[This Transcript is Unedited]

National Committee on Vital and Health Statistics

Subcommittee on Populations

November 4, 2004

Hubert H. Humphrey Building
Room 425A
200 Independence Avenue, S.W.
Washington, D.C. 20201

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


Call to Order and Introductions – Dr. Mays

Update on Mental Health – Subcommittee

Discussion on Quality – Subcommittee/Workgroup on Quality

Discuss Joint Hearing with the Subcommittee On Privacy and Confidentiality – Mr. Rothstein/Subcommittee

P R O C E E D I N G S [3:10 p.m.]

Agenda Item: Call to Order and Introductions – Dr. Mays

DR. MAYS: — there’s another meeting after us so I don’t want to run that
over, and then we have Mark coming in and Mark asked so that he didn’t have to
just sit around for an hour if we could really have him at 3:30 rather then
making him wait until 4:30.

So here’s what I want to do, let’s start by call to order so we know who’s
here and then I’ll go right into what some of the issues are on the agenda. Why
don’t we start with Edna, let’s start down there.

MS. PAISANO: Edna Paisano from Indian Health Service.

MR. HUNGATE: Bob Hungate, Physician Patient Partnerships for Health.

DR. MAYS: Nancy Breen?

DR. BREEN: Nancy Breen, National Cancer Institute.

DR. CARR: Justine Carr, Beth Israel Deaconess Medical Center.

DR. MAYS: Vickie Mays, University of California Los Angeles.

MR. LOCALIO: Russell Localio, University of Pennsylvania School of

DR. VIGILANTE: Kevin Vigilante, Booz-Allen & Hamilton.

MS. GREENBERG: Marjorie Greenberg, NCHS, CDC, and executive secretary to
the committee.

DR. MAYS: Telephone?

DR. HOLMES: Julia Holmes from the National Center for Health Statistics,
I’m on the Quality Workgroup.

DR. MAYS: Thank you.

MS. BURWELL: Audrey Burwell, Office of Minority Health.

DR. MANDISHITE: Ron Mandishite, Center for Mental Health Services, I’m here
for the mental health update.

MS. GRANTHON: Good afternoon everyone, Miriam Granthon with the Office of
Disease Prevention and Health Promotion and staff of Populations Subcommittee
and staff person for the mental health workgroup.

DR. MAYS: Great, thank you everyone. One of the things that I sent you an
email about that I said that we did not do and there is very little time for
doing it but I wanted to check in was to talk about the research agenda that
you were sent. I think you have it in here, Gracie, is that true? I emailed it

MS. GREENBERG: It was actually in the agenda book.

DR. MAYS: Okay, so that’s the other place it would be but there’s some
people here who don’t get the agenda book, some people here don’t have the
agenda book but I did email it twice. So I’m going to take that we are going to
fly by the seat of our pants and use what we have that’s in there, if not I’ll
get comments from you outside of this meeting, that would be my suggestion.

Here’s what we’ll do is why don’t we do the discussion of quality, let’s
follow that by the update so that we can do the mental health part, and then
let’s move into our discussion about our hearing with the Subcommittee on
Privacy and Confidentiality. That then, I know Bob wants to get to something at
4:20 so we take care of that, we take care of Ron’s on the phone for mental
health and Miriam, so we take care of them, and then we have, we also can try
and accommodate the schedule for Mark. So how’s that?

MR. HUNGATE: You might want to do the mental health one first, that would
be fine.

DR. MAYS: Oh, okay —

DR. BREEN: And Vickie? Are we talking at all about our report on this
agenda? I had just a couple of questions —

DR. MAYS: Let’s see if we can at the end steal a few minutes because we may
be able to move along.

Agenda Item: Update on Mental Health –

Miriam, we’re going to start with mental health, Miriam and Ron we’re going
to start with mental health so that I think that really facilitates your
schedules. Is Cille, I don’t know if Cille Kennedy is coming.

Where we are, Don’s not here but where we are, this is just an update when
we talked about mental health at our meeting and there were a lot, there was I
think a pretty good discussion, Miriam do you want to talk a little bit about
it since I don’t have notes?

MS. GRANTHON: Sure, and we did a summary for whoever wants an electronic
copy of it. The first conference call was actually in August, the
16th, we had I think we had somewhat better representation across
the department, not complete but really trying to talk about what sort of
mental health statistics are collected in departments and whether those on the
call, it wasn’t clear, so surveillance of mental health there were and then
what prevalence the other was, we had a discussion about that and also defining
mental health, and also it was suggested that perhaps it would be a worthwhile
effort for everyone involved to do sort of an inventory list of what data is
being collected. And also at last week’s Populations Committee there was a
discussion too about doing a little bit more education for our committee in
terms of inviting different folks from the agencies so they that they can help
us to become more informed of what statistics are being collected within the
department. And it was also suggested that there was a previous report done by
someone, I apologize, I’m blanking the name, who was on the committee about ten
years ago but there was a report on mental health statistics and we’re trying
to get a copy of that —

PARTICIPANT: David McCannick(?) was the person.

DR. MAYS: Yeah, we actually have a copy of that, that was supposed to be in
our, I got a copy and I made sure that a copy was faxed to Don and what I’d
asked for was a copy in our folders so it’s not here but we’ll make sure
particularly those of you on mental health get it. I even suggested to Don when
you look at the mental health charge it’s very clear that it’s almost that we
should probably pick up the same charge because it’s asking the same questions,
for example, about an inventory of mental health statistics, some of the issues
that surround what mental health is. And I think Cille is very good about
bringing us some materials in the last meeting of the Population Subcommittee
that pose what some of the issues are in terms of mental health, the definition
of mental health. So I think that that’s part of what we’re also going to be
struggling with here too.

MS. GRANTHON: And also identifying additional folks to serve on this
workgroup, there were several needs that were shared and because not everyone
felt that we had all the correct experts involved and so that was also
suggested. And again anyone liking a copy of the summary please email me and
I’ll make sure you get it.

DR. MAYS: No, actually what we’ll have you do is email it, we’ll have
Gracie have it because I haven’t gone through it but we’ll send it to Gracie
and then we’ll send it out, I think that’s the best way to do it.

Ron, I want to thank you for actually coming on, if you could talk a little
bit about the statistics that are collected at your agency and your role that
would be very helpful to us so that we can kind of get you integrated into this
workgroup. I saw Cille’s email to you and so I’m very happy that you were able
to be on the call.

DR. MANDISHITE: Okay, well let me give just a little bit of background on
the federal program and then on mental health in this committee. First on the
federal program there’s a very long history to the federal program of mental
health statistics, in fact it goes back about 175 years, and it covers both
population epidemiology and service statistics. That program originally was
operated by the superintendent of the Census, then they moved to the Public
Health Service, then it moved to the National Institute of Mental Health, and
then it was subdivided about ten years ago where population epidemiology stayed
with NIMH and all the services work came to SAMSHA, which is where I am and
I’ve been associated with this area for about 20 years. I’ve published Mental
Health United States where we publish both population epidemiology work,
services statistics, financing statistics, new work in performance measures and
so on, so there’s quite an array of things.

In addition to that we work very closely with NIMH and Ron Kessler on the
National Co-Morbidity Survey which will be the next generation of our
epidemiological work. We are working with CDC in Atlanta on some population
based work and are designing some new approaches for community surveys. We also
work with NCHS on the National Health Interview Survey, we’ve had mental health
content in that survey ever since the time David McCannick was the chair of the
Mental Health Committee and the National Committee on Vital and Health

So there’s a long, long history and I don’t want to bore you to death with
this but a lot of this history has been written up, we can provide some of the
historical materials, we can provide current electronic copies of Mental Health
U.S., we can describe some our new initiatives, decision support 2000 plus and
so on. So that’s kind of the two minute version of our history there.

In terms of National Committee on Vital and Health Statistics, if you go
back ten years, there was in fact a subcommittee on mental health that we used
to staff, there was a vast reorganization of the national committee as I
remember about 1994 or 1995 where all of the existing subcommittees were ended
and a new set of committees was structured including after a period of hiatus
of maybe a year or two the Subcommittee on Populations, which I guess is the
committee I’m talking with today. Some of these specific fields have had
difficulty taking root in this new organizational structure and I will cite
mental health as one of those, I probably would cite substance abuse as
another, and there may be others, public health and so on. So we’re very happy
to be invited back into the committee and we look forward to working with you.

DR. MAYS: Thank you very much. I think, here’s the next step, Don who is
not here is actually going to be one of the lead people in the mental health
area so he and I are going to be talking soon and I think we probably should
make it the three of us, because I think it would be very useful for you to
update us on exactly, I mean I’m a psychologist by training and I did not know
about, what is it, Mental Health U.S. —

DR. MANDISHITE: Mental Health U.S. has been published by the government for
the last almost 25 years.

DR. BREEN: Must be one of our more low profile efforts.

PARTICIPANT: How do we access that?

DR. MANDISHITE: You can go to the website, www.samsha.gov, click on the left on mental
health statistics and up will come the cover of the 2002 version which is the
last version that’s actually out there. It’s a 21 chapter 400 page document and
I’m not working on the 2004 version.

DR. BREEN: You publish it every other year?

DR. MANDISHITE: Publish it every other year, would love to publish it every
year, don’t have the horsepower to do it.

DR. MAYS: Just so that you know because one of the members, Eugene,
reminded me, is we actually get Health U.S. sent to us and we’re asked to
comment on it so part of probably maybe getting you back into the loop might be
for you even to think about whether you want us to, I’m not going to make
promises since you just told us it’s 400 pages but just so that you at least
get a little PR —

DR. MANDISHITE: We’re happy to work with you.

DR. MAYS: Okay, great. Okay, so what we’ll do is next step will be a small
group conversation around these issues and I think that will help us enormously
in how to move forward, and then we can get from you whatever documents that we
need and I think Gracie will pass around because I do have a document that is
probably about ten years old and it was when David McCannick, 1992 when David
McCannick was the chair, that is kind of the mental health work that took place
back then.

DR. MANDISHITE: No, no, I understand, and who’s taking right now, Vickie

DR. MAYS: Yes.

DR. MANDISHITE: Okay, good, so it will be you and Don Steinwachs and I and
I’ll be happy to work with you.

DR. MAYS: And the staff, right. Okay, thank you very much. Okay, so that’s
your update on mental health folks, if the mental health folks wants to move on
in terms of the call they should feel free.

Agenda Item: Discussion on Quality –
Subcommittee/Workgroup on Quality

DR. MAYS: Okay, let’s do quality, Mark, we’re going to do quality and then
right after we do quality then you’re on.

Okay, let’s get started, I mean we can actually do this, I mean I’m pretty
good about doing this without the notes.

Bob, part of what we need to I think have a discussion about is what we’re
doing jointly, how we want to do it, when we want to do it, we had a pretty
good conversation. Justine was there, Julia was there, they called in for our
meeting of October 26th and maybe what we should do is try and recap
that just a bit so that we can talk about where we’re going with it and then
see kind of where we are relative to what, you all meet tomorrow so that’s why
I was like trying to make sure that we at least had a discussion today about
these issues.

It became clear in our discussion that what you’re calling candidate
recommendation 11 and 12, they do to some extent I think fall into an area that
populations can I think assist you in terms of moving those recommendations
forth. And I forgot, some of you probably don’t know what they are —

PARTICIPANT: Were they supposed to be in the folder also?

DR. MAYS: Lots of things supposed to be in the folder. I have everything, I
think you all don’t have some of the things but the notes —

MR. HUNGATE: Want me to read them out?

DR. MAYS: I have my copy of it so let me go to mine but I realizing you all
don’t, that’s why I suddenly realized when I said that that you didn’t have it.

Candidate recommendation 11 is modify existing mechanisms for reporting
race and ethnicity of subscribers and dependents on the HIPAA enrollment
transactions. And priority area is reducing disparities in quality, data need
is the data on race and ethnicity of all enrollees, strategic focus is existing
data systems, potential options to be considered, is consider revising the ANSI
X12N 834 implementation guide and whereas the relevant public and private
sector initiatives, I don’t know what PHDSC is but there’s IOM and PHDSC.

MR. HUNGATE: Public Health Data Standards Committee I’m guessing, pure

DR. MAYS: It’s almost like you stick around here enough, hey, the acronym
list. And 12 is investigate how best to capture race and ethnicity on the
standard provider transaction. So again here we’re talking about data on
ethnicity of patients, we’re talking about the strategic focus here is in
existing data systems. Potential options to be considered, consider revising
NUCC and NUBC datasets and ANSI X12N 837P and I to report race and ethnicity on
all 837 claims, and on 837P claims containing a CPT ENM code for new patient
service, and add race and ethnicity to the claims attachment standard or
develop a new standard transaction. And relevant public and private sectors,
CDA, Institute of Medicine, CCR, and PHDSC.

DR. BREEN: I think we get the general gist of that but HIPAA, I’ve signed
several things related to HIPAA when I go to the doctor now, are you suggesting
collecting race and ethnic data when you sign? Is that what the recommendation
11 is about or what does that mean? I’m kind of asking you, Bob, because I
think you’re familiar with this then even Vickie.

MR. HUNGATE: The form that you fill out when you go to your physician for
HIPAA is not an enrollment form per se —

DR. BREEN: But you don’t enroll in HIPAA, it’s just a protection.

MS. GREENBERG: People in health plans fill out an enrollment, usually it’s
through their employer or directly with the health plan, this is not when they
go to the provider, this is when they actually enroll in a health plan, so
that’s 11.

DR. BREEN: So everybody who’s already enrolled in a health plan would never
do this.

MR. HUNGATE: That’s correct, you probably don’t do it again unless you
change health plans and then you probably do it again.

DR. MAYS: Or I would assume that what would happen, and again maybe I’m
making an assumption, is that if the health plans decide to do this that there
would be some point at which they would try and gather this information, would
that not be correct? If you were already enrolled, you were not a new enrollee,
that there would be some point at which when you come into the system either
for a visit or something they would want to collect this.

DR. BREEN: Well, that’s the problem, you don’t go to your insurer for a
visit, so I mean that is the issue though, coverage, would we be getting data
on the entire population by collecting it that way, what would we have at the
end of the day —

[Group discussion.]

MS. GREENBERG: I believe the enrollment transaction currently includes race
ethnicity but it does not, I don’t think it includes the current OMB standard
for enrollments, and also it doesn’t include information on dependents, I think
only on the subscriber. So that’s why the candidate recommendation was to
modify the existing mechanisms and I think there’s more discussion in the
actual report.

MR. HUNGATE: Which I don’t have with me.

DR. MAYS: Okay, here’s what I’m going to suggest —

MS. GREENBERG: — Nancy mentioned though, because of those limitations
that’s why there’s also number 12, which is about the transaction actually
filled out on every encounter —

DR. BREEN: So that’s the whole idea is to get full coverage of race
ethnicity via insurance companies, so this would just be for private insured

MS. GREENBERG: Except that Medicare does not use this enrollment.

DR. BREEN: But Medicare collects data on race and ethnicity —

MS. GREENBERG: They don’t, they get data, for Medicare this is a long
standing —

[Phone interruption.]

— when in fact many of the really elderly enrolled only black, white, and
other was collected. However they have done various things over the last 20
years I’d say to try to improve that data but they still have limitations in
Medicare and for that matter Medicaid data on race and ethnicity because they
do not routinely, they don’t even use —

DR. MAYS: Okay, here’s what I’m going to suggest which is I think that in
order for us to work well with you that we need a little tutorial because one,
I think many of the numbers, the alphabet soup and the numbers, the X, the 12N,
only some people are as up to date on that so if we’re going to work together
what we’ll ask is that if you all can plan at some point to have us in a little
tutorial we’ll be happy to do it, you can either make it an online tutorial
where you send it to us and we read it, or you can do it on conference call.
But if it’s on a conference call the only thing I’d ask is that you send it to
us before hand so that we can actually look at it.

DR. VIGILANTE: And if you missed the call it would be nice to have
something to review anyway.

DR. MAYS: So the how, what we know is we need to collect this and part of
what we need to deal with is the how, how and where. And what I think probably
is good for us to do is for now given we have limited time is rather then
trying to deal with the technical aspects why don’t we walk through the broad
aspects of what the plan is. And then I think you all are having your meeting
in the morning, right, and then maybe you can then decide what you want to do
with us or whether you need us or is it just, you want us to comment on
something and you hold your own hearings, or we’re holding them together, just
we need a little guidance.

MR. HUNGATE: Let me talk out loud and see what comes back.

DR. MAYS: Okay, as you talk out loud then I’m going to ask people because
I’m going to try to stay to my time, I’m going to ask people to be looking at
this document as you talk out loud, so go ahead, the document talks about,
these are the notes from Justine and I acknowledge that I made some additional
comments so if anything’s a problem don’t hold her to it, hold me to it.

MR. HUNGATE: I’m working from an inherited agenda and because of that I
have a need to check priorities, so I need to understand what’s the relative
emphasis of things. In the context of the Populations Committee I see a lot of
work that’s gone into the generation of a complete document about surveys —

DR. BREEN: It does focus on surveys.

MR. HUNGATE: It’s focused on surveys. Now the question I want to ask is
this, if the point of the information is to reduce disparities —

DR. MAYS: You’re talking about the information about the surveys?

MR. HUNGATE: The information on race and ethnicity, with two datasets,
there’s an administrative dataset, there’s a survey dataset, which dataset is
more critical for reducing disparities? That’s question one. Because I think
that’s a prioritization question, to me the way I see the system work is that
the survey questions are excellent baseline, they create the baseline here’s
what things really are at some macro level, but the change to address
disparities is done at a micro level, in a community I use levels and arbitrary
examples, it might point me between the Navajo and the Hopi in the State of
Arizona, or there may be disparities between those two tribes where most of the
people that serve are Navajo and the healthy are discriminated against, I don’t
know. So my suspicion is the disparities exist in communities with their own
culture where the perception of race and ethnicity by another observer is what
is the basis for whether disparities exist, that the cultural expectations
result in disparities. And if we’re going to reduce them then it’s got to be
measured at a local level which probably is administrative data.

DR. BREEN: I think in theory the premise is very sound that you just laid
out but in practice we don’t have two datasets, we have some surveys, some
national surveys which allow us to get national data. We have one dataset that
gives us a survey that gives us data on states which is more or less
comparable, I’m thinking of the behavioral risk factors surveillance system. We
have one state that gets really good local data from a survey. And then we have
the Medicare dataset, the administrative dataset that we were talking about
which is again a national dataset with millions of records in it, and some
issues with racial ethnic data, for example what Marjorie just said, that it’s
not collected recently or in a way that conforms anywhere near to OMB standards
necessarily, and probably under counts American Indians big time and some small
populations, Asians I would guess are badly under counted in that dataset.

And the other administrative datasets that we have are sort of here, there,
and everywhere, I mean we don’t have them, and I think this is one of our
biggest problems as a nation, that they’re not organized into a single dataset
unless you know something or have some plans that I don’t know about.

MR. HUNGATE: Let me give you a contrary view —

DR. MAYS: Let me jump in here in terms of how we’re going to use the time,
I would suggest that it appears to me that there needs to be these
philosophical discussions and if we need to have that then we don’t need to
talk about hearings yet. It means we have to have a meeting to sit around and
talk about this, that it doesn’t sound like we’re ready to move yet. When you
talk about either/or in terms of one or the other it’s kind of like for
example, I’m kind of say, kind of piggyback on what Nancy is saying,
theoretically I think we could do that but in reality it’s almost like if you
don’t have these national surveys CDC then who is responsible in this country
for prevention and who then allocates the resources down to these local levels
to deal with these problems, it doesn’t have a guiding rationale for being able
to do that. So if anything those surveys really guide quite a bit of resources
and health planning that takes place for health departments, state health
agencies —

MR. HUNGATE: I hear you. The point I’m trying to get into the discussion —

DR. HOLMES: — surveys add information about people who are not accessing
or in the health care system and in that way also add information about
disparities in health care across the U.S.

MR. HUNGATE: Part of what I’m trying to get at is that there are also
resources within health plans, within community health centers, that there are
the public health resources that you’re referring to and there are the private
health resources that are applied to populations. And the benefit of
administrative data is in the local use of information to identify disparities
in treatment within a geography and address the disparities.

DR. HOLMES: — I think that the discussion is perhaps become convoluted and
the issue is I think that we all agree that there needs to be added, whether
it’s to surveys or whether it’s to administrative data collections, we need to
have good data on race and ethnicity regardless of what those two different
kinds of data collection systems apply that information to, and that we don’t
need to separate them out in terms of A, whether it’s a good idea to gather
data on race and ethnicity, or what data should be gathered or how it should be
gathered, there may be differences but overall I think that the importance is
the same regardless of what kind of data collection system you’re talking

DR. MAYS: Kevin and then Marjorie.

MS. GREENBERG: We had some of us philosophically —

[Group discussion.]

MS. GREENBERG: I was going to say we had some of the “philosophical
discussion” at the last meeting of the subcommittee and whether it be the
philosophical discussion, unless, I mean I guess if you aren’t ready to say you
even want to investigate this area and these candidate recommendations, then
fine, we just defer it. But the best way to get information first of all about
the current state of things as opposed to staff running around trying to do a
tutorial is to do what we did at the last work meeting of the workgroup and
that is to bring in the people from ASC X12 who know and understand the 834 and
the 837 the best, better then any of the staff do, and who are not only
responsible for, who not only understand the current 834 and 837 but how it is
changing or evolving and what the opportunities are. As well as provider groups
and employer groups or whoever to find out what the pros and cons and costs and
benefits are of pursuing these two recommendations. So the only reason not to
do that at this point I think, not to plan a hearing to do that, is if you
really don’t think it’s worth doing anyway so why have the hearing. But I think
this could be knocked around philosophically in this subcommittee for a long
time without maybe getting more clarity from the outside, the people both who
are responsible, these two recommendations are about particular transactions
and considering revising them, and it seems to me that the next step is to
bring in the maintainers of these standards as well as the users of these
standards and see what the options are.

DR. MAYS: I guess I was talking about something very basic and that was the
alphabet, that I think people need to know that and see the forms. Some of us
actually have read —

MS. GREENBERG: Well they aren’t forms because these are electronic
transactions and they’re not that easy to just, I mean frankly it’s not like
looking at a form.

DR. BREEN: I was just going to say first I apologize for getting us off
track because the person on the phone was dead on, I feel into a trap, I mean
we weren’t competing over whether we wanted to talk about surveys or
administrative data, I think right now we’re talking about administrative data
and how we want to handle that in terms of improving racial ethnic categories
on it. And then I think that’s what we’re talking about, right? How to improve
race and ethnicity on administrative data? How to collect it better, better
quality? And then the numbers that you just mentioned, Marjorie, are those how
you collect data on administrative data?

MS. GREENBERG: Yes, you either collect on patients, there are, clients, or
whatever, there are two ways of collecting information, all the other HIPAA
transactions have to do with payment status and this that and the other, there
are only two. There’s an enrollment transaction which is inconsistently used,
and there’s a encounter transaction which we heard now 98 percent of all
Medicare encounters are using and a growing percentage of non-Medicare. So
those are the only two ways in the administrative transactions that you’re
going to collect, potentially collect information on race and ethnicity.

DR. BREEN: And those are the two 800 forms you just mentioned?

MS. GREENBERG: The 834 is the enrollment transaction, the 837 is the
encounter transaction.

DR. HOLMES: And Marjorie, is the encounter transaction, does that cover
both inpatient and —

MS. GREENBERG: We’re also, we are of course every state, not every state,
the vast majority of states have hospital discharge systems, they aren’t
surveys, they are all the encounter data, all the encounter records within the
state and the majority of those do collect race ethnicity, they don’t collect
ethnicity very well, they do collect race —

DR. HOLMES: Not all of them, Marjorie —

MS. GREENBERG: Not all of them, no.

DR. HOLMES: For example in the National Hospital Discharge Survey here at
NCHS we use in part some of those hospital discharge abstracts that are sent to
us and sometimes they’re missing data on race approaches, it’s all missing —

MS. GREENBERG: I was thinking of the HCUP, the Hospital Cost and
Utilization Project, which is about what, 20 states?

DR. HOLMES: They have a subset of states where they perceive that they do a
good job in terms of reporting race ethnicity and those are now up to I think
about 20 or 22 states.

DR. BREEN: And can those data be linked? Do they have some sort of
identifying information?

MS. GREENBERG: No, they can’t be linked with anything.

DR. HOLMES: Because they’re discharges, they’re not people.

PARTICIPANT: Link to what?

DR. BREEN: Well, for example the Medicare data has an identifying number
which is not the person’s name or address or anything that you or I would be
able to make heads or tails of but it’s unique and it allows you to link it to
other datasets.

PARTICIPANT: Relevant to that person.

DR. BREEN: Correct.

MS. GREENBERG: You don’t have any longitudinal data —

DR. HOLMES: These are discharges and they’re not necessarily people.

DR. LENGERICH: Marjorie I think there was a question on the phone and I’ve
got the same question, so does the 837 apply to both inpatient and outpatient?

MS. GREENBERG: There’s the 837I which is institutional and the 837P which
is professional, and the I is used for all inpatient and for hospital
outpatient, and the P is used for all other outpatient and the physician,

DR. MAYS: This is when I said about tutorial, this is actually what I mean
is bringing us up, because I think we know the other stuff but we didn’t know
the specifics

MS. GREENBERG: This recommendation was only about, it was about the I and
the P if you look at 12, the enrollment doesn’t differ, I mean it’s just
enrollment in a health plan but if you look at recommendation 12 it refers to
the 837I, which is the institutional claims, and the 837P. But the P, the
recommendation was more nuanced that it would only be for new, somebody
identified as a new patient. So if a provider has seen this patient before, I
mean presumably they only would have to collect this once, if Mrs. Jones comes
in monthly for her blood pressure or blood sugar check or whatever you would
collect it the first time she came in and then you would have it in your record
and you could report it but you wouldn’t have to collect it every time.

DR. BREEN: Would you grandfather old patients in?

MS. GREENBERG: All of that would have to be worked out obviously.

DR. MAYS: Let me make a suggestion and see because we’re kind of doing
parts of process here that I don’t know if we know where we’re going with it
because I thought from our meeting on October 26th when we had a
discussion that we were like maybe ready to make some headway. Maybe what we
need to do at this point is whether the population putting this out there that
it needs to be the Quality Workgroup comes back to us with what you want to do,
I mean I think in our subcommittee, the Subcommittee on Populations on the
26th, I think we laid out several issues, I think that those are, I
mean you have the wisdom of our thinking, our willingness to move forth on
this, but I think maybe what we need is to just really be specific from you as
to what you want. We see the importance of this and we think the issue of race
and ethnicity in collecting it in these areas is important. We can either
pursue this question, if not a Quality question but pursue this question as a
data collection question purely with the health plans, or Quality can determine
what it would like to do and if it wants to involve us. At this point I’m just
pretty confused as to how we can be of assistance to you because we’re spinning
our wheels here.

MR. HUNGATE: I’m going to express my disappointment for not being here on
the 26th because I didn’t have the flavor of that. What you just
said about the Population Subcommittee wishing to pursue with health plans the
collection of racial and ethnic data is I think the whole subject that the
Quality Workgroup has proposed and so we’re entirely in support of that.

DR. MAYS: Okay because we wrote a letter some time ago actually saying that
we thought that that was important, which is I think part of what helped to
facilitate in these other arenas —

MR. HUNGATE: If it’s important, I think that the next step that Marjorie
has outlined is the appropriate right step for getting to the what in
implementation, I think it will require a hearing —

MS. GREENBERG: The what, the whether, and the how.

DR. MAYS: The only part that I’m disagreeing with is that I do think that
this group needs some simple things, like what Marjorie just did, before we’re
sitting in a hearing with the people who are the experts. So again, maybe
tutorial is like being the wrong word but it’s like just simple things like
that helps you to make sense of it. And the reason I’m saying that is because
I’m now set on several phone calls and hearings and stuff and I’ve kind of
learned by quick catch up and I even go on the internet and look for this and
ask people and stuff so I’m a little caught up and still a little confused but
a little caught up but I think the rest of the individuals sitting at the table
unless this is what they do I don’t think they are. So I think to make it
easier —

MS. GREENBERG: We can provide something by email, I’ll have my colleague
who’s very, very knowledgeable about the X12 transactions write something up
and we’ll send it to you.

DR. MAYS: And that’s I think the level at which I’m saying, when someone
comes in then we can actually be in a position to not ask really dumb
questions, it’s okay to ask a dumb question but a really dumb question because
you didn’t do anything beforehand, no.

PARTICIPANT: Or only dumb questions.

DR. MAYS: Yeah, you stay dumb the whole time.

DR. BREEN: Vickie, one of the things we’ve used the hearings for in the
past is to bring us up to speed and I think people have always had different
levels of knowledge pertaining to different parts of the talks, some people
could have given the talks and other people really were pretty clueless or
hadn’t ever heard of the survey before for example. And then we all went away
from the hearing with Pacific Islanders feeling like we needed a geography
tutorial, just about everything tutorial related to what we were hearing there
but we still got a lot out of it. So I’m wondering if maybe that information
could be conveyed at the very beginning of the hearing that you’re talking
about, I don’t know how long this hearing would be or exactly what it would be
but maybe that would be the way to do it.

MS. GREENBERG: That’s what I had suggested but I can also send you

DR. BREEN: I think you should have a handout.

DR. MAYS: I agree that I think we can have the hearing and do that but at
the same time I think some of the individuals that you want to invite, I think
they kind of did this already so let’s just make sure they understand that it’s
a different group, we’re kind of the special ed group, we need to catch up with
them and that will be fine. Because in a hearing on the 14th some of
that did happen. Justine?

DR. CARR: I’m just taking two steps back trying to integrate what we’ve
discussed earlier today with now and I think that we have the big picture that
this is important and why it’s important. And we have a plan of where we can
get the data in some ways. And we know that that will not be perfect and so it
gets back to, but the question is is it good enough and it gets back to what we
were saying, or what I was saying this morning, which is to move this ahead
what is it that we need, do we need detailed data or do we need simple data and
do we need complete data or incomplete data. Because I think having that sort
of understanding of what will take us to the next stage will inform this
particular, and maybe it will help formulate some of the information we’ll get
at the hearings but from just what we’ve already heard we will get this, some
of it exists, some of it will new, some will be old patients, some will be
different —

DR. MAYS: And some of it will be not so good.

DR. CARR: Yeah, and that might be okay but I think we need to have a point
of view that what we’re setting out to do, this particular intervention of
putting it on these forms is going to make us better how, or inform us better,
or it’s going to move the whole agenda ahead in what way. And so that, I would
just put that out there that we understand that before we have the hearings.
And I think in terms of taking, who takes the lead on it, Bob and I were
talking about this earlier that really the expertise lies in Populations and
the recommendation from Quality is a technical approach to getting that but I
really think and I think Bob agrees that the context and the understanding of
the universe is really your expertise and so this would be one technical way of
moving it ahead. But I still think in terms of philosophical discussions, not
so philosophical but how will we be better, what’s enough to get us to the next
level is a question that we need to understand.

DR. MAYS: See and I think, I agree, and I think that where, we’re starting,
I think it’s the issue of where we’re starting and it’s like we don’t know the
technical part and that’s almost where we’re starting and so we keep saying if
you want us to help you with that then we need to know a little bit more about
the technical parts to know what questions to ask. When there was kind of like
this discussion previously about health plans the question really became things
like what can they collect, what’s the burden in terms of the collection, we
recognize that for example the issue between how data is collected at a federal
level versus how data may be collected at a state level, and how the states may
want to display it quite differently then they do at the national level, that
those are really some of the issues around how good is it, how good is it if
you’re sitting in Massachusetts or you’re sitting in California and you’re
sitting in California the best you can tell me is black, white, Hispanic, and
other, it’s not good enough to some extent because knowing for example that
someone is El Salvadorian they tell me, I need to ask another set of questions,
I need to pay particular attention to, there are characteristics that may be
associated with knowing something else about your race and ethnicity that we
know from place and migration and things like that.

So I think that that’s how good it is may also depend upon a lot of factors
and that I think is part of what we would want to spend, we would want to spend
time in a hearing asking for example health plans to come in and tell us what
is the burden, who typically does this, what it is that you think that you
need, are you trying to do this by science, if you’re trying to do it based on
science then we know, then we actually don’t have scientific data that’s all
the way down to some of these smaller populations. But if what you’re going to
try and do is to be able to better determine from a clinical database, better
determine things that you should do, then you probably do need that.

So I think these are kind of some of the decision making points that I think
we would think we might need to pursue, which is different to me then the

DR. BREEN: One thing I’m thinking that it might be useful to have at this
hearing, we spent some time when we had our other hearing talking about what
would be the things we would need to know from this hearing and set up a series
of questions and then we went out to find people who could answer those
questions for us. And I’m thinking of one of my colleagues who might be a
useful person to present at such a meeting because what she does is she’s in
charge of the linkage between the CIR(?) data and the Medicare data, the cancer
registry data and the Medicare data, and she also works with a lot of grantees
on analyzing the data and so she’s used the race ethnicity category in analysis
and she’s familiar with it and the extent to which the data is actually
collected, race is collected by both CIR and Medicare and so at some point
they’ve got to resolve issues that come up between, inconsistencies.

And so I think if we, most of us are, on this committee I think are data
users and to some extent data producers but we don’t know like what Vickie
said, the technical aspects of sort of obtaining the data in a clinical setting
and what you might want to feed back to clinicians. This is like a whole set of
things that we really haven’t dealt with on this committee and so I think we
want to know that because part of it is going to be the purpose to which it’s
used, who’s collecting the data —

DR. MAYS: How perfect it needs to be —

DR. BREEN: Right, and not only how perfect it needs to be but how detailed
it needs to be because you were saying El Salvadorian, that’s really boring
down pretty detailed —

MS. GREENBERG: It not only isn’t going to be perfect, it’s probably not
going to be close in administrative data but Julia, are you still on the phone?

DR. HOLMES: Yes, I am.

MS. GREENBERG: Do you want to comment on just how imperfect it currently is
for like doing the disparities report?

DR. HOLMES: Well, I mean the issue of it being imperfect is that there’s so
many groups for whom there is too little data to report on, in other words low
event rates say of diabetes combined with small size groups mean that we cannot
report on the percentage of folks from a certain race ethnicity who have all
the recommended diabetes tests simply because in the surveys the sample size is
too small. So the problems in the disparities report are primarily not
necessarily inaccurate data but missing data and simply not enough cases for
groups that have only small representation in the population. But some of the
issues for the administrative datasets would be a little bit different in terms
of but we at least want them to be comparable in terms of how data are
collected on our national surveys as well for comparability purposes. You can
collect a lot of detail but they should be able to be collapsed down to for
example the 1997 OMB race and ethnicity categories.

And that’s why these are the sorts of the questions that we need to have
answered in a hearing about how that might be accomplished and I would think
you would want to speak to folks that do have experience in asking people about
race and ethnicity not in a doctor’s office or in a hospital or at the point of
enrollment but for example folks from the Census and people from the National
Health Interview Survey, not only the national one here but for example the
California National Health Interview Survey where they are asking people very
detailed questions to drill down from just simply Asian Pacific Islander.

MS. GREENBERG: you clearly can do things in a survey that you can’t really
in an administrative environment.

DR. MAYS: But again, I think we’re maybe pre-deciding because it’s like if
somebody is going to collect this data you want to know what they want to use
it for and how much they want to invest in it, I mean it may be that if you’re,
remember, we really have to, I think in this issue we really have to keep
separate that the private sector and it has marketing needs, I mean it has a
whole different set of needs I think then we’re talking about in terms of at
the federal level and also in terms of national data. We have to remember,
maybe I should send that to people again is, remember Carl Volpeg(?) came in
and gave a presentation that is based on kind of the experiences of Well Point
and he took all those data points and turned them all around and said we’re not
concerned about this but we are concerned about this, marketing is a big issue
and if collecting the data also helps them in terms of marketing for all you
know they may want more detail —

DR. HOLMES: I don’t think, I still think that a part of the hearing should
be, we discussed this at the 26th meeting Vickie about trying to
develop a business case and a justification to encourage all health plans and
providers to collect data on race and ethnicity so many of them do not
currently do so nor are they mandated to do so.

DR. MAYS: I agree.

MR. HUNGATE: That articulates the purpose of the hearing, build the
business case for the health plan collection of race and ethnicity data for the
use that benefits them.

DR. HOLMES: That’s right.

DR. MAYS: And I think that’s helpful because we’re very specific because
we’re talking about health plans and the only difference we’ll talk about is
our, later we’ll say whether or not it’s just private or if we’re talking about
also —

MR. HUNGATE: If we start with that narrow we can broaden that as we need to
but let me make the suggestion of the next step, that the Quality Workgroup
will make an outline of the hearing based on the past three hearings that we’ve
done that are a little bit parallel and say this is what we think for starters
might be what’s involved, and then we can share that —

DR. MAYS: You mean questions —

MR. HUNGATE: Questions —

DR. MAYS: The people stuff I think, we let you all fill that in later but I
think what’s important for us is to understand what it is that you think needs
to be asked so that we understand.

MR. HUNGATE: I think that both the questions and the people that have to
answer them, I think you need both together in order —

DR. HOLMES: This is Julia, I didn’t get Justine’s note, did you, Justine,
from our 26th meeting where we had come up with a whole list of
potential organizations to testify —

DR. MAYS: I don’t have a working email list for you all because I don’t
always get emails, I get them sent by the staff to me afterwards. She needs it
sent by email to her electronically.

DR. BREEN: Julia, it’s NCHS, AHRQ, private, large employer groups have
difficulty requiring race ethnicity on claim form or enrollment, that’s
probably where we really need to target the business case. Health plans, and
consult with Harry Reynolds on his ideas, he works at Blue Cross and Blue
Shield in North Carolina, state, implementation provider challenges, and that
would be the American Hospital Association and the American Medical
Association. And then the pharmaceutical industry, pharma represents that
industry, enrollment in clinical trials, difficulty in using categories
offshore, I’m not sure what offshore means here. Oh maybe, I see, I guess it’s
drugs brought into the United States that are not produced here or are sold
through the internet or something. And then the VA and CMHS, I’m not sure what
that H is —

DR. HOLMES: What is it?


PARTICIPANT: MHS would be Military Health Service —


DR. HOLMES: — Center for Medicare and Medicaid Services —

DR. MAYS: Yeah, that’s what it’s supposed to be, it’s probably a type, CMS.

DR. BREEN: So that’s pretty comprehensive —

DR. HOLMES: — in terms of making the business case not only to for example
large employers but also to their workers and dependents and I’m talking
primarily about large employer groups that have a unionized workforce, I wonder
if we might want to think about asking a union —


DR. HOLMES: Say the UAW to also represent their perspective because as I
think I suggested before in a former life at Blue Cross Blue Shield of
Michigan, and they handle the business, the health care administration for the
auto, we were not, we did not collect data on race and ethnicity because it
was, we were not allowed to. And I think one of the concerns has been
historically the use to which data on race ethnicity are put and I think it’s
for example large unions that might represent those concerns and I think we
would have to deal with them at some point.

DR. MAYS: Okay, here’s what I’m going to suggest because I think that we’re
getting into actually designing who and the who is going to come from the
Quality Workgroup so I think in the morning then you all should come up with
the who and questions that you want to send to us. And then also I think it
would be great if you all also discussed a mechanism for how you want to work

MS. GREENBERG: Who’s you all, us?

DR. MAYS: Them, the Quality Workgroup, sorry, I’m being very southern here,
the Quality Workgroup, so that we have a sense of who for example you want to
see brought in, how you want to do this, whether each time your whole Quality
group comes to Pops, or the whole Pops goes, we need to decide whether we’re
going to take a subsection because we just need to have a process by which we
need to do this. And on our side we need to make sure that, I mean Julia is
being great but we also need to make sure that there’s staff from Populations
Subcommittee that participates in Quality Workgroup activities so that we have
that continual cross fertilization. So I’m going to try to bring this item to a
close, today is one of those days where we have such limited time because of
whatever reason, we don’t have our usual three hours. Is it clear what we need
to do and how we’re going to proceed? You all have your meeting in the morning
and I will come to your meeting in the morning so is it clear what the next
steps are?

DR. HOLMES: I think so.

DR. MAYS: Okay —

PARTICIPANT: And the meeting was changed from 8:30 to 8:00 for Quality —

Agenda Item: Discuss Joint Hearing with the Subcommittee
on Privacy and Confidentiality – Mr. Rothstein and Subcommittee

DR. MAYS: Okay, no California time for me. Alright, folks, we have a guest,
let’s treat him like a guest, that’s the way we can get him out, otherwise
we’re going to keep sitting here like Mark’s a part of the committee but let’s
treat him like a guest —

MR. ROTHSTEIN: Oh, you’ll want to get rid of me after —

DR. MAYS: No, no, just as long as you don’t bend over too far to touch

I want to switch now and start talking about some of the concerns that we
have about privacy, confidentiality, and how several things that we’re doing
now these issues are coming up. We had a meeting with the National Child Study,
people from NICHD, and again the issue of collecting biological samples, I mean
they’re worried about these things but they also need some guidance on them. We
had some joint worries with them about for example whether or not when they
collect data on racial and ethnic minority groups, whether or not they can put
into place now procedures, while it protects the person at the same time
there’s accessibility for the data. So they would love for example if there are
any hearings or will be hearings on some of these issues if they can
participate with us if we on their behalf for example could start asking some
questions about this issue. They’re pulling ethicists in but it was clear that
kind of some of the broader rules and regulations around federal data still
need to have I think more discussion because the protection needs to be there
but the people on the other end, the researchers, the community are saying but
we get no access without coming into a data center.

For the subcommittee we have been worrying because we keep hearing from
groups that we go and see, we had a hearing with the American Indians, we’ve
had hearings with the Asian Pacific Islanders, and people say there’s all these
activities that are going on in the collection of data, I filled something out,
I talked to somebody, but I can’t get it. And even for us as data users for us
to get it for some of the federal statistics it would involve that we would
need to go to Rockville to be able to get access to it. So what’s happening is
it’s not being used and then Russell who as a part of his life as this other
background of reading these statutes and what have you, actually decided that
we needed to write a piece about these issues of privacy and confidentiality so
Russ, I’m going to kind of put you on the spot and ask you to share those with
the broader committee. Because it wasn’t that Jennifer disagreed, Jennifer
Madans at NCHS didn’t disagree with us around these things, she just wanted us
to kind of say them a little differently. So NCHS is concerned about some of
these issues. Russell.

MR. LOCALIO: Well, thank you for coming and I’ll try to make it very brief.
As I looked at the problem I saw this inherent, as everyone has seen, this
inherent conflict between the need to be able to ensure to respond and said
their information would be kept confidential, and making, giving access to the
information to people who need it. Now we have these different rules that
people operate under, for example if you want some information from an NCHS
survey and you want a patient, person level information, and you want place,
chances are you cannot get that. However, with a limited dataset you can get it
from a health care institution. And I think our talk with the National
Children’s Study, I think it’s called, National Children’s Study, just nine
days ago was that comes under the auspices of NIH, different set of statutes,
they haven’t even thought about this although they fully intend to issue data
at the respondent level soon after it’s collected.

Now to the best of my knowledge as of about two days ago, and this is
subsequent to our meeting last week, it’s my understanding that the data
processing for the National Children’s Study is going to be, perhaps is going
to be contracted out, so it would be a private contractor doing all the parts,
but anyway.

So here are some of the issues, by the way, just to make sure you
understand, we have here I believe a member of the PCDS, Privacy
Confidentiality and Data Sharing Workgroup of NCHS who has more experience then
I do on these issues. Are you still there, Marjorie? You still in PCDS?

MS. GREENBERG: Am I still what?

MR. LOCALIO: Are you still on PCDS?

MS. GREENBERG: Public Health Data Standards Consortium?

MR. LOCALIO: No, Privacy Confidentiality and Data Sharing Workgroup.



MS. GREENBERG: I don’t think there is such a workgroup at NCHS.

MR. LOCALIO: There’s not anymore? It’s listed on your website.

MS. GREENBERG: Let me see, I’m not on that, no, no, no, this is the Public
Health Data Standards Consortium, yes, there is a Privacy Confidentiality and
Data Sharing Workgroup of the Public Health Data Standards Consortium, which
used to have its website on the National, National Center for Health Statistics
hosted that website, it was not about NCHS data.

DR. MAYS: Are you on it?

MS. GREENBERG: I’m associated with that, yes, but I can’t —

MR. LOCALIO: There are organizations such as PCDS whose mission it is to
represent public health and health services research interests on privacy
issues including attendance keevies(?) and acting as a form to explore the
implications of the HIPAA privacy rule of public health and health services
research. So upon reading this I said the right hand ought to know what the
left hand is doing and everybody should be at least working, reading from the
same hymnal or reading from the same prayer book on this, but I get the feeling
we have inconsistent policies.

Now there are lots of new developments in this area, technical
developments, you heard my question to Chuck, and I asked are you releasing the
data at the person level and he said yes but it would be appropriately
disclosure limited I think is the term. In the last couple years there’s been a
lot of work on statistical disclosure limitation which actually I’ve only read
the technical papers on it, I don’t know what’s going on policy wise, but
there’s a lot of work in this area that may give rise to some other policies or
a series of policies that we may want to look into to ask given the big expense
there is in collecting all these data and the severe problems in collecting
data that are good and accurate how can we be most, how can we provide guidance
or recommendations that would make these data most available to the people who
need it, and the people who need it are two classes and I think Vickie touched
upon them.

We heard from a lot of people, when I was in California people said I’m a
member of a sub-sub group and we need data on Japanese American men who live in
Seattle, it’s not just race ethnicity, it’s place, and sometimes you can get
the race and ethnicity but you can’t get the place. We talked a little bit
about GEO coding which is the ultimate places that distinguishing between where
Gene is sitting and where I am sitting, to be able to say that’s where a person

Now when you factor in place it becomes extremely difficult on how to reach
these data in a way that is not overly burdensome but in a way that protects
against non-disclosure, protects against disclosure. Now excuse me, just a word
of clarification, we’re not talking about direct identifiers here, we’re
talking about identifiability, in other words the ability using other datasets
to be released and record linkage algorithms to break the law and do record
linkage and then identify people.

So I think that’s a thumbnail, is that okay as a thumbnail of some of the
issues that we have here.

DR. MAYS: Part of what we struggled with is that there is data that might be
there but because of the size of the data that’s collected we can’t do much in
the way, even if you have the ability to access the data it’s like as soon as
you try and cross it with age you can’t, as soon as you try and cross it with
geography, and geography is actually the biggest one, as soon as you try and
cross it with geography it’s like saying all of a sudden you can comment on the
Native American, the American Indian that’s in Colorado that has HIV, whoa,
it’s like you can’t.

And part of what is frustrating is that the policies will differ if we’re
talking about NCHS, if we’re talking about Census, if we’re talking about NIH,
so they all have the basic protection in place but what you can do differs. And
I think the extent to which there is a capacity to utilize the technology also
may differ, if you talk about, and I think Russell has read the technical
papers, I’ve read the less technical papers at some of the ASA in terms of
statistical needs, I mean this year I sat and sent for several of the papers,
I’ve kind of looked through them, and it’s like you look at NORK(?) and some of
the others, they’re actually like trying to use technology in order to be able
to see if they can release one more, like using it in terms of remotely, that’s
one of the things, or putting it up on like the California Health Interview
Survey they have an Ask CHIS thing where you can go and do cross tasks to get
to it, so there’s some ways in which I think there’s possibilities to get some
data but what we don’t know, and this is where I think Privacy can be helpful,
is what is really a regulation versus a policy and whether or not there can be
some changes to policies.

We all are very much in favor of regulation because it only takes violation
of one person that makes the news for all of us to have difficulty in terms of
accessing one day but what we’re not clear on is that we’re at the state of the
art relative to being able to do both at the same time, accessibility and
release of data and at the same time protection of individuals. Gene.

DR. LENGERICH: Thank you, I think you put me on your mailing list for your
institute newsletter —

MR. ROTHSTEIN: You’ll be getting your bill shortly.

DR. LENGERICH: Well, believe it or not I just did read the most recent one,
which had to do with urban health. Now one of the articles in there was one of
your colleagues writing about communicated based participatory research in
Louisville, so that’s, and that’s the perspective, that’s what I do in my day
job. Our purpose is to improve the health of specific communities by working
and engaging with them in participatory research. So I think what we have heard
and seen happening is there are national or state level surveys which collect
data, which have data about communities, but when the communities want to use
those data as baseline or as measurement to direct action plans, intervention
plans, they can’t get them because they’re not available because they don’t
know how to access them and where to go to them. So I think to me the purpose
is to, I mean there are data which are about a particular community, they need
to somehow be made available to that community and helps that direction target
interventions to help improve the health of that community.

DR. MAYS: So where I think we need your help is that we all want to do
what’s right in terms of privacy but we just don’t know if it needs to be what
it really is and whether it’s time to rethink a lot of these issues or not.

MR. ROTHSTEIN: Well, it seems what you’ve described is a huge project and
issue because the first step I would think is to try to figure out all the
different major data sources that, all the HHS agencies, Census, all the
sources of data, and figure out what their rules are on access. And then not
only what the rules are, whether they are legislative mandates, whether they
are regulations, or whether they’re policies or whether it’s somebody’s been
doing that for the last ten years. And then only at that point do you get to
the issue of is it possible at all to come up with some general principles of
balancing privacy and confidentiality interests against the interest of the
individuals and the researchers to get access to the information in some sort
of form that will be most valuable to them.

So that’s kind of the way I see that and sort of the purely privacy aspect
is kind of at the end because what, I don’t even have the faintest idea about
all the sources that you’re talking about or what their rules are or what the
implications of those rules are, whether they are turn out to be too
restrictive or not restrictive enough in terms of access to the information, so
the first step would be getting lots of information about what the current
status is and then possibly having a hearing and hearing from all the people
who are in charge of these various databanks and hearing from some privacy
people and maybe hearing from some representatives of your sub and sub
population, finding out what their interests are, and then trying to see if
it’s possible to make some recommendations about some general principles in how
we can make the rules or regulations comparable across the various agencies.

DR. MAYS: Marjorie, is it possible to ask ASPE to undertake a contract to do


DR. MAYS: Isn’t that what you all are called?

MS. GREENBERG: ASPE? Dale’s group?

DR. MAYS: Yeah, Dale and Jim’s group —

MS. GREENBERG: You could ask them to do that —

DR. MAYS: That’s what I’m saying, is that something that they could do?

MS. GREENBERG: They provide reimbursable funds to NCHS annually to support
the committee, and particularly to support research projects of the committee,
such as the, well the cost benefit study that was done on ICD-10, that was, or
the PMRI study, or the patient medical record information, and the
e-prescribing, so those have been supported by funds received from ASPE but the
contracts have actually been awarded either directly by NCHS or through our
logistical contractor, sometimes the Office of Data Policy, so I mean I’m just
thinking, so yeah, you could ask them, they may say one of three things, they
might say yeah, we’ll do that, they might say well that’s something, this is
something that the national committee is requesting that should be done through
NCHS with reimbursable funds we’ve already providers, or neither of the above.
But it does seem to me that a contract is needed, I don’t think you could get
this systematic information through a hearing.

DR. MAYS: We don’t want to touch it, we don’t want to touch it.

MR. ROTHSTEIN: I mean the hearing would be the second stage —

DR. MAYS: That would be what we would do here but —

MS. GREENBERG: I was listening to all this and saying to myself what you
need is a contracts committee. Now there have actually been some, I mean Jim
Scanlon is definitely the one to talk to, I mean there’s been some study of
this in the past but I don’t think, disclosure policies and all of that, but I
don’t think it’s anything real recent, definitely Jim Scanlon is the one to
talk to.

MR. LOCALIO: I think you may be referring to the study that was done in ’94

MS. GREENBERG: It’s about that long ago.

MR. LOCALIO: This was report on statistical disclosure and limitation
methodology for example —

MS. GREENBERG: Well, that’s methodology, I was thinking —

MR. LOCALIO: That was May ’94, then there was an Institute of Medicine
report in ’93 —

MS. GREENBERG: That’s the one I was thinking of, an Institute of Medicine

MR. LOCALIO: But there’s been a lot of development since then —

MS. GREENBERG: It’s been quite a while.

MR. LOCALIO: Including HIPAA, for example, HIPAA hadn’t been dreamed up yet.

MS. GREENBERG: Of course HIPAA really is clinical data —

MR. LOCALIO: That is true but we’ve been talking about data collected by
health care organizations and that might be a source of information and then
maybe that’s not the best source of information but the question is whatever
sources of information we have we ought to be able to figure out whether the
limitations that we’re going to have if it were collected. Here’s an example,
supposing we were to make a recommendation to do something very expensive for
the collection of data and race ethnicity, place, socioeconomic position. It
might be very good but it would sit in somebody’s computer someplace and never
be used, we don’t want to have that. So I think we’re now at the point where we
have so much discussion on getting this information from different sources
maybe it’s the time to revisit this.

DR. MAYS: But I think that clinical data should also be included because I
think there’s a very big issue here that’s not clear in the field, for example
when a health plan collects the data who are they responsible for sharing that
data with and how easy can you cross boundaries. The provider collects the
data, it’s given to the health plan, the health plan collects the data, do they
give it to the provider, I mean I think, again it would be like we could start
out with something and then it would be like there’s all this guidance that’s

MS. GREENBERG: But those are really different issues, I guess the same
contractor could look at both but —

DR. MAYS: About privacy and confidentiality?

MS. GREENBERG: Well, the issues related to sharing and disclosure of survey
data and how the policies differ and among agencies and the extent to which
those differences reflect different legislative requirements or just different
cultures or kind of history, I mean that’s a whole set of issues and I can see
getting a contractor to do that, to look at the major national datasets,
certainly those within the department but also the Census and some others and
categorize all that.

Now then you’re talking about this whole issue of health plans and providers
and all of that, I mean that really is a different set of issues, the actors, I
mean the players are different, the controlling the regulations are different,
etc. —

DR. VIGILANTE: But on National Children’s Study you’re collecting clinical
data. I mean it’s clinical and you’re going to be collecting visit data for
routine visits and emergency department data or for 20, 30 years for these
kids, I mean ultimately they want to be getting it out of electronic health
records in a way that doesn’t require manual data collection.

MS. GREENBERG: I mean that’s still in the context of a survey —

DR. VIGILANTE: Or a study, I mean not a survey because it’s done passively

MS. GREENBERG: I think thought Vickie was talking about not research and not
surveys but what goes on just routinely —

DR. VIGILANTE: I thought we started with the National Children’s Study,
isn’t that how we started this conversation?

DR. MAYS: Well, it didn’t start it but it got piled on. It started with the
surveys and then they came in and they need help and it’s like much the same
issue and then I think as we were talking here my sense is that if we can also,
I mean again, we’re about to like try and push health plans to do something
where we don’t know, I mean we have no advice to give them about how easy it is
to go across these barriers and there are barriers between what’s collected in
a clinical setting and belongs to the clinical setting and what’s collected by
a health plan and what the health plan can share, I mean it would make a
difference I think to some extent if we find out that these are not easy. I
mean the Secretary can be useful I would think in trying to make some
recommendations about how to do this, both at the level of the industry and as
well as at the federal level —

MS. GREENBERG: I mean it’s, Mark, the Privacy’s, is the Privacy Subcommittee
meeting in the morning?


MS. GREENBERG: I mean can you get kind of a pulse of how important they
think this is?

MR. ROTHSTEIN: I can tell you that we’ve got seven other things on our list
and I can’t imagine that this issue in its current format, I mean given the
data limitations, that you can’t hand me a list that shows some sort of a
chart, I can’t imagine that it would be moving up the charts. I mean we have
already on our list things like, we’ve got hearings scheduled two weeks from
today on e-prescribing and security, then we’ve got two days of hearings on
patient controlled limits on the contents of their own electronic health
records, we’ve got limits on disclosure of protected health information of
third parties by compelled authorizations, we have a hearing on decedent and
archival health information, we’ve got, I want to take up the issues of radio
frequency ID chips —

MS. GREENBERG: Well, I guess the reason I ask this is because this would, I
don’t know how expensive this would be but it wouldn’t be cheap I mean because
collecting information, and particularly as you’ve expanded it, so we don’t, we
have a modest amount of funding for contracts —

MR. ROTHSTEIN: Maybe I answered the question incorrectly —

MS. GREENBERG: If this contract were undertaken —

MR. ROTHSTEIN: If the contract were undertaken and done then I think there
would —

[Group discussion.]

MS. GREENBERG: If this contract were done would the Privacy Committee then
feel this was something they were interested in pursuing?

MR. ROTHSTEIN: It’s premature now but if we had data I would want to pursue

DR. MAYS: What I think is that at this point it would be if Mark could act
as a consultant in terms of a charge that would go, or a scope of work that
would go for example to a contractor, so that we could make sure that things
that you thought you would need to know, that that’s being asked. I mean we
have a sense of what we think but we also would want to make sure that since
we’re getting a contractor to do this, if we get permission to do this, that
the right questions are being asked. And that the scope of work, see I thought
you were asking a different question that had to do with whether or not he
thought now going into this other area, which is the more clinical side —

MS. GREENBERG: No, no, I was just asking whether in fact if such a contract
were done the Privacy Committee would consider this something they wanted to

MR. ROTHSTEIN: I would, I’ve learned not to speak for the subcommittee, I’m
often the one dissenting vote on the subcommittee, but I think it’s a very
interesting issue.

DR. MAYS: And see I think the thing to determine is whether or not the
talents to do the two are the same or whether or not it’s broken into two
contracts. I mean I think if that’s what you’re concerned about well it’s a
whole different set of plays and maybe a different set of people, then I think
that’s something we ask actually I think the Quality group might kind of kick
around and let us know what you think about that.

MS. GREENBERG: It could well be done in the same thing, I mean I thought
they were kind of differently focused but I think the thing to do is if you
feel that this is something you really want to see done is to bring it up
tomorrow and you even make a recommendation to the department that it conducts
such an analysis.

DR. MAYS: We think that this is critical to our integrity, every time we go
out it’s the same questions that are coming up by small groups and we do need
to make I think some genuine effort to try and do something about it, or to be
able to offer insight as to why we can’t, and right now we can’t do that so I
think it’s important.

DR. LENGERICH: I think this is timely from the GEO coding perspective
because the contract that Dale reported on as being finished now was mandated
by all, mandated from the Department of Interior I think that all datasets
maintained by the federal government with geographic information be coordinated
through the geospatial data warehouse, so that contract that Dale is just
finishing has already recorded the geographic detail for all of the national,
all of the health surveys that are done, so we’ve already got a listing of what
they are in the data that are there, so that could go forward as trying to
articulate what are the problems, what are the regulations that regulate each
of those datasets.

MR. LOCALIO: I also just want to say the e-prescribing is very much related
to this because the enormous amount of work that is done using prescription
drug data and to the extent that that would have information on race ethnicity,
place, socioeconomic position, the same issue is going to come up. But I know
you’re not talking about that issue, I just want to say that those things come

I have to agree with you, when you GEO code something you can identify
people very closely and what are the issues related to that, so I’m a little
surprising that in GEO coding this hasn’t come up very —

DR. LENGERICH: Well, it comes up in the context of problems all the time,
right, and they just phased the contract to list all those datasets and what
variables are in there so we can build on that.

MR. ROTHSTEIN: Vickie, may I suggest an issue that I think, not to be a bad
guest, but I’d like to suggest an issue that you may want to think about
jointly working with us on and it has to do with research and under both the
HIPAA privacy rule and the common rule it’s possible for researchers to get
access to records and samples of individuals that are not keyed by name but do
have race and ethnicity information without any authorized or informed consent
requirement. And if you’re doing for example behavioral genetic research or
drug abuse and so forth you could, the researchers could make findings that
have great impact on particular subpopulations and it’s a widespread concern
and in fact there’s a lawsuit now that’s pending you may be familiar with by
the Havasupi(?) Indians against Arizona State University for using their
samples, even though there were not a lot of sample, in ways that they didn’t
want, and I think it’s an important sort of cross section between privacy and
populations but your plate may already be overflowing but I just wanted to put
that on your list.

DR. MAYS: Maybe what we can do is, at another point what I can do is at
least bring up, Edna, do you know about this? I mean I know a little bit about

MS. PAISANO: Yes, I had sent some information I thought would be forwarded
to the subcommittee, this was several months ago and it was sort of like an
incident when a tribal member had been somewhere and they were talking about
different health characteristics and it was like they had drawn blood for
something and then the researchers had used it for two or three other different
purposes without the individual tribal people agreeing to that.

MR. ROTHSTEIN: It was supposed to be a study on diabetes and then it went
into other things that were very controversial and not authorized by anyone but
it’s anonymous —

DR. MAYS: It had to do with the samples being anonymous but what happens is
it’s a vulnerable population so it should have probably come back again in
terms of for vulnerable population but as I understand it one also twist to
these things is that particularly if you are a researcher who is in that
particular center or place or that the data has been collected that under HIPAA
you have access to it, that normally if I just wanted to come and do a study I
wouldn’t but because I’m part of the hospital staff, even though I’m a research
I’m part of the hospital staff —

MR. ROTHSTEIN: That’s only for reviews preparatory to research, so if you’re
a member of the hospital staff you have access to the files to see if there are
a enough cases of X to do a study on. The problem as our committee saw it was
under HIPAA at that point if you’re still a member of the staff you can
actually approach the people and try to recruit them and to sign authorization
to be in the study even though it’s never gone to an IRB —

DR. MAYS: Because you’re part of the staff, yeah, see that was the part I
was trying to remember is that because you’re part of the staff it’s an
interesting, again, it’s like this was probably done to facilitate things but
on the other hand it’s clear that there is just something, a bit of a gulf
between the common rule and HIPAA here that does need to be plugged up in some

I think what I was going to say is that I think that that’s something that
we may need to discuss and to see where we are with this, we do have a full
plate but again these are some important issues that we need to think about
having some time to prioritize whether we can address them.

Anything else before Mark goes? I think that this has been very useful,
Mark, and even though I think now is not the time for us to pursue it with you
I think that if you will act in a consultation role what we will do is to see
if can’t get the information because Russell’s been trying to do it here or
there but I think you’re right, that we really need a very thorough task,
thorough assignment of a task to someone to actually collect this and then to
do an examination of it and give us that and then from there for us to decide
what’s the next steps and the next steps might be calling people then to talk
about it. So I think that’s probably where we need to head.

MR. ROTHSTEIN: Thank you for letting me —

DR. MAYS: Thank you very much.

Anything else that you want to bring up before we adjourn? I think that we
quickly went through our entire agenda —

DR. BREEN: Can I just revisit that one item related to our report?

DR. MAYS: Oh, yes, I’m sorry, that was, yes.

DR. BREEN: The report, I wasn’t completely, I didn’t quite fully understand
that the report was just covering the first hearing and I’ve been thinking
about the report and I was reading the latest version and Anne’s outline and I
think it’s coming along really nicely and I’m pretty excited about it and I’m
especially excited about it as a federal staff member who can take reports like
this and actually use them and say oh, we need money for X because this report
said there’s a gaping hole and we’re in a position to fill that gaping hole and
so therefore I propose that we do the following RFA or contract data collection
or something, or have a workshop, whatever. So that’s the way we can use those
things in the government.

So I would really like to see a document that will provide mandates or, I
mean the best mandates actually come from Congress but mandates from places
like a subcommittee like this don’t hurt because it’s an expert subcommittee,
it’s convened by NCHS, it’s a very legitimate committee, convened by the
government for the purpose of providing expertise to us. So I’d really like to
see the report be able to do that and so that kind of made me wonder if we
should revisit the business about just focusing on the first hearing because I
think we’re going to be doing a report on the racial ethnic populations but I’m
wondering if we’re going to be talking at all about the state hearings because
those data are actually where we want to expand and that was the state
governments were very interested in having federal guidance and federal funding
and federal standardization and there were a lot of things they were requesting
from the federal government that we don’t normally do because we don’t want to
be butting in in places where we’re not wanted.

So I think it’s quite important to make those things clear in the report and
so I’m hoping that we can include those parts of the hearing somehow in that
report as well, or if you have other plans for it but short, pithy, focused,
all those things help us get things done.

DR. MAYS: Another thing I can say is that in the state hearing it’s almost
like apples and oranges in the sense that what they want is different then what
the people who run the surveys want and that’s the tension that’s there.

DR. BREEN: I didn’t see it that way.

MS. GREENBERG: I don’t know, I was not at the state hearing but I guess I
was surprised too, we had to go to the NHII, I was surprised myself that you
were really only focusing on that first hearing. Were you planning other
reports on the other hearings?

DR. MAYS: What we had talked about before, I mean this but this has kind of
been one of these traumatic experiences, what we talked about before was we’d
even talked about having a series of reports and that if you remember when we
first went, that this first report was supposed to have kind of the structure
to it that allowed you to then kind of know that there would be other reports
coming and that you could use the information from other reports but that it
didn’t, it wasn’t like it was just all folded in. See I think that the other
hearings kind of help us to understand and focus on one thing which is the
population surveys and then I think when you go to the other hearings that they
have a different set of issues. I’m not sure at this stage that we can like
just make a big turn and pull in all the information from those hearings, it
means reviewing the transcripts, it means a lot —

MS. GREENBERG: We have a contract now I think actually —

DR. MAYS: Yeah, but I mean it just means a different set of —

MS. GREENBERG: If it isn’t part of what the contract is I don’t think we
should pull it in —

DR. BREEN: Is there a plan then to do something on the state hearings?
Because I mean we’re moving so slowly on this first one, I’m not optimistic
that we’re going to do four reports.

DR. MAYS: Believe me I wasn’t optimistic either with what it took to get
this one out, I think what needs to be expressed is that if you think this
needs to be done then we have to be a lot clearer about what we suffer from is
resources, and I’m talking about just basic level staff resources here and
that’s what’s needed to move these things along, I mean I’m more then willing
to try and do a lion’s share of them but when I’m doing nothing but clerical
work on moving it along in the sense of writing contracts and —

DR. BREEN: Well we need to have this discussion when Marjorie is here, she’s
the one in the position to improve that —

DR. MAYS: I mean that’s a little part —

MS. JACKSON: Along with directions because we’ve had various writers over
the years — [off microphone] — certain directions of what needed to get done,
so pulling all the information together as everyone knows at the same time —

DR. MAYS: But I mean I think the other hearings are if anything are less
complicated, I think this is one of these politically complicated reports very
honestly when we site what it’s about, I think the other reports will be less

Okay, we stand adjourned.

[Whereupon at 4:58 p.m. the meeting was adjourned.]