more than 10 recommendations, or
12 or so, it would be good if the Committee could agree to some prioritization,
‘cause if there were 40 – I mean, it’s not that they’re – it’s
good to know what would be – if you could do the first three or four
–

DR. MAYS: Okay. ‘Cause you’re all very helpful. Okay. Anything else?
Yes?

MR. REYNOLDS: Yes. As I try to look at this from a standpoint of if somebody
handed it to me, what would I do with it, Section 3 almost says overwhelming
challenges rather than – I don’t see anything in the report talking about
that there have been success stories, people have been able to overcome these
things that are challenges. It seems to me they just kind of throw it up in the
air, and if you’re reading something this big, if I got to Section 3, by the
time I got to Number 6 in Section 3, I would – I might walk away.

DR. MAYS: Okay. That’s a good suggestion.

MR. REYNOLDS: Everything that you’re saying is valid, and I think it’s been
fascinating but the possibility of two – maybe even, I mean, if you turn
that around to where you collect this data appropriately, you have to deal with
these things. But the way it seems to say, it just seems to just kind of jump
in your face, like you’re never – you’ll never get through all this. So I
know there are issues and whether or not there are success stories which
obviously is –

DR. MAYS: Yes.

MR. REYNOLDS: — and certain surveys – you think are done well,
highlighting any of those to help somebody say, this has been – these have
been overcome in places –

DR. COHN: Yes.

MR. REYNOLDS: Might be an approach that puts a little, maybe a little more
–

DR. MAYS: Yes. Positive spin –

DR. COHN: Yes.

DR. MAYS: More hope, little more –

MR. REYNOLDS: Yes, yes. Hope would be good –put in some hope!

DR. COHN: Harry, I think you bring up a very good point.

MR. REYNOLDS: Open up.

DR. COHN: A I understand it, I think health plans are increasingly
collecting this data to help them –

DR. MAYS: Yes.

DR. COHN: — care for special populations or disadvantaged populations, and
I think that somehow reflecting on that, I mean, begins to give some reality to
maybe the recommendations. I mean, maybe there’s a linkage there that isn’t
immediately apparent. They might say, oh, geez, there’s best practices –
they need to be spread, as opposed to, yes, we can fix that idea if you do X, Y
and Z –

DR. MAYS: Yes.

DR. COHN: Anyway, just a thought.

MR. REYNOLDS: Can I have one follow-up on this?

DR. COHN: Yes, please.

MR. REYNOLDS: Because if I think about as a health plan, I think about our
enrollment form – wow! You got a lot – it’s got a lot of stuff here
we might not even be thinking about.

DR. COHN: Yes.

MR. REYNOLDS: And having those put down in a way that somebody could think
about them might help people go ahead and go this themselves without even
waiting for all this to happen.

DR. COHN: Exactly.

MR. REYNOLDS: It’s important enough, which it is.

DR. MAYS: I think that’s a key way to rethink that, because I don’t want it
to be here are the issues, and everybody goes, well, hey!

DR. STEINWACHS: Forget it!

DR. MAYS: You know, like: I neither have the time, the staff, the money
– you know.

[Laughter.]

DR. MAYS: Like: Next! You know, bring in the next memo and see what I have
today.

But that’s a – you know, and I think that we’re in danger of presenting
it in that way. And I know earlier we kind of did a round where we were making
sure that where progress had been made, particularly in terms of like NCHS,
that we were saying it. But there’s progress in the field on a lot of this. I
mean, it’s not perfect, but, you know, again it’s like, let’s give people
– this may be where we insure that we have the little references and some
of the point people to – even specifics about, you know, this primary
race. We can point them to NHIS’s data – this is what they do with this,
this is how they handle it.

So I really think that that’s probably one of these make-or-break things,
so I really appreciate it.

DR. COHN: Yes.

DR. MAYS: Justine? I’m going to call on you, since he’s talking.

DR. COHN: We’re going to do Mark and then Justine.

DR. MAYS: Okay, okay – Mark is here.

DR. COHN: I was going to say: Mark and Marjorie and Justine, okay. Mark?

MR. ROTHSTEIN: I just wanted to add on to what Harry said, and I think if
we’re going to set out some examples about how it was done right, it would be
very valuable if we took it the next step and said, as a result of doing it
this way, we were able to then get data that led to the following intervention.
So in other words, why all the effort to get these data were worth it. And I
think that’s valuable.

DR. MAYS: Well, I think we can go to the agencies and ask them if they can
give us some examples, ‘cause that’s hard. But it’s very good.

DR. COHN: Okay. Marjorie, and then Justine.

MS. GREENBERG: Okay. I was just thinking that in Tab 4 we have the
commentary that it did go forward from the Committee. And Jim, have we received
a response to that yet?

MR. SCANLON: I believe I’ve seen one that –

MS. GREENBERG: Oh, really? ‘Cause I –

MR. SCANLON: — coming through –

MS. GREENBERG: — didn’t hear. I don’t – okay, I was going to say I
asked staff but I didn’t –

MR. SCANLON: It may not have been signed yet.

MS. GREENBERG: I mean, there are some issues addressed in this commentary
that will also be addressed in the report, and I was just thinking that it
might be useful to get some feedback on the commentary specifically from the
Data Council, which is made up of all the agencies that, you know, at least in
the Department, would collect these data.

And I wondered even about the possibility of maybe having a presentation on
this to the Data Council –

MR. SCANLON: It’s a good idea.

MS. GREENBERG: — because this report won’t actually be due until March
– won’t be – come to the full Committee till March, right, although
there’s – obviously it’s going to be worked on. But if you get some
feedback on these recommendations from the Data Council, that might feed into
the report and the final recommendations.

DR. MAYS: We could do that soon so that we can kind of stay on schedule, so
maybe we can figure out a way.

MR. SCANLON: Yes, we could even think about the December meeting or
something like that. These are very familiar. I mean, many of the themes here
are things that the Data Council invented.

DR. MAYS: Yes. I think many of the times that these recommendations were for
previous iterations. They went right to Data Council. So maybe if Data Council
had them, they could even comment on what has happened to some of the
recommendations from – you know, from their perspective.

DR. COHN: And maybe there’ll be a letter by that time. Maybe they’ll help
them get a letter out.

[Laughter.]

DR. COHN: Justine?

DR. CARR: I wanted to underscore what Harry and Mark were saying, which is
alignment of incentives, that getting local buy-in to this is going to have
some intermediate entity that advocates that this is a good thing and that
there’s a clear connection with how it’s good.

DR. COHN: Okay.

DR. MAYS: I just want to say I appreciate the comments, because I think
they’re very good, very useful. I’m glad that we had the time to be able to do
this so that as we bring a report, rather than having you do 60 pages in it,
things that aren’t there or something like that, that at least we try and
capture this, and it’s – I do appreciate the comments.

DR. COHN: Sure. And this is as someone who has to read the 60 pages, I’d say
if you make it 45 pages, we’d even be happier.

[Laughter.]

DR. MAYS: Believe me, we’ll click. We’ll try and get it down as much as
possible ‘cause we have to reedit it, too.

DR. COHN: Okay. Well, listen, with that, why don’t we take a lunch break? We
have a presenter at 1:15, so it gives us around 50 minutes for the break and
we’ll see you back at 1:15.

DR. MAYS: Can I make one – one thing I neglected.

DR. COHN: Sure.

DR. MAYS: Just make one introduction. And that

is, because people probably don’t know. We have a new writer that’s working
with us and she actually came for

this. Ann Rogers, could you stand up? She’s working with us, so at the point
if you have comments also that you want to give, you know, feel free to share
them with her. And thank you, Ann, for – she helped us to reach the stage
of this outline; thank you very much for being here today.

Thank you, Simon.

[The Committee breaks for lunch at 12:27 p.m.]

A F T E R N O O N S E S S I O N

MR. COHN: Okay, this afternoon we’re going to be having a presentation on
the American Community Survey. Following that, we’ll be talking about a letter
that’s coming before the full Committee for discussion and action during this
meeting.

Before we move into the presentation, however, I just wanted to take a
minute, and I think Marjorie and I will be sort of doing this together. I think
as you all may have heard, Jackie Adler, who has been with us –

MS. GREENBERG: Since I think about 1986 maybe –

DR. COHN: — yes, has announced her retirement for January, which means this
will be her last meeting being here with us. We obviously want to thank her for
her great contribution and help over many years. I don’t know if I would have
made it through my tenure so far with the full Committee without her help
personally and I’m sure that may be true for many of you, and I know for John
also.

I was actually going to first of all suggest, and I would hopefully get a
concurrence from the full Committee, that we actually put together a letter,
and I hope you will allow us to – the executive committee to wordsmith it
and whatever, thanking her for many years of service and help to us, so do I
have a motion for that?

[MOTION is made and seconded.]

DR. COHN: Okay. All in favor?

[MOTION is carried.]

DR. COHN: Okay, so we have that.

[The Committee then discusses plans for a members’ dinner that evening at
Georgia Brown’s restaurant honoring Ms. Adler.]

DR. COHN: Okay, well, thank you. And again, Jackie, thank you for your help,
and we really appreciate you being with us all these years. You notice that
she’s not coming forward; she’s staying there in the far back, but hopefully
glowing.

Okay. Anyway, with that, why don’t we turn actually back to our business,
and obviously we want to thank Chuck Nelson for coming and joining us and
obviously briefing us on the American Community Survey. So thank you for
coming.

Agenda item: American Community Survey –
Presentation by MR. NELSON

MR. NELSON: Okay, thank you very much. And don’t be afraid to stop me as I
go along if people have questions. I mean, we can make this kind of informal,
if you want.

Okay, I’m here to talk about the American Community Survey, which is a major
new data gathering effort by the Census Bureau. I’ll give you some background
and talk about our plans for this survey.

First of all, what is the American Community Survey?

I guess to really talk about this, the ACS or the American Community Survey,
you have to go back to the Census. In 2000, you all probably got a Census form
in the mail, and some of you got the long form, the Census long form, which
– so in addition to answering questions about age, race, sex, you also
would have answered questions about education, occupation, income, disability,
work experience and so on. And that’s the way we’ve collected the sort of
– we’ve dealt with the nation’s small area data needs, survey data needs,
is with the Census long form.

Now, the Census Bureau has a new plan that will replace the Census long
form starting in 2010 and beyond and that’s this survey, the ACS. It’s a large,
continuous demographic survey, and by large I mean about three million
households a year, so that’s very large.

It will produce annual and multi-annual estimates of the characteristics of
the population and of housing. There are housing questions on this survey as
well. It won’t replace the Census, obviously; it won’t – there’ll still be
a 2010 census, but the 2010 census would only ask about age, race, sex and so
on, 10 year, and the ACS would produce information for every area that the
census produces information for, down to the tract or even block group level.
And this information will be updated every year, although for small places, it
will use multi-year averages.

And as I alluded to before, this is a key component of a reengineered 2010
census and beyond. The Census Bureau wants to get out of the – out of this
sort of phase of building up a large army of people to take the census every 10
years and then sort of dissolve it and then bring it up again 10 years later.
We want to have at least part of these operations be continuous. Makes it more
efficient and should make for some better data quality.

The problem with taking the short form and long form census at the same
time is the people that get the long form don’t – tend not to return their
questionnaires and that means it costs the Census Bureau a lot of money
‘cause we have to go to their houses.

And so – and it also affects the data because the long form data, like
income and poverty data, get affected by the fact that census is taken under
extreme time deadlines and so it makes it difficult to collect this kind of,
you know, long form data in the same period that we’re collecting short form
data, so it really has effects on both parts of the census.

So, what is – what would the – be the effect of the ACS on the
2010 census? Well, there would be a long form; there’d be a very nice, short
census form for everybody in the country. The fact that we would have
interviewers in every county of the country, in every area of the country,
means that we’ll have staff and have local knowledge about the entire country;
we won’t have to build up this staff of local knowledge every 10 years. We’ll
have the staff on the ground already. And there’ll be ongoing contacts with the
local governments through our field staff, so that will help us in 2010 take a
better

census, and that’s part of what we’re trying to do here, is to take as good
a census as possible and meet all the data needs, both long form and short form
needs. And we’ll have this experienced staff that will also help out on the
short form census in 2010. So really – so we’re looking at something that
really will affect both ends.

And so the ACS – we’ve been actually testing this survey since 1996 in
selected sites throughout the country. Since 2000, we’ve actually been
conducting this survey of a very large national sample, around 800,000
households across the country, and there are data at the Census Bureau website
for these estimates that I’ll tell you about later.

We’re planning for the full survey to start next year, in 2005, okay? And
this would include the entire household population of the U.S. as well as the
group quarter population and it would be planned for a 2005 start, assuming the
Census Bureau gets funding. I’ll talk about that a little bit later, too.

The first ACS data products would come out in the summer of – or the
full ACS data products would come out in the summer of 2006. So basically we’d
collect the data throughout the year and then the following summer we would
release the estimates for that year. That’s the way, so it’s a very quick
– one nice thing about the ACS is it’s a relatively short, simple survey
and we can put out data relatively quickly from this survey.

And then every year after that, we would put out updated information, so
there would be more information in 2007, ‘8, ‘9 and so on. So it’s a
very ambitious schedule.

And the way we plan to implement the survey is that in ’05 there will be a
national sample of around three million addresses, and these addresses will be
in every county, every American Indian and Alaska Native area, Hawaiian Home
Land, and also in Puerto Rico, so this will take place in all the U.S. as well
as Puerto Rico. Right now, there aren’t plans in sort of what they – these
called the outlying areas like Guam and the Marianna Islands.

PARTICIPANT: How about the Virgin Islands?

MR. NELSON: No, there aren’t plans now – I mean, people have been
talking about it but there are no definite plans for that, the Virgin Islands.

I volunteered already, yes.

And through this survey, we would produce profiles every year for every
community in the U.S. that has at least 65,000 people. So there would be annual
estimates for every county, every city, every area in the U.S. that has 65,000
or more people.

And then for areas of under 65,000, it would be either a three- or a
five-year average. We’d have to wait until we can accumulate enough data to put
out reliable data for those areas. And if the population is between 20,000 and
65,000, we can put out data based on a three-year average. If the population in
a place is under 20,000, then they’d have to wait for a five-year average.

Basically, after five years, you have 15 million households, which is
roughly equivalent to the census long form, about. It’s a little bit less, but
it’s pretty close. And that’s the idea.

Let’s see – so here’s a sort of a schedule of the way we will do it.
Now, yes, as I said before, we’ve been putting out annual data, and from our
national sample that we have out in the field right now, we’ve been producing
annual estimates for areas with a population of 250,000 or more and those
estimates – you can go to the Census Bureau website right now and find
data for, you know, the city of Cincinnati or Baltimore, or large county, for
every year since 2000.

In fact, in the last Presidential debates, one of them was in Cleveland,
and, you know, Kerry mentioned the fact that Cleveland had the highest poverty
rate in the U.S. That was based on ACS data – is – our rankings of
cities of 250,000 or more showed Cleveland with the highest poverty rating in
the country, and, you know, we — we’re sure that’s where he got that
information from.

MR. BLAIR: Is the census intended to gather data exclusively from American
citizens –

MR. NELSON: Oh, no.

MR. BLAIR: — or of residents?

MR. NELSON: No, it’s residents.

MR. BLAIR: It’s residents.

MR. NELSON: Yes. Anyone who lives at an address who gets this form will be
expected to answer those

questions, in fact whether they’re a citizen –

MR. BLAIR: If they’re a citizen of California and New Mexico, for example do
you make it available in Spanish?

MR. NELSON: Yes, we do. Yes, Spanish is the only – right now, Spanish
is the only other language that we have printed forms for. We do have – we
try to have field staff available for other – we try to handle other
languages through having field staff that, you know, in local areas that know
those languages.

There are thoughts about having more languages in the future and it’s been
discussed in – in Spanish, we certainly have those. In fact, all of our
surveys have Spanish versions right now, even like surveys like the current
population survey are in Spanish versions because there are so many people out
there. Yes.

DR. HARDING: Just for perspective, you said you did three million
residences. How many residences are there in the U.S.?

MR. NELSON: Number of households? I think it’s 115 million, maybe, 115.

DR. HARDING: About three percent?

MR. NELSON: Yes, it’s around three.

DR. HARDING: And then how many entities of counties and so forth are you
talking about?

MR. NELSON: There are about 3900 counties in the U.S., I believe.

DR. HARDING: Plus the – so there are about 4,000 entities that you
–

MR. NELSON: Well – right. I mean, if you look at beyond county, you’re
talking – I mean, we’re even talking about tracts. I mean, there’s, I
don’t know, 20 – there are a lot more than that; I mean, for areas in the
country that we’ve been putting out data for, eventually we’re talking about,
you know, 40,000. It’s a lot, it’s a lot. Yes, because every tract and every
– and actually we block it – blocked area in the country, which is
smaller than tract areas, so we’re talking about, you know, small places.

Now, the way this – so the way this works is, so, assuming we –
that we do start in 2005, we put out data starting in 2006, annual data, for
every place with 65,000 or places – if there was a place that had, let’s
say, 30,000 people in it; they’d have to wait until 2008 because we’d have to
wait until we conducted interviews in 2005, 2006 and 2007 to produce their
estimates.

And also, if they’d – if they want to compare their data to the state
data, we’d also put out state estimates and, you know, larger area estimates
for the same three-year average so people can make comparisons and, you know,
be on the same footing.

And then for really small – for places that – with populations of
under 20,000, they’d have to wait until 2010. And those estimates will be based
on an average of 2005, ‘6, ‘7, ‘8 and ‘9.

And then, every year after that, there would be a new set of data. There’ll
be a new set of both annual data and there’ll be a new set of three-year
averages and a new set of five-year averages, although there’d be a lot of
overlap and there’ll be two years of overlap between one-year of three-year
estimates and the next. And for five-year data, there’d be four years of
overlap between one year of five-year data and the next year of five-year data.
But that’s the way it works. It’s a continuous survey. We’re out throughout the
year and throughout the decade and every year there would be a new set of
estimates for every place in the country, really starting in 2010.

Does anybody have any questions about that?

So – now, there are budget issues here, obviously.

[Laughter.]

MR. NELSON: Now we – what we’ve said is that – what the Census
Bureau has tried to say, that because we save – because the census itself
is such an expensive operation and we save a lot of money on the census by
doing this, it’s actually a – this is actually a much more efficient way
of operating and actually saves money, although you have to invest in this
throughout the decade. So it’s a different way of spending money but it
actually saves the government money in the long run and produces better data.

And we’ve gotten a lot of support for that. The President’s budget in ’04,
in the budget that we received in ’04, gave us enough money to sort of start
the process of beginning this big survey in ’05. You have to do a lot of hiring
of interviewers and a lot of training, so we actually have got the money in ’04
to start the process.

In ’05, things are – I’m sure for a lot of people, things are up in the
air. Our House mark – the President’s budget had 165 million, which is
what we said we’d need to do this survey in ’05. The House mark was about 146
million, which we said was enough to get us started. You know, we’d have to cut
some corners, but we could do it.

The Senate mark was the same as ’04; we were straight-lined at ’04, about
60, 65 million. And then for that we said we just can’t – we just couldn’t
do the survey, we just could not do the full survey, and that – and the
Director of the Census Bureau has said that he’s, you know, willing to shut the
survey down and plan on a long form census if – unless the final, you
know, mark is close to the House mark.

So this is kind of a make-or-break year for the ACS. Basically what we’ve
said is that if you – if we put it off too long, there’s a risk of not
having – not being able to – not to do a good long form in 2010 or
not having ACS. I mean, at some point there, you know, the government has to
decide which way they want to go, and if you wait too long, then that’s going
to hurt the data no matter what we do, so this is really the kind of
make-or-break year for the ACS and we don’t know what, you know, we don’t know
when Commerce’s budget will be finalized. We’re part of State-Com, we’re part
of – I guess, State, Commerce and Justice are all together, and so we’re
kind of waiting to see what’s going to happen. So that’s still unknown.

Now, the ACS methodology is a lot different than the surveys that we work
on, than most of the surveys that most of us work on. It’s a mail out-mail out
survey, so it’s similar to the census.

Basically, we would – we take – so, let’s say, three million
households; we take one-twelfth of that number and mail them a survey and
– with the paper form, and they fill it out, and generally in the tests
that we’ve been doing, some, it like 56 percent or, you know, in the upper 50s,
close to 60 percent, mail out – mail back their form.

Now this is – and part of the reason is that when you get this in the
mail, it says – since we’re part of the census umbrella, it says that it’s
mandatory that you answer this form. It’s — you know, just like the census is
mandatory, this is a mandatory survey. And that helps – we’ve done
testing, and that helps, the fact that, you know, you have something –

DR. STEINWACHS: You want to use incentive!

MR. NELSON: Oh, yes. We would never actually send anybody to jail. You know,
it certainly helps.

Now, for those who don’t respond, we try to find a phone number for them and
using a laptop, a computer- assisted questionnaire, we try to conduct that
interview over the phone. And those – for those who – and that’s the
second month of the cycle.

And then the third month of the cycle, for those people who we still don’t
have interviews for either through the mail or over the phone, we then knock on
their door and conduct a personal interview, again using a laptop and using a
computer-assisted questionnaire. And actually, that third, you know, that third
part, is the real costly part, so we actually take a sample of nonrespondents.
It’s about a third, one out of three, and we go and knock on doors and get the
interviews that way. Yes?

DR. MAYS: When the census is done, there’s a lot of advertisement. I mean
– and as a matter for fact, in Census 2000, there were major campaigns
that were done particularly in terms of racial and ethnic minority communities
and there were community groups that were enlisted in order to get the level of
participation really up.

When you switch to this, I guess I’m wondering several things. In the
testing, have you tested whether or not it looks like without a campaign that
you’re going to get the same level of participation in terms of special
populations?

Two, is there a difference in who responds to the mandatory, because we’ve
actually been having – our census guy, Jamie Christy, has actually been
doing a lot of visiting and work and wanting all of us to let everybody know
about this, and I’ve actually heard people who were more concerned,
particularly people who are immigrants et cetera, about this being mandatory
and not understanding what might happen. You know their living situations, you
know, in which there not supposed to be in this garage in the back and all that
kind of stuff. So I’m wondering the extent to which those in the testing tried
to determine if there are differential response rates.

MR. NELSON: Well – yes, we’ve certainly done a major study because
Congress told us to do a major study on the effect of voluntary versus
mandatory and what – and our answer and our conclusion was that as a
voluntary survey, the costs would go with – will be tremendously more
expensive because more – because there wasn’t a definite – I don’t
have the numbers in front of me, but there was a definite difference in the
response rate. When you just sort of sent people the survey in the mail and
didn’t say it was mandatory, as you may expect, there was certainly a
difference.

We’ve – yes, and that’s the – and one of the differences between
the census and the ACS is, you know, the fact that there’s – we’ll make a
concerted effort so we can count people, you know, count everybody.

We certainly will, once – you know, once we know that this is really
going to happen, we will try to – try our best to publicize the survey and
tell people that their answers will be confidential and they won’t – they
can’t be used against them and so on. But you’re right; it won’t be

the same, certainly won’t be the same as the census. It will not be the same
kind of effort. One thing, there’s – I mean, the funding isn’t there to do
the same kind of – that was a tremendously expensive thing, you know,
campaign we ran in 2000, so – but to the extent we can, given our budget,
we will try to make it, you know – publicize the fact through our Public
Information Office, go out to local newspapers, you know, tell them that we’re
coming, and then – and also make part of that message the fact that, you
know, their answers are confidential and no harm will come to them for
answering this survey.

So, it’s – but it’s certainly an issue, and certainly we know there
– you know, there are all those areas in the country where it’s more
difficult to do this and we – you know, we try our best to – you
know.

DR. MAYS: Well, it kind of concerns me because I do think that you have a
proven method that increased participation and you are just about to walk away
from it at a point where you’ve got great response and in particular – but
it was a big budget item. I realize that. So now to walk away from it really
concerns me about – we already have been talking about the quality, the
data we’re getting. In a sense, it was so critical to almost everything else we
do.

MR. NELSON: Well, we’ll say that even –

DR. MAYS: That’s just a note of concern I think we should go forth.

MR. NELSON: Yes, I know. I mean, I will say that we – that this survey
– I work on surveys as well besides working on the ACS, and I worked on
surveys like the current population survey, so a voluntary survey – I
mean, the ACS coverage rate – I mean, very few people end up not getting
–I mean, the response rate is very good for this survey. So we don’t have to
– usually there’s this sort of nonresponse adjustment that you have to
make to reflect the fact that you missed – these you know, some people
that just refused to be in the survey. That adjustment for the ACS is very
small; it’s only a couple of percentage points.

So we’re capturing the people that we set out. I mean, we’re capturing
interviews for a very large percentage of the people that we mail these
questionnaires to, even though it’s not uniform for every city across the
country. Obviously the goal was to make it as uniform as possible for every
city and every area across the country because that’s what data quality depends
on.

Okay, I’ll talk a little bit about ACS content. ACS content is – it’s
similar to the long form, so people who have worked on that know – you
probably know what the content already is: education, migration status,
disability status, employment status – whether or not somebody worked or
is unemployed in the week before the survey, industry, occupation, weeks and
hours worked the previous 12 months, whether or not somebody received food
stamps, there’s a full set of income questions, you can get poverty statistics
from this survey, and you can get housing characteristics such as number of
rooms, utility costs, property values, mortgage costs.

A big difference between the census and the ACS is the fact that the census
– you got your census form on April 1st. You know, some people
answer on April 1st, but it’s – a lot of people do, and that’s
sort of – so your employment status and your disability status and your
educational attainment status is all kind of based on that date. The ACS is
taken throughout the year. One-twelfth of the survey is sampled every month.

So when you come out with a – you sort of come out with an estimate,
with an employment estimate, in the ACS, it’s really sort of an annual average;
it’s really a 12-month average, and it’s not a certain point in time.

When you come out with an income estimate from the ACS, it’s based on –
it’s actually a little complicated; it’s based on previous 12 months’ income
taken from survey that’s taken throughout the year, so the reference period
actually covers almost two years, so it’s actually a little complicated to show
in a table, whereas for the census, it’s this previous count of your income, so
any census, any number that you saw from the 2000 census on a poverty rate was
based on your income in 1999 where for the ACS, it’s actually kind of a rolling
reference period based on the previous 12 months of income for a survey that’s
taken 12 months throughout the year.

But the content itself is very similar because – and I’ll cover that,
cover the reasons why, in a little bit. To see what the content is on this
survey, we went through a process where every agency, including HHS, was asked
what their data needs were and basically, these needs that we received –
so we received responses from HHS, HUD, Education, Agriculture and so on –
those responses were used to justify every question on the ACS.

So because we’re a mandatory survey, we have to do – we have very
strict guidelines about what content, what the content is, and every question
has to be a – there has to be a demonstrated need, small area need, for
every question that’s on this survey. And that’s similar to the long form, I
mean, so basically the same agencies went through the same process in the long
form and a lot of them just gave us the same package for the ACS because it’s
kind of similar.

But that also tells – I come to groups and obviously there are lots of
things we’d like to ask on the survey, but for surveys, we can sometimes help
out agencies who need to know – who need information, but on the ACS it’s
a much higher bar. You have to really demonstrate, you need to show us
legislation or regulations that mandate the collection of this information, and
you also have to show how those needs can’t be met through other means, such as
national surveys.

So it’s a very high – it’s a very difficult thing to do. But surveys
– agencies do it, and certainly there are some discussions about some
further data needs with agencies, and as long as they can identify them and can
– and it kind of fits those rules, we do entertain requests for new
content, but it’s a very sort of a tough – it’s a very high bar, mostly.

And – yes, and part of the reason I’m here today is that there is the
– we’re trying to work with all the federal agencies, particularly those
who are already working on – using census long form data, to make this
transition between the long form and the ACS. And part of the reason I’m here
is that I’m one of the HHS representatives from the Census Bureau on this
program, so we’ve been talking to HHS for quite a while now, telling them that
what – about the survey and what’s going on and trying to help
particularly in the parts of HHS that are using long form data and try to help
them make this transition.

There’s a lot of information out there right now that kind of helps this. I
mean, people can look at the numbers right now since we’ve been collecting the
data on this big national sample since 2000. Certainly, agencies are able to
kind of take a look and see how their – how the estimates they use are
compared to the long form and so on, so it’s a –

The big issue, I think, is this sort of multi-year averages, the fact that
the census data is very simple to work with. You just have this big sample that
was out there, you know, about 17 million households and it was out there at
the same time period and it was relatively simple to deal with. You didn’t have
to worry about sort of multi-year averages. The ACS, the only way you can do
that, you can get analysis for small areas is the easiest multi-year averages,
and a lot of agencies just haven’t had a lot of experience with multi-year
average on the data, so I think that’s one of the things we’re trying to do,
and we certainly are doing our best and you can certainly talk to us and talk
to me and I can put you in touch with the right people if you have specific
questions about your data needs from this survey.

And with that, there’s a website, there’s an ACS link on the Census
Bureau’s website; it’s part of the handout, part of the slides. And you can
contact me if you have some, you know, further questions after this meeting, if
you have – I can certainly either help you out or point you in the right
direction if you want a hand on understanding this survey or the context.

And the other handout is sort of – this is something that we give to
communities and local officials, sort of highlighting the advantages of the
ACS, you know, the fact that there’ll be updated information every year and
it’s the sort – just a handy little guide. It’s sort of a general guide to
what the ACS will do, so I bring this along just so you can kind of see the way
it works, sort of trying to publicize the survey in local areas.

We try to have a big presence in local area conferences, mayors, and, you
know, local area governments, to tell them that this really, you know, really
had a tremendous potential to give them much more current information.

Another problem I really haven’t talked about, but it’s an obvious problem
with the census, is that if you’re a local area and you’re – and you want
to qualify for federal programs, say, that are based on poverty rates, you
know, if your poverty rate is just – if your area has changed a lot since
the 2000 census, for example, you know, these formulas, that’s what they’re
based on because that’s the only source of data, so you’re stuck for the next
10 years.

The ACS, you know, because it will be out there every year, it will –
you know, there’s a lag, obviously, but, you know, it will be current –
much more current than 10 years ago and so I think the people most excited
about this program are local area officials who now, you know, see the
advantages of having this sort of current data and

have them updated every year. We think it’s a tremendous new program and
we’re doing our best to make it happen. And I guess we’ll probably know in the
next couple of months whether or not this really is going to happen.

DR. COHN: Oh, Chuck, thank you very much. Actually, I have a question, and I
know Vickie and Russ have questions. I’m – I will apologize; I’m first of
all not a researcher, as I may express my – I’m going to show some of my
ignorance about the uses of census data. But it’s an ongoing question I’ve had,
and since you’re representing the census today, I thought I would ask it.

There’s an ongoing angst both in this Committee and it seemed to me in many
areas about the issue of race and ethnicity data and health care disparities,
and it is an ongoing conversation that we’ve had for a long time. And of course
a lot of it centers around to this, again, ongoing angst about the collection
of race and ethnicity data.

MR. NELSON: Right.

DR. COHN: And of course I look at obviously using administrative claims data
and things like this is probably not the optimal way to go out and try to
capture this; I’m very well aware of obviously the census data having a lot of
that. Is there an opportunity or some way for us to put that data together to
shed light about, you know, health care disparities? I mean, what’s going
– not being a researcher, I don’t know all the rules, but the sense I have
is that there’s an impenetrable barrier between health care delivery data and
census data and somehow there’s no way to take it —

MR. NELSON: Are you talking about actually matching the –

DR. COHN: Well, actually, I’m not even sure that I’m talking about matching.
I’m just asking the general question of, you know, are there solutions that
don’t involve lots of efforts of manual redundant data entry is I guess what
I’m asking. If I’m somehow just misrepresenting this, Vickie, I will apologize,
but am I asking such an obvious question that –

MR. NELSON: I mean, certainly the fact that the ACS data, that the ACS is
this large sample in which we are collecting pretty detailed information about
race and ethnicity as well as other information, I mean, it can certainly be
used to produce, you know, profiles of areas that can be combined with other
information about that area. It’s tough to do it correct between the two, but
certainly I think the ACS data could be used in concert with other information
to produce sort of a better look at something like health care disparities, but
you need some other information beyond the ACS in order to do that.

And how that information can be combined, I guess I – you have to know
more about the information you’re – you know, you’re – this other
information that you’re talking about, but –

DR. COHN: Maybe I’ll ask Vickie since we spent an hour on it this morning.
Am I asking the right question, Vickie?

DR. MAYS: You’re in the ballpark.

[Laughter.]

DR. MAYS: I feel really bad. Let me reframe it. That warms my heart. It
means all these population discussions are just like being spread all over, so
it warms my heart – [laughs].

DR. COHN: Okay. So I’m going to leave the ballpark and didn’t actually hit
the ball or –

DR. MAYS: Here’s the question.

DR. COHN: — something like that.

MR. NELSON: Yes, sure.

DR. MAYS: For example, there are lot of census data centers, and as I
understand it, for example, there are these incredible firewalls of even being
able to share in those data centers putting in NCHS data and putting in the
census data; they can’t even be in computers saying any – and there have
to be major firewalls.

Let’s move from that – I mean, also what we can do is take the census
data and use it as like contextual data. The question would be, you know, the
extent to which it can be linked to MEPS or NCHS or things like that I think is
what’s being raised, and the extent to which – and this is the question I
was going to ask, so we’re kind of there – the extent to which the census
has at various points kind of asked a different question than it usually
doesn’t ask, like, for instance, it did one about same sex partners, it did one
– I mean, it – they’re almost like each time it has something
different it does.

And the question is whether or not there’s the opportunity for this group
or for HHS to ask for a particular thing to put on. I mean, we may only get it
for like census – when’s the next one? 2010? Okay, I remember, Census
2010, that we may only be able to get it once, but that may be worth it to us
to do that. So the question is, what is the opportunity for something to go up
the line to make a request for that to occur? ‘Cause you said it had to be
mandated or regulated; we have to show it can’t be captured in other way.

MR. NELSON: Right.

DR. MAYS: Wait – we can make a case here, but – so I just want to
know what –

MR. NELSON: Basically, yes. I mean, I can give you some examples of what
agencies have used as – for their rationales.

DR. MAYS: That work?

MR. NELSON: That work.

DR. MAYS: Oh, okay.

MR. NELSON: I mean, certainly if you have a funding formula, right, that was
allocating money to small areas based on the characteristics of that small
area, like poverty, and that’s a reasonably asked income question, is –
because there are tons of formulas out there, tons of – including HHS
formulas – that where the allocation is, you know, based on local area
poverty data or based on the difference between a local area’s poverty rate and
the state poverty rate, or, you know, something where the only way – and
Census – and certainly the Census Bureau, you know, certainly is of the
opinion and, you know, that the only way you can collect poverty information is
by taking – is by asking a full set of income questions.

So those are the kinds of – you know, those are the kind – and
there’s, you know, legislation that this agency can point to that says, you
know, this money will be allocated, you know, based on this formula, and they
often say, you know, based on census results on poverty, of census poverty
results at the local area level.

So those are the kinds of – I mean, those are the kinds of things that
agencies use to justify a question, either an HHS regulation or legislation
that says, you know, something will happen, you know, based on our knowledge
of, you know, or based on this sort of small area data need.

And it has to be – and I can tell you, based on past experience, that
it has to also – has to be a data need and it has to be a local area data
need because obviously, you know, the first, you know, if a need can be met
through a national survey, you know, then it will – it won’t be a
candidate for a survey like the ACS.

So – yes, I think people – I know we’ve talked to some health
people in the past about the fact that the ACS will be a – is a – can
be a great screening survey for like a follow-up survey, and you can use –
if there were certain universes that you wanted to get to, the fact that we’re
asking, you know, that we’re taking this survey of three million households a
year, you know, that means no matter what characteristic you want, you could
find a pretty good sample in there and then you could use that to make –
to sort of design a very efficient sample for your own survey. So people have
talked about that possible use of the ACS.

The area about data linkage, I mean that’s obviously – I mean, I know
we have talked – I mean, we do have data sharing agreements with some
agencies. SSA I know we have data sharing agreements with, you know, and
certainly I could put you in touch with the people at the Census Bureau who
work on these data sharing agreements to see if there’s – I mean, it’s not
my area but I could certainly – I know the people who do this and I could
certainly put you in touch with the people who deal with these things and
– yes, I know there’s always a – you know, health data, there’s
always a sensitivity about health data and obviously you have – I mean, I
think a lot of health agencies have their – have guidelines that are, you
know, just as strict as the Census Bureau guidelines for sharing, if there’s
certainly, if it’s demonstrated that there’s a great benefit can come out of
it, you know, I think it’s certainly worthwhile to have discussions with the
Bureau.

We have a policy office that deals with these things and I could certainly
give you the names of those people and, you know, the contacts so that we
could, you know, just later if you want.

But, yes, there’s no reason, I mean, why it has to be that, you know, this
firewall you talk about has to always be that way. Well, maybe there is –
[laughs] – just don’t know. But certainly it’s worthwhile to talk about
it.

DR. MAYS: Can I ask my second question? Will you be – you know the
reports that come out from the census and there’s, you know, just consistently
the same series of reports, will you able to do that with the ACS?

MR. NELSON: Yes. Yes, you do.

DR. MAYS: Oh, okay.

MR. NELSON: Right. That is the plan. The plan is to produce, you know, every
– and obviously reports. I mean, I think we’re getting – you know,
we’ll probably – as time goes on, it may be less printed and more impact,

DR. MAYS: Right, right. But still it will be –

MR. NELSON: — you know, but the equivalent information will be there. For
every geographic area that right now you see long form data for, you would see
ACS data for eventually. Yes, that’s one of the Bureau’s commitments, and
there’s no doubt about that, yes.

DR. COHN: Okay, Russ?

MR. LOCALIO: I’m Russell Localio. Just a clarification of a previous
question. Correct me if I’m wrong. This survey is not going to be released with
three million individual observations; it will be released only in tables,
isn’t that correct?

MR. NELSON: No, there will be a public use file, a microdata file that
–

MR. LOCALIO: Right.

MR. NELSON: It would not be based on all three million households. It would
be based on a sample of those households.

MR. LOCALIO: Is it going to be masked or synthetic or anything like that?

MR. NELSON: There would be disclosure proofing applied to that microdata
file.

MR. LOCALIO: Thank you.

MR. NELSON: Sure.

DR. COHN: Okay. Bob?

MR. HUNGATE: Bob Hungate. Question really related to the disability status
content that’s in the form. Is the questionnaire itself on the website so that
one could see all the variations of that?

MR. NELSON: If you give me – I could send you the link. Well, I didn’t
get to it through here. If you have trouble –

MR. HUNGATE: Through there.

MR. NELSON: Yes, give me a call or send me an email. I can get to the link.

MR. HUNGATE: That starts to get into the health area pretty quickly and
there are some things that you could do against that that might be informative.

MR. NELSON: Right, yes. And there’s actually some discussion about making
modifications to the disability question. There was a little dissatisfaction
with the 2000 census disability questions and there’s a committee – it’s
actually chaired by people at NCHS. It’s sort of an interagency committee
that’s looking to make some modifications to that – to those questions.
These are six questions.

MR. LOCALIO: Okay. I wanted to see if this is possible.

MR. HUNGATE: This committee has had quite an interest in the reporting of
functional status and disability is a piece of that, so if that could mesh into
the ICF kind of coding system, then it might be a big stride in some of the
usability of data in a broader way.

MR. NELSON: Okay, yes.

DR. COHN: Yes.

MR. NELSON: Yes, do you know Jennifer Madans? She’s involved in that effort.

DR. MAYS: Yes.

MR. NELSON: She’s at NCHS. I know she’s

one of the people involved in that, yes.

MS. GREENBERG: I have a comment on —

DR. COHN: Well, did you have a comment on this one specifically, Marjorie?

MS. GREENBERG: Yes. I might —

DR. COHN: Just ask the question.

MS. GREENBERG: I don’t know myself, the exact question that, or a set of
questions. It’s several questions that are being asked.

MR. NELSON: Yes, it’s a three-part question or a four-part question.

MS. GREENBERG: All right. Actually, Jennifer is also leading an
international effort to develop a small set of questions on disability that can
be asked — you probably know this just from other people’s information as well
– that can be asked in any census throughout the world, and then that
effort also includes people who are working on an extended set that could be
asked in household surveys.

And the international classification of functioning disability and health is
the framework for this international work, so interesting to see how this
–

MR. NELSON: And certainly the same people are involved with both of these
efforts, so hopefully they’re thinking about 30, yes.

MS. GREENBERG: Maybe – I mean, we could – if you’d like us, we
could have Jennifer come and talk about both of those activities either to the
full Committee or to the quality work group and subcommittee on populations.

DR. COHN: Okay – Jim?

MR. SCANLON: Yes – very informative. As I understand it, as I look at
this, all of the information is going to be collected from individual
respondents who are selected based upon their household?

MR. NELSON: Well, it’s a household survey, so we – when we choose a
– an address, we would then ask that – the people living there would
fill out the questions for everybody living at that address.

MR. SCANLON: Okay.

MR. NELSON: So it’s –

MR. SCANLON: In your brochure, it sounds almost like you’ll also be
collecting information about the individual community as well, the amenities or
lack of amenities that may exist in the community.

You talked about roads and schools and after-

school programs. So are you also collecting data about the communities?

MR. NELSON: No, I think that they were trying to say that this survey will
help these communities plan for roads and, you know, but, no, we’re not
collecting anything – we’re only collecting the information about the
people living in these communities, and then we hope, hopefully, that those
communities can put that data to use, you know, along with the other
information that they have about the amenities. But maybe it’s not so clear
– [laughs]. Obviously it’s not the end.

MR. SCANLON: Well, I just got a little bit –

MR. NELSON: It’s not that clear from the brochure, but that –

MR. SCANLON: — confused because as I thought it was the – people are
the respondents, and they’re talking about their households, not about –

MR. NELSON: That is true.

MR. SCANLON: But you’re also not collecting information about the community.

MR. NELSON: No.

MR. SCANLON: And I began to think about health care access, availability to
every –

MR. NELSON: Right.

MR. SCANLON: — but you wouldn’t have any of that information.

MR. NELSON: No, it was strictly this – it’s information about the
people living in this community, yes, right.

DR. VIGILANTE: Given the financial condition of many of the states, it may
not be the best time to ask this question, but since – it seems like a lot
of the benefit, or some of the benefit, would accrue to states and localities.

MR. NELSON: Yes.

DR. VIGILANTE: Has there been any thought to sharing the funding for this
ongoing process at that level rather than risk using the whole –

MR. NELSON: Not to my knowledge, not to my knowledge. I think this has been
– you know, since this is a sort of a federal government responsibility to
provide this information, it was – you know, it was – we’ve
considered it a part of – and we’re trying to get the funding to our
agency to do that.

I think we’ve discussed the possibility of – you know, if states wanted
to, you know, do something additional with this survey that they might be able
to, you know, add, you know, add on, or you know, follow-on surveys or things
like that, but we haven’t – you know, beyond just general talk, we haven’t
really, you know, beyond that, but that –

I would think that as time goes on, if this became a reality, I can see
some of that happening. I can see states saying, hey, this is a great –
because within any state, there’s a lot of people we’ll be interviewing, this
is a great – and could be a fairly cheap to find out some additional
information and go beyond the sort of basics that we collect.

So – but I think you have to wait until the survey’s up and running and
to show you can put out the information before states will trust us to do those
kinds of things.

DR. VIGILANTE: ‘Cause one could argue that, you know, certainly a
10-year snapshot might actually be adequate at a purely national level. But at
the local level it certainly isn’t.

MR. NELSON: Exactly, exactly.

DR. VIGILANTE: And the incentive is actually getting it in a more local
level or more local rather than national so that one of the incentives would be
with the outcomes.

DR. COHN: Vickie, I see your hand raised. Did you want to bring up the last
question, and then we’ll wrap up?

DR. MAYS: Sure. Can you tell me if this change in any way impacts some of
the NCHS work? I mean, there’s two issues that I’m familiar with, which is,
one, that they get their sampling frame from you for other studies, and two,
they also get their workers from you for – several people, I think,
actually use the census to actually collect data for them.

So going to this, does it in any way impact NCHS?

MR. NELSON: Well, actually, yes. We think that our ability to choose
sampling frames for surveys will be better due to this survey because part of
this – I mean, part of taking a successful, large scale survey like this
is having good addresses, and not having good addresses, you know, once a
decade but having good addresses throughout the year. And so we think that
– and part of the funding for this includes doing a better job throughout
the decade of making sure that our address list really is the best it can be.

So we think that will help our – you know, help us for our clients,
like NCHS or –

DR. MAYS: And in the work force, ‘cause you – it sounds like you
have less people there?

MR. NELSON: Well, I think the work force – I mean, it actually gives us
the ability to do a better job in local areas because we’ll have a lot more
people throughout the decade. And they have people who work on these different
surveys and, you know, people – so it probably helps, I think, helps to
the extent that these people, you know, that they’ll be trained interviewers
out there that will be available to work on other surveys, right? I mean,
probably a lot of interviewers work across surveys. So I think it actually
probably helps if – but that’s what you’re talking about, is that the
trained – you know, the trained interviewers actually gives us a bigger
pool in trained interviewers.

You know, one – and that’s a big advantage I really didn’t talk about
before, but one really big advantage of this survey is that we’re out there
once a decade and you just hire a bunch of people, you know, to take the
survey. Yes, they’re pretty good on collecting the data on how many people live
at a house, ‘cause that’s the basics; that’s what the census really is all
about. But the not – but it’s not as good for subject areas like, you
know, what is income and what is an income, what’s work and what isn’t work,
what is looking for work and – you know, these – it’s hard, this
temporary work force that you hire once every 10 years, to get, you know, to
get them to sort of understand those kind of concepts. And this survey, I mean,
it really is like a survey; these people, once they start working on it, work
on it every month and they really understand the concepts. So we can tell that
the data quality on the ACS is generally better than the census because you
have this trained work force that works on these surveys every month of the
year as opposed to a temporary work force, so it really does a better job.

DR. COHN: Well, Chuck, I mean, thank you very much. It’s been an actually
very informative discussion. Vickie, I guess I would ask, in terms of our
discussions for the first hour, you know, the hour before lunch break, is there
anything here that reflects or is a crucial opportunity in terms of your
report? I mean, is there a section here that might in some way reference
opportunities related to census data or is that a completely different report,
another day, another conversation? Somebody’s laughing here. Don, was that you?

DR. MAYS: I think that maybe another day.

DR. COHN: Okay.

[Laughter.]

DR. MAYS: I hope to think about it more, though.

DR. COHN: Okay.

DR. MAYS: Thank you.

DR. COHN: Well, thank you. And Chuck, thank you very much.

Agenda item: Quality Workgroup Letter on Diagnosis
Indicator, ACTION 11/5 – MR. HUNGATE

DR. COHN: Tthe last item of the day is the letter from the Quality
Workgroup, and Bob, I’ll let you present that. I think the question is
obviously we’ll look through it and talk about it. One of the questions is
whether with some wordsmithing whether or not it needs to be – whether
it’s good enough to go or whether it needs to be brought back tomorrow, but
we’ll talk about that as we go forward. Bob?

MR. HUNGATE: I’ll ask you all to turn to Page 5, or Tab 5, I guess is the
more appropriate way of describing it, where the letter is presented.

The letter is an outgrowth, just for a brief context, of three hearings
which were conducted primarily by Quality Workgroup but jointly with Standards
& Security. Harry Reynolds and Simon Cohn and Steve Steindel joined us for
the first two hearings and Steve stayed for the last of the hearings, which
covered registers, providers and designated standards maintenance
organizations.

The other parts of the context were that we started out working on eight
recommendations and one of those surfaces in this letter. The deliberations on
the rest of the content will be ongoing with emphasis on race and ethnicity and
claims data as one focus, functional status as the second focus, and a third
element that came up in the discussion as a surprise, I think, of almost
everyone was the linkage of the UB-04 with the ICD 10, and I’d like to talk
about that a little bit after going through the letter, and maybe we can –
if things go smoothly, we’ll have time for that; if not, we’ll defer that till
tomorrow.

In the letter itself – I’ll read the letter paragraph by paragraph
– there are a few places where I think a word change would improve it, and
I’ll include that as we go. Is that acceptable?

[Reading] “In the May 2000” – well, “Dear Secretary
Thompson:

“In the May 2004 report titled Measuring Health Care Quality:
Obstacles and Opportunities, the National Committee on Vital and Health
Statistics (NCVHS) identified 23 opportunities described as candidate
recommendation. The Quality Workgroup together with the Standards &
Security Subcommittee of NCVH has conducted hearings centered on the first
eight of these candidate recommendations. Candidate recommendation 3 –
‘Facilitate the reporting of a diagnosis modifier to flag diagnoses that
were present on admission’” – and here’s the first change –
“’in secondary diagnosis fields’” – “or” instead of
“in” – “’all inpatient claims transactions’ was universally
sought by the purchasers of care, endorsed by the providers of care and is
poised for implementation by the Designated Standards Maintenance Organization.
The NCVHS recommends that the next version of the Uniform Bill for Hospitals
(UP-04) and the ANSI ASC X12N 837I HIPAA Implementation Guide be revised to
facilitate reporting of a diagnosis indicator to flag diagnoses that were
present on admission in secondary diagnosis fields or all inpatient claims
transactions.”

MR. BLAIR: Bob, I think instead of being “ANSI X12,” that should
be “ASC X12,” which stands for Accrediting Standards Committee.”

MR. HUNGATE: So the “ANSI” drops out?

MR. BLAIR: In fact, the “ANSI” drops out.

DR. COHN: Well, I think we can check that one. Jeff, I thought it was
“ANSI ASC.”

MR. BLAIR: No, it is – matter of fact, ANSI has tried repeatedly to get
that corrected, ‘cause ANSI is the American National Standards Institute.
That’s the accrediting organization. But the actual name is ASC X12N, which
stands for Accredited Standards Committee, X12N.

DR. COHN: Well, we’ll refer that to the executive committee to clarify that
and look at the HIPAA implementation guides, what –

MR. HUNGATE: I will ask for Standards to resolve this.

DR. COHN: Yes, exactly.

MR. HUNGATE: I’ll leave this to the Quality Workgroup to – and we’ll
make that be consistent and correct. Any other comments here?

MR. HOUSTON: — comment. I’m not sure whether it’s obvious or not, but in
the first sentence, “opportunities” – the end of it says
“opportunities described as candidate recommendations.” Should –
is there a “for what” that needs to be an end to that sentence,
“recommendations.” I mean –

DR. MAYS: I can’t hear you.

MR. HOUSTON: Sorry. I’m just – at the end of the first sentence.

MR. HUNGATE: That could be “to correct gaps in information.”

MR. HOUSTON: The end of the sentence says “opportunities described as
candidate recommendations.” I’m just questioning whether there should a
–

MR. HUNGATE: Purpose of the recommendations.

MR. HOUSTON: — purpose of the recommendations. So, for what, for measuring
health care quality or what’s the – what are the recommendations related
to, I guess is my point. Maybe that needs to be –

MR. HUNGATE: They relate to the correction of

gaps in data, obstacles and gaps.

MR. HOUSTON: So that’s what the 23 opportunities are related to?

MR. HUNGATE: That was the subject, right.

MR. HOUSTON: I think we should say at the end of the sentence that these are
recommendations related to –

MS. GREENBERG: This was the first sentence?

MR. HOUSTON: Yes.

DR. COHN: We’re talking about the first sentence, yes, at this point, so —

MR. HOUSTON: Yes, that’s correct. That’s all I’m saying.

DR. COHN: So, the recommendations for – and I think the Workgroup needs
to come up with that, those words.

MR. HOUSTON: Right. That’s all – I think it just ends.

MR. HUNGATE: All right, we will find the words to put in there.

MR. HOUSTON: Okay.

MR. HUNGATE: Right. Anything else in this

paragraph? The next paragraph:

“The secondary diagnosis indicator which flags a condition present on
admission can help differentiate pre-existing conditions from those that may
have been the

result of medical errors or ineffective processes of care. The ability to
distinguish between conditions present on admission from the development of
conditions during a process of care can enable both case mix/severity of
illness adjustment at admission and quality improvement opportunities in care
processes. There is experience collecting this information in New York and
California hospital discharge systems, where it has proven valuable for risk
adjustment” – “for both risk adjustment and outcomes
assessment.”

Justine?

DR. CARR: Yes. I want to make two suggestions. In sentence Number 1,
“The secondary diagnosis indicator which flags a condition present on
admission” – you don’t have to say “can help to.” It does
differentiate. So, say “differentiates pre-existing conditions from those
that may have been the result of medical errors.”

And I would leave out the word “or ineffective processes” because
I think that there are other things in the process of care that can happen to a
patient and I just think that it’s too judgmental to just say that anything
else that occurred was ineffective care, or a complication.

So we take out “ineffective.”

MR. HUNGATE: Does the change – I –

DR. CARR: Take out “or ineffective.” I would say may – can
distinguish between medical error – I mean, I don’t know. This sounds kind
of punitive to me. It sounds like complications of care, iatrogenic or
otherwise, is really what we’re trying to say. I don’t know. Other folks, is it
not –

DR. COHN: Very good sense.

MS. GREENBERG: You’d keep in “processes of care” but just take out
“ineffective?”

DR. CARR: I mean, I think I would be – it differentiates pre-existing
conditions from those that may be the result of, you know, the best processes
of care. I guess, what’s the intent of the sentence? Is – there are many
reasons to want to differentiate pre-existing from those acquired during the
admission. This has a negative tone to it. If the idea is that because medical
errors is a high priority item and this will give it more support by saying
that will take out medical errors – that would be one reason. But I just
– you know, but this suggests that anything else that happens during the
admission was bad care, and that’s not –

DR. COHN: Justine, I’m just – I sort of agree with what you’re saying;
I’m just not sure what the

modification you’re suggesting –

MR. HUNGATE: Could we try this? I have a suggestion. I think you wanted to
strike the word ‘can help to” –

DR. CARR: Right. That I would –

MR. HUNGATE: — and then change that to just “differentiates”
–

DR. CARR: Yes.

MR. HUNGATE: — “pre-existing conditions from those which may have
occurred during the process of care.” Would that be correct?

DR. CARR: During the in-patient stay?

MR. HUNGATE: Would that fix it?

DR. CARR: Other processes.

MR. HUNGATE: Well, whatever – something of that sort.

DR. CARR: Yes.

MR. HUNGATE: Marjorie?

MS. GREENBERG: I think the reason it was phrased “can help to
differentiate” is because the way this indicator works, you either can
say, yes, it was present on admission, no, it was not present on admission, or
don’t know, because you always have to give people that option because
sometimes the record is not clear as to whether the person came in with the
condition or it – and it developed during the stay. Or it seems unlikely
that they – it would have developed during the stay, like diabetes or
something but they didn’t know about it when they came in.

So I think that’s why it’s hedged —

DR. CARR: Okay.

MS. GREENBERG: — a little bit, but –

MR. HUNGATE: Is that all right? Okay.

DR. VIGILANTE: But they’re thinking of experience of the undiagnosed
condition that was there. They were just unaware of it.

DR. CARR: Right. Okay.

MS. GREENBERG: So then, the rest of the sentence – would it be more
correct to say “differentiate pre-existing conditions from those that may
develop during the admission” –

DR. CARR: Or something like –

MS. GREENBERG: — because this is specifically inpatient care. It’s patients
in a DRG so the process of care goes on for a long time, but this is just those
that develop during that particular admission.

MR. HUNGATE: So the wording would be “that may develop during the
admission” –

DR. COHN: Yes. Judy, do you have a –

DR. WARREN: I have a question about the whole sentence. I mean, are we
trying to focus here on medical errors or are we focusing on care delivery? If
all we’re trying to focus on is try to differentiate what the patient came in
with and what they left with, I mean that’s one thing; if what we’re trying to
do is assist in identifying medical errors, then I think we need to say medical
errors. I just don’t know the purpose yet of what the – what do I look at.

MR. HOUSTON: The next sentence, though, discusses the fact in – t

“The ability to distinguish between conditions present on admission
from development of conditions during a process of care can enable both
mix/severity of illness adjustment at admission and quality improvement
opportunities” which I think would – medical errors and ineffective
process of care would fall into. So the next sentence sort of describes what
the purpose of this is, and I –

DR. VIGILANTE: Well, the basis for patient’s adjustment is – do we need
the sentence that we’re talking about?

MS. GREENBERG: First of all, this language is actually from the quality
report, sort of rewriting history but it wouldn’t be the first time.

The other thing is, I mean, the main purpose of this in states that have
implemented it has been to differentiate pre-existing conditions from
complications. But, I mean – because in particular when you don’t have
something like this, people – institutions actually can get credit for
having more complex cases when in fact the complexity was created by the
institution. Now that’s only part of the picture, but in fact that’s kind of
– so that’s – this is the way it was – I think this was taken
right out of the quality report that, you know, was –

MR. HUNGATE: The more scrutiny anything gets, the more care it gets.

DR. VIGILANTE: Complexity can certainly emerge during the course of their
hospitalization and it could be all the same (?), some ill-conceived process,
or it could just be complication –

MR. HUNGATE: Sure.

DR. COHN: Yes.

DR. VIGILANTE: The example was (?).

DR. CARR: Yes, I mean, that’s –

DR. VIGILANTE: If it’s really got case mix adjustment, then it really
doesn’t apply if it’s not focused on those per se although that could be
included here.

DR. COHN: Yes. Well, I think it handles – I think what we’re coming to
is both medical errors and processes of care which – Justine was saying it
sounds like the subcommittee – the workgroup needs to look at this a
little bit, but I’m not seeing anything horrific going on here. It’s just that
it handles both. So, Bob, are you comfortable with that?

MR. HUNGATE: I’m not uncomfortable with changing the first sentence here to
say “the secondary diagnosis indicator which flags a condition present on
admission can help to differentiate pre-existing conditions from those that may
develop during the admission” and leave that statement that way.

MS. GREENBERG: Repetitive there then of the next sentence, but –

DR. CARR: You might want to make the next – maybe we should work on
this in a quality group tomorrow morning. We could add the point about medical
errors, ‘cause I think that is a high priority. I just wouldn’t want it to
look like that’s the only – that the only thing that happens is a medical
error.

MR. HUNGATE: We’ve allowed time tomorrow morning to do that further
wordsmithing.

Third paragraph: “While testimony received on this recommendation has
in effect made a business case for including the diagnosis indicator on a
claim, the Department should work with interested stakeholders” – it
probably should be “in a claim” –

DR. COHN: Yes.

MR. HUNGATE: — shouldn’t it? “In a claim.”

“With interested stakeholders to more explicitly describe the elements
for effective implementation. Testimony observed that variation in definition
among users can preclude the benefit to be gained from the information
collected. The National Uniform Billing Committee and ASK X12N should work
together to specify the code set, reporting conditions and use cases for the
indicator. Education and evaluation will be needed to assure that the
information is recorded consistently and accurately.” Questions?

Final paragraph: “The Workgroup received supportive and informative
testimony on several other candidate recommendations but there was no consensus
on adoption. The Workgroup plans to develop a report summarizing the findings
from the hearings and will continue to liaise with the key organizations in the
public and private sector that are addressing the information for quality
measurement. Sincerely” – that’s it.

DR. COHN: I guess the only question I have about the last paragraph is do
you even need it?

MR. HUNGATE: Pardon?

DR. COHN: Do you need the last paragraph for any particular reason?

MR. HUNGATE: I think that, you know, there was a lot more covered in the
work than is reported in the letter and that maybe at least alluding to that in
a fairly brief manner as we have here is appropriate.

DR. COHN: Okay. I mean – okay. Well, in case – in that case, let
me ask about the last couple of words in this paragraph which is
“addressing the information for quality measurement.” What exactly
– that probably needs some wordsmith to state essentially what it means.

MS. GREENBERG: The information it needs – needed?

DR. COHN: Is that the information need for quality – okay, maybe that’s
what –

MS. GREENBERG: Yes.

DR. COHN: Okay. Okay, thank you.

MR. HUNGATE: So add “needed” after “information?”

DR. VIGILANTE: Yes, “information needed.”

MR. HUNGATE: Right. Any other –

So we will take that into the letter, reword it and come back tomorrow with
a similar, almost identical letter.

DR. COHN: Yes.

MR. HUNGATE: Now, the ICD-10-CM issue came up in the third of the three
hearings when George, I guess, made the statement that the implementation of
UB-04 was to be concurrent with the implementation of ICD-10-CM. This came as a
bit of a surprise. Maybe that’s an understatement. Marjorie, go ahead.

MS. GREENBERG: Actually, that has always been the plan for the UB-04. I
think what he said was a major change in the – on the UB-02 was the
ability to accommodate the IC-10-CM. Then what – so that wasn’t the
surprise. But what he said was, without any indication that ICD-10-CM was ever
going to be implemented, it phrased out about whether those hospitals should
actually be asked to move to UB-04.

MR. HUNGATE: Okay. I misunderstood. I understood that implication, but I
thought there was more than that and I misunderstood the rest of it.

DR. COHN: You know, I – well, Harry may have a comment; I have a
comment, too. But I mean, neither was I a part of the National Uniform Billing
Committee. Marjorie is, of course, so you have to defer to her, but –

MR. REYNOLDS: With or without ICD-10, you would hope that UB-04 would align
the paper document much closer to the HIPAA required field.

MS. GREENBERG: That’s also a goal.

MR. REYNOLDS: Yes, I mean, that to me should be the number one goal, because
the problem we have right now is HIPAA 1500 and UB-92 or UB-whatever have no
correlation to the information that we’re passing back and forth –
electronically through HIPAA, so the world is still running two incredibly
disparate types of data coming through. So that’s – that is a major
concern.

Whether it would accommodate UB – ICD-10, whether it could accommodate
it versus require it, is another question.

DR. COHN: Yes. You know, I guess the good news in all of this is that most
hospital discharge claims forms are electronic and not paper. I mean, a very
small number of them currently exist in paper. So I guess that is a good thing.

I think the part in all this confuses me a little bit is that my
understanding is that this modifier that we’ve been talking about was
accommodated on previous versions of the UB –

MS. GREENBERG: No, it’s never been – it’s not part of the UB. It is
actually in the 50-10, I believe, version of the electronic –

DR. COHN: Well, I guess the confusion is that both New York and California
have used it in both paper and electronic versions for many years, so I guess I
always presumed that they were able to do that and if there were some –

MS. GREENBERG: Well, that’s because we used to have state-assigned fields
and they had a way to collect it, but it’s not accommodated on the basic UB-92
with the state-assigned fields, plus actually California does not use the
actual bill. And even New York uses – I mean, they use the UB data set and
they spin off the UB data set but they don’t just get a copy of the bill. They
get a –

DR. COHN: Okay. Something different.

MS. GREENBERG: Yes, a tailored version that is very similar to the UB-92
–

DR. COHN: Okay.

MS. GREENBERG: — but it’s somewhat different.

That actually is more difficult with – they can’t really use
state-assigned fields on the UB now after HIPAA but they can get – they
can request something different, you know, from the providers because it’s for
separate –

DR. COHN: For separate reporting, okay.

MS. GREENBERG: They try to align themselves as

much as possible. Most of the states try to align themselves as much as
possible with the UB because that reduces the burden for providers and, you
know, includes quality and –

DR. COHN: Sure.

MS. GREENBERG: The UB-92 does not have any – it has the diagnoses and
it doesn’t have anything after each diagnosis that would – secondary
diagnosis – that would allow their reporting of this indicator.

DR. COHN: Well, I guess my understanding was always that there was a way by
a modifier on the diagnosis that they always used to do it in California, that
was for California.

Having said all of that, I guess the bottom line – the point that I was
actually going to make, and once again I’m not a member of the NUBC so
everything I’ve heard is secondhand and hearsay, but I think there may have
been some members of the NUBC that were surprised by the testimony of the
Chair. They’re having a meeting next week and I guess the question would be, as
I know Marjorie will be there and maybe we can ask, maybe we can ask Marjorie
to ask them if there’s a clarification of understanding on that that occurs
from that meeting, that maybe they could send a letter to the Quality Workgroup
to help clarify that issue, to see if really George’s – George Arges’s
testimony really reflected the intent of the NUBC on this one.

MS. GREENBERG: I know this is going to be discussed at the meeting next
week.

MR. HUNGATE: I think that’s right. But to Harry’s point that the electronic
transmissions that are taking place are not in conformity with UB-92 now, did I
understand correctly?

MS. GREENBERG: Well, yes. Part of the –

MR. REYNOLDS: You’ve got a distortion going on as it goes forward.

MS. GREENBERG: Part of the revision of the UB-92 for the UB-04 was to
accommodate, or to try to align it more with the electronic version, including
like accommodating the NPI, the national provider.

MR. REYNOLDS: Yes – well, anything, as long as there’s a significant
difference in it.

MS. GREENBERG: Those are not totally different because really they –
the 837I draws predominantly on the UB data set. But they’re not aligned —

MR. REYNOLDS: They’re not aligned.

MS. GREENBERG: – because the UB-92 was developed obviously before the 837I.

MR. HUNGATE: Well – and in terms of the information base that people
use, the 837I is the dominant

form, right?

MS. GREENBERG: Is what?

MR. HUNGATE: Is the dominant form?

MS. GREENBERG: Yes.

MR. HUNGATE: — if everything’s going in the

electronic form, then that’s the 837I, right?

MR. REYNOLDS: Well, for hospitals.

MR. HUNGATE: The HIPAA 1500 is for physicians.

DR. COHN: Yes.

MR. HUNGATE: And the HIPAA 1500. Those two then are the dominant data set
and not the UB.

MR. BLAIR: Why does the –

DR. COHN: Jeff, do you want to get closer to the microphone so we can
broadcast your question?

MR. HUNGATE: Didn’t hear you, Jeff.

MR. BLAIR: I feel like this question reflects my ignorance, like there’s an
obvious reason and I just don’t know it. Why do we still have a UB?

MS. GREENBERG: Why what?

MR. BLAIR: Why is there still the UB forms?

MS. GREENBERG: Why is there still a form?

MR. BLAIR: Yes.

MS. GREENBERG: There are actually a number of small, rural hospitals that do
not report electronically, do not bill electronically. They don’t account for
the vast majority of patients but they’re rural small hospitals. But they
– there are – I don’t know what the number is; I think the American
Hospital Association would tell us – but there is still a need for a paper
form.

As you know, providers aren’t required to go electronically. If they bill
electronically, they’re required to use the HIPAA standard. But there’s a
non-trivial number of small hospitals that just – they still use the UB.

And furthermore, even if everybody was billing electronically, I think the
National Uniform Billing Committee would still continue its work because they
are the content committee for the uniform bill for hospitals.

MR. REYNOLDS: In this area, we get about eight percent of our claims,
hospital claims, on paper – 92 percent electronic, eight percent paper.
And those paper come in on the UB-92.

MR. HUNGATE: Thank you. Well, I ended up feeling that we had a strange
system here, that it wasn’t a very clean methodology. As I think about the
collective work, listen to the hearings, I feel like we got more heat

than light, that there’s a lot of energy going in and not much result coming
out, which –

MR. BLAIR: [Laughing} Good way to put it!

MR. HUNGATE: Well, on the whole ability to move the information system
forward in a content sense, you know, I felt a lot of – I heard a lot of
frustration and I don’t see that we relieved much of that by our – we’ve
moved one, ended at recommendation. We have no way of moving the other five
that are really related to it, I think. And if it’s all held up by basically
the implementation of ICD-10-CM, then we’re talking about things that are way
off in the future with as yet no clear statement by the Department that it’s
going to happen.

I end up a little uncertain on whether we’re just working on something where
we’re spinning our wheels.

DR. COHN: Well, Bob, I guess I have a couple of comments on that. I mean,
first of all, I actually would be expecting that you might receive some
clarification that the comments made about a new UB form being dependent on
ICD-10 implementation may be – may have been an overstatement –

MR. HUNGATE: Okay.

DR. COHN: — of the Committee, but that’s just supposition. I think that
clearly the intent is to accommodate ICD-10-CM but I don’t know that everyone
unanimously feels that it should be delayed. Now, that relates to that one.

I think you were also commenting about frustration about all the other
recommendations, and I guess one can take that two ways. I mean, either that
there’s a lot of resistance to the industry and that there’s – you know,
and that it’s going to be hard to make those happen.

But I, having attended at least one day of those – or one session, I
think that there was an ongoing question about the value added in the business
case behind a number of the recommendations, which I think led you to only be
able to go forward with one of them, at least at this point.

MR. HUNGATE: Yes. There are two element. One – that’s one element of
it, one aspect of it. The other is that until HHS clearly states its direction
on ICD-10-CM, there’s a certain waiting that takes place on all the work to go
forward, it seems to me. At least, that was reflected. People say people are
waiting to see.

MS. GREENBERG: That may or may not be the position of the National Uniform
Billing Committee after it meets next week. It is a fact that this Committee
made recommendations on ICD-10 exactly a year ago –

DR. COHN: Sure.

MS. GREENBERG: — and when it was reported on, those NPRMs or kind of rules
that were in progress, there was quite a bit of silence but – on that
topic, but, I mean, that’s the committee to decide whether you are interested
in slowing up on that or not.

MR. HUNGATE: I guess that’s my question.

MS. GREENBERG: But regarding the – well, there are two things that I
think are – you know, that – the cup can be half empty or half full.
I like to look at it as half full. But two things, is that the UB-04 probably
will be implemented in the next – well, it’s not going to make it in ’04.
I would suspect it will probably be implemented in the next year or so because
of the need to align with the 837.

And not – several of the first eight recommendations, or candidate
recommendations, will be facilitated by the UB-04 even without any further
recommendations probably of the Committee. This particular one, I think the
Committee felt was – the work we thought was very timely, particularly
because the 837I 50-10 is also going out for review, and so, you know, it’s
kind of a good concurrence.

But also, AHRQ is planning to do a – I think award a contract or grant,
I’m not quite sure, but to study the cost and benefit of adding selected
clinical information to the administration standards which would be both, I
guess, paper and electronic, in order to inform this whole area.

And they are certainly well aware of the Committee’s report, and since the
report pretty much included all the things that have been recommended in that
regard, I think they will be shedding some interesting and important light on
the business case for or against some of these recommendations which I think
the workgroup felt could not really be made for some of the other ones.

I mean, there were people in favor of it, there were people that weren’t,
and the whole question of the cost versus the benefit just couldn’t really be
resolved by the hearing, so I think a lot of – I think the hearings were
very informative and that’s the benefit of doing what it says in this last
paragraph, which is summarizing the findings without any actual
recommendations.

But I think they – this work by AHRQ will certainly be – you know,
help inform the business case for or against the recommendation. So I think
there’s – you know –

MR. HUNGATE: I wasn’t aware that work was going forward. That’s very
encouraging and will help resolve the uncertainty that I’ve –

DR. COHN: Good. Well –

MS. GREENBERG: I don’t think they’re going to be addressing the ICD-10-CM,
though.

MS. GREENBERG: That’s not part of it.

DR. COHN: But we already did a cost benefit on that one.

MS. GREENBERG: We’ve done a few, yes.

DR. COHN: Yes. I know of one personally.

MR. HUNGATE: We’ll rework the letter and be back with it tomorrow morning.

DR. COHN: Okay, great. Bob, thank you very much. And pretty interesting
conversation. Now we’re going to adjourn the full Committee. I believe that
this room will have the Subcommittee on Standards & Security. Let me just
sort of go through the list here.

Vickie, your Subcommittee on Populations will be in 425A and then at 4:30
the Workgroup on NHII was again in this room. So, thank you all very much. Dr.
Lumpkin will be back in the morning, cheerfully chairing, and we will adjourn
the meeting.

[The session adjourned at 2:58 p.m.]