Transcript of the October 26, 2004 NCVHS Subcommittee on Populations Meeting

We’re going to get started, those of you who have the expanded agenda,
I don’t know if there are people that are listening through the internet
but we’ve changed the agenda that’s on the internet. The first thing
we’re going to do is actually talk about the workgroup process, where we
want to go, prioritize activities and clarify some of the staff assignments.
The start here really has to do with kind of listening to you the members and
knowing that we’ve had new staff join and feel as if we really need to
spend a little time checking in to talk about things like what topics staff are
assigned to because we’re going beyond just the lead staff here,
we’re talking about the other staff. Brenda, for example, you’re
doing yeoman’s job, you’re online, but we also want to make sure to
get you involved, we don’t want to take a year to do that.

So we’re going to talk a little bit today about what projects we have
going on and then talk to people about their involvement in the various
projects because I think we will do well in the sense of having a couple of
staff for each project, the staff being able to work together as a group,
instead of if one staff has something to do there’s no one and then
they’re kind of pressured to play catch up. And we don’t need to do
that because I think we have several new people but I wanted to take the
opportunity to get them involved and find out a little bit about what are these
priorities that we’re going to talk about that they’re interested in.

Now one of the ways to figure out some of our priorities or the actual
activities that I’ve put on for us to discuss today, because part of our
goal today is to try and jump start several of these activities. Mental health
is one that we have worked a bit on, the question in mention health really has
to do with what is the portfolio that we would like to see the federal
government have in its collection of health statistics on mental health. If we
say the federal government in general there’s a lot of things it collects
but the question really is what ongoing statistics do we want to see.

And then I think the other is what, in what way might this committee be able
to influence or to take leadership in helping to pull health and mental health
together. In many ways we have statistics that are being collected but what
doesn’t happen is that those statistics are not necessarily being brought
together to have a full impact on not just physical health but also how the
well being and mental disorders interact with some of the physical health. So
that’s one of our topics.

Privacy and Confidentiality, we’re preparing for a meeting with them
because this committee has had a long history in its work on racial and ethnic
minority data collection of being concerned about how do we get that data more
accessible, what are some of the issues in terms of when you have small
populations and you collect the data, how can you GEO code it, so we thought
our colleagues that are in the Subcommittee on Privacy and Confidentiality
could be helpful.

Quality, we’ve reached a point where for a lot of my time and probably
Gene’s time Quality has been a workgroup that was generating quite a lot
of activity kind of separately but now that it’s come in with a set of
recommendations what we’re finding is that for Quality there’s a lot
of overlap so we want to work with that particular group.

The National Child Study is something that we’ve been concerned about,
it’s the potential to mount, let me get my numbers straight, a 100,000
children who will begin in utero through age 21, and again, what this committee
is trying to do with that study is to ensure that many of the things that we
have learned through our hearings can be interested into the study at the start
as opposed to at a later point in time to ensure that we will have
participation and data for racial and ethnic minorities.

So those are major things on our agenda, Gene, is there anything else that
I’m overlooking that has slipped through the cracks that we want to do?
The reason I’m asking, it’s not because you’re just the only one
in the room but the reason I’m asking has a lot to do with where we were
going to go relative to some of the GEO coding stuff and whether we wanted to
try and do that separately or do that as a piece of the privacy and
confidentiality.

DR. LENGERICH: I think that it’s a part of the privacy and
confidentiality, it’s a special part of it in that it is data about
probably small groups or at least geographically distinct groups, individuals,
and so I think the issue becomes, I’m under the assumption that the data
can be collected and I presume that that is the case, then it’s a matter
of how is that data made available to and about those individuals from which
it’s collected, which really borders into the privacy and confidentiality
issues. So I think we’ve heard recommendations from some of our hearings
that compared to the small groups, compared to the entire population, do want
that data and so the question is how do you make that available to them without
violating the confidentiality that they gave to provide those data. So we had
heard recommendation, we heard the recommendation about go ahead and collect
the data, we presume that that’s going to be done. And then the question
is how do you make that data available to people, we heard about
recommendations for training centers or some sort of community or local access
data centers as well.

So I think that we have heard, and we recognize some of the issues around
that about various datasets that can’t reside on the same server, that can
reside and how that all goes about. So I think that there are some, there are a
number of methodologic system issues that need to be addressed. So I think it
can be housed within the privacy and confidentiality but it goes back to me to
making the bigger issue of how do you make data about small groups or
individuals available and useful for knowledge about the health of that
population.

DR. MAYS: Thank you because I think those, fleshing that out is really
helpful because it’s about, we did, we talked about the data center once
before and so this, I think we just need to be clear about the range of things.
But that’s good, I think that we agree then on a good way to kind of house
this.

Part of what I think we want to do is to try and prioritize where we want to
go, and I think the question becomes where can we make an impact and kind of
the order in our ability to be able to make that impact. So for example in the
Quality, candidate recommendations for Quality, I think that that probably
should definitely be prioritized because one, we want to stay in step with the
workgroup but second, the work is done and we just have to figure out, I mean
they’ve come forth with these candidate recommendations. What we need to
do when we have the discussion about that particular topic is to figure out
with them which of the recommendations do we need to have hearings and which do
we have enough information from our previous hearings that we could do is
assist them in jointly writing a letter and putting something forth.

In the Quality, in the instance of Quality I think that what we want to
think about is what would be the specific changes that we think if they were to
come about would make a difference in terms of the quality of the delivery of
health services, health care, and health status of vulnerable populations. And
this is where I think hearing from our colleagues a little bit of the work
that’s taken place in the health plans with the collection of the data on
race and ethnicity and where that’s likely to go. There were some
excellent comments made at the September 14^th hearing and then you
have the transcript for the other hearings, so I think it’s interesting
we’re at a point, I mean it just seems like we arrived here a little bit
longer then it takes some of the other committees but I think we’re at a
point where actually working with changes to some of the HIPAA as well as some
of the health plans should result I think in potential change that would make a
difference.

Anything else in terms of quality that is a priority? What we have to
realize is when we come to the table with Quality, Quality is bringing us some
candidate recommendations but in terms of the hearings that we’ve had this
is a good time for us, if there’s anything that we think Quality should
pursue that we have heard in our hearings this would be a good time to also ask
them to pursue those agendas.

One that I have down here that came up on the call, and again I’m going
to ask the Quality people to talk a little bit about it, was when Ken Kizer(?)
was talking he brought up the issue of functional status and not having
information on functional status. And so that’s why I put that back on to
our agenda so like we got involved in functional status several years ago and I
think that really occupied this committee’s workload for quite some time
but we need to remember when we have that expertise to kind of be on the
lookout for points at which to raise it relative to other things. So that was
the only other one that I could think of but I guess we really should hear from
Quality so I would say that that’s probably our first priority in the
sense of our ability to move some agendas forth that where people are waiting
to receive them.

The other that we should talk a bit about is probably what we want to do
with the Privacy Subcommittee because we’re almost close to that
particularly agenda item anyway. And I think in that, do you have, I passed
out, you passed folders out? Russell, good morning.

MS. EVELYN: Vickie, I’m sorry to interrupt, this is Brenda, you keep
fading in and out so I don’t know what the difficulty might be.

DR. MAYS: Let me know when it happens and that will help Don when he’s
monitoring the technology to figure out what’s happening because we think
everything’s working here but it may be the transmission line.

Russell, good morning.

MR. LOCALIO: Good morning.

DR. MAYS: Welcome. Can you introduce yourself and then I’m going —

MR. LOCALIO: I’m Russell Localio from University of Pennsylvania
School of Medicine, I am a member of the subcommittee.

DR. MAYS: Okay, Brenda, before we move on I just wanted to ask you in terms
of for FDA are there particular issues that you think we should be looking at
that we haven’t, are there any particular topic areas that we’ve
talked about that are of concern to your agency?

MS. EVELYN: Of the ones that are listed here we do so incredibly little
with data collection on any kind of populations because basically what we are
looking at is information that supports safety and effectiveness of drug
products or medical devices. So we don’t do anything with respect to
collecting data, not even on usage of the drug products. We do things with
adverse events but we have no way to track it by population type or race or
ethnicity or anything like that. Nor do we track anything with respect to
populations within the clinical trials that are submitted to support the safety
and effectiveness of the application for the products that they’re seeking
approval for. So right now I’m just sort of trying to listen because
I’m having a difficult time trying to find where we plug in and fit into
this.

DR. MAYS: Well let me raise some questions. I guess I thought that for FDA
there was now the push, I know in terms of women there was a push that when,
and I’m not going to use the right terminology here but anyway, when drugs
are submitted for approval that at the approval stage they were now requesting
information on gender and that also you were being requested to bring in
population specific information. Is that correct?

MS. EVELYN: We ask that sponsors tabulate the participants by age, race and
gender and we get it some of the time but we don’t really track and we
don’t have any specific requirements as to percentages of population
groups that need to be in clinical trials for any type of products. So as long
as the sponsor tells us that they included women and they’ve included a
mix of racial groups then we really don’t track the numbers per se, we
basically sort of just look at it to see whether or not people are included, if
they’re not included we may ask the sponsor to include a broader mix of
people but we don’t keep statistics on exactly who’s in the trials
and who aren’t.

MR. LOCALIO: If the FDA does not keep data on this who does?

MS. EVELYN: That’s an excellent question, we as I said look at the
applications to make sure that people are in there but we don’t have any
type of database, as far as I know the sponsors probably themselves know but we
don’t have the capacity at this point to keep that in any kind of database
that is retrievable and that we can look at it. Our office actually has been
taking samples along the way of say IND annual reports, we might take like a
five year sample and look at some of those that are submitted and see to what
extent people are reporting but it’s not a very sophisticated way of
tracking the entire drug industry or even what’s been submitted to us in
drugs or devices. So to answer your question who does, I don’t think
anybody in the federal government does at this point but that’s sort of
where we are right now. And it’s not been an easy topic to get through
this agency that that’s something that we really need to do.

MR. LOCALIO: Let me ask a follow-up question then, you said it’s not
going to be easy to get it through the agency, what’s required to get it
through the agency? Would you have to go through formal rulemaking or would
this be done with policy or guidance?

MS. EVELYN: Actually it probably would have to go through formal rulemaking
because we did issue a draft guidance trying to outline to sponsors how we
would like to see the data collected basically using the OMB recommendations of
race reporting, but there tends to be difficulty with people doing that because
so many of the clinical trials are held, are conducted in Europe or other
places that don’t conform to our way of reporting. So a lot of the
difficulty has been trying to merge or trying to get some consistency between
what other countries do as far as reporting race and ethnicity and the
categories that we use, that’s part of the difficulty. And so in order to
get it where people would report and we would be able to keep the statistics in
a certain way I think it would call for rulemaking because we have guidance
documents, draft guidance documents already but guidance documents don’t
hold the same force of law as regulations and in order to get a regulation to
do that we would have to go through the process of rulemaking.

DR. MAYS: I think what we might want to do as a subcommittee and I just
want to see how the two of you think is to talk with Standards and Security
because they do patient safety, there’s a lot of things where —

MS. EVELYN: Okay, I lost you again.

DR. MAYS: I turned my head, now we know what the problem is, I’m
turning my head. What we might want to do is to think about whether to bring
this up as a subcommittee with Standards, I think in Standards they’ve
often talked about some of the patient safety, they might have, it might be a
good mix for them and us for us to have a conversation with them and see to
some extent if this fits with them. What do you think?

MS. EVELYN: That sounds okay to me, would you like us to prepare maybe a
presentation to talk about what some of the issues are with our data
collection?

DR. MAYS: Let us talk to the other subcommittee and then lets figure out to
do this. If for example we wanted this at the November 4^th breakout
could you do it then? Or is that too close? That’s too close, I just
realized what the date is, that’s too close —

MS. EVELYN: — from the agency who have a little bit more direct expertise
with this then I do.

DR. MAYS: Okay, we’re going to put this in the hopper of one of the
things that we want to look at so we will work with you on when is a good time
to do this.

MS. EVELYN: Okay, and if you could give me some specifics about exactly
what you’d like to have raised beyond what I’ve talked about today
would be helpful as well.

DR. MAYS: What we’ll do is I think we’ll have a discussion with
Standards and see where we are and see if this is something that we want to
pursue a little further, rulemaking is pretty hard but at least some clarity as
to what these issues are and ways that we might be able to make a difference is
something I think we’ll have to think about.

DR. LENGERICH: This is Gene Lengerich and I guess I’m wondering that
if this is going to come up for an information session, a briefing from FDA to
us, I wonder if it would also be appropriate to include informed consent in
that discussion as well because if many of these trials that they’re
talking about are being conducted overseas in international populations
I’m just wondering about the informed consent process if that, how that is
handled as well.

DR. MAYS: Tell, me, I’m not sure what the informed consent in terms of
when the trials are conducted over there?

DR. LENGERICH: Yes, so what are the rules and expectations of informed
consent for trials overseas from FDA.

MS. EVELYN: If those trials are conducted under an FDA issued
investigational new drug application, which means it’s going to be
marketed in this country, then the same rules apply to those sponsors there
that would apply here. So all of those regulations under 21 CFR Part 50 and 56
would apply with respect to informed consent. Now if the trial is not being
conducted under an FDA issued IND and they’re just doing it in their own
country for their own use then I don’t know what regulations would apply
in that case. But they would be consistent if they’re applied an IND. And
we can talk some more about that if you want that included as well.

MR. LOCALIO: Let me ask another follow-up question here. I guess Pfizer has
recently announced that it wants to do a new study on I think Celebrex and
enroll 4,000 people. Now I’m not sure whether that’s for safety or
whether for a change in labeling but that’s a lot of people and would the
FDA have any idea as to the population characteristics of those 4,000
participants or not?

MS. EVELYN: We would not know the population characteristics of those
participants unless the sponsor submitted detailed information on those
participants in their IND annual report. What we have found over the years sort
of looking at that is that sometimes they tell us and sometimes they don’t
and dependent on what the indication is and what meetings that they have with
our Center for Drugs the agency might tell the sponsor we don’t, tell us
about your racial and ethnic mix of these people. And the sponsors usually have
it, sometimes like I said they’ll include it in the application or not but
if the agency sees that they’re using, let’s say it’s not
Celebrex, let’s say it’s something for diabetes for instance and a
sponsor comes in with a large group of people in diabetes and none of those
people or very few of them appear to be African American or of Hispanic
descent. Then the agency might say well you need to include more of those
people because the disease is very prevalent in those groups. But as a rule we
sometimes get, that’s the best answer I can give you, we sometimes get it,
we sometimes don’t, at first we thought it was because we didn’t tell
people exactly what categories to use to report the information and so
that’s been out there for a couple of years now even though that draft
does not finalize it, at least its been out there and talked about and still
sometimes we get and sometimes we don’t.

DR. MAYS: Okay, I think this has been helpful, we should I think a
discussion about this at the November 4^th breakout.

MS. EVELYN: I didn’t hear you —

DR MAYS: The subcommittee would meet in a breakout on November
4^th so I’m just starting to keep a list of potential issues for
us to discuss on the November 4^th breakout and so I’m going to
put this down as one possibility for discussion at that meeting. It will just
be a discussion about raising other people’s consciousness about this
issue and adding it as a potential area of interest for us.

Thank you very much for that, Brenda, that was very useful.

Let’s move to talking about our meeting with Privacy, the Privacy
Subcommittee.

Agenda Item: Collaboration with the Privacy
Subcommittee – Dr. Mays, Dr. Lengerich, and Mr. Localio

DR. MAYS: One of the things I’ve distributed, Russell, I distributed
in the packets the piece that you wrote on privacy, do you have it, Gene?

MR. LOCALIO: Let me qualify what you call the piece, the piece is a rough
draft of some issues put into some, just put into words, to elicit comment,
criticism, and additions and subtractions, it is another words a place to
start.

DR. MAYS: Yes, okay.

MR. LOCALIO: And nothing more.

DR. MAYS: Alright —

DR. LENGERICH: Vickie, excuse me, I don’t think I have a document, I
had seen it in the email, okay, that was what you just distributed.

MR. LOCALIO: By the way as of yesterday I checked around to see if there
were some new issues or new developments since August and I couldn’t find
anything and the person I asked who knows a lot about this issues had not heard
of anything so this is reasonably up to date, there have not been any new
developments that I know in terms of just the issues.

DR. MAYS: Here’s what I think would be helpful is lets flesh out what
the issues are so we have a sense of what it is we’re going to raise with
Privacy, what some of the suggestions are, and is Jennifer Madans online? Okay.
Jennifer was going to try and join us but in her absence we’ll talk about
the email that I got and what she’s talked about relative to these issues.

So Russell let’s start with you in terms of really outlining what are
some of the issues for the Subcommittee on Populations around privacy and
confidentiality that applies to racial and ethnic minority populations, and to
population health I should say in general.

DR. LOCALIO: Well, okay, the overriding issue as I see it is very simply
this, we have talked a great deal about collecting data on race and ethnicity,
not only on race and ethnicity across the country but race and ethnicity by
region and by sub-races and sub-ethnicities, so the problem is what happens if
we collect the information in that detail, it may include GEO coding for
example, but because of existing statutes, rules and policies on the privacy of
that information the use of it is very limited, so then the question comes well
why are we going to collect the information if people can’t use it.
That’s the overriding issue.

Now in terms of access to the data we have discussed this several times
with NCHS people and there have been some traditional policies to make the data
available principally opening a data center in Hyattsville where people can
come to Hyattsville and use the data in a specific protected location using
specific analytic tools and then being able to leave with a certain number of
specified analyses. There have been some developments, some technical
developments recently that may offset some of the problem but what struck me
initially as being somewhat inconsistent about the policies is that I can go
into a hospital and provided I have IRB approval get a limited dataset, which
has an enormous amount of detail on people, their diagnoses, their procedures,
their dates of birth, their ages, and their zip codes, as part of a limited
dataset, and I could get that with an enormous number of people across the
country and I could get that with other protected information that’s
covered by HIPAA but if it comes from a survey that is done by NCHS for example
I can’t get any of that detail. So it struck me as being inconsistent and
why was it inconsistent, should it be inconsistent, and what do those
inconsistencies do in terms of restricting access to information.

By the way I haven’t seen this before. This is an email from Jennifer
Madans, September 13^th, oh I agree the issues raised by the paper
are not straightforward and would need discussion but that’s why
we’re here.

So the purpose of having the Privacy Subcommittee join us is to say okay
this is the situation, I would hope somebody from NCHS would be available to
us, is to say this is the situation, why is it like this other then being an
artifact, an historical artifact of well, these rules grew out of this statutes
and these rules grew out of this rulemaking procedure and the fact that
they’re inconsistent we can’t do anything about, that’s not
terribly helpful.

DR. MAYS: Gene, in terms of GEO coding are there specific barriers that you
can think of that have to do with privacy and confidentiality that we should
raise with them?

DR. LENGERICH: I think GEO coding is a method by which the data become
identifiable and so it goes back to I think the issue Russ is raising, that
that one variable, particularly if it’s very specific, is sufficient in
itself to identify the person and even if it’s not totally specific in
combination with other ones may make the person identifiable. So I guess, I
think there is great move afoot in the federal government to enhance the
ability to analyze data to place, we know from studies that place is important
in health and health outcome so it is a part of factors that are important,
possibly modifiable, in improving health. So Vickie, I think it comes back to
being it is one of the very specific variables, it is a variable that can add a
great deal of identifiability to the data and yet we see great effort from the
federal government, the Federal Data Geospatial Data Clearinghouse effort for
example is attempting to bring great consistency and greater use of geographic
identifiers. So in a sense there’s another sort of tension exhibited here
between greater specificity, use of geospatial data, but also ensuring
confidentiality.

And as I said before we heard from, when we were conducting our hearings,
that the groups that we’re talking about want to be able to use these data
so I think that we’re reflecting some of the community, some of the
population’s desires and to me it’s a question of what is preventing
it and how to make that, what are the barriers and if there’s anyway to
overcome those barriers.

DR. MAYS: I think a strategy might be when we’re talking to Privacy is
emphasizing this issue of how important place is to health and health outcomes
and it’s almost like that should be a guiding force, if there is enough
research data and we’ve heard it in testimony about the importance of
place relative to health and health outcomes and that should be driving some of
our efforts I think relative to trying to ensure that we have place and that we
have place particularly for these subpopulations. And then to try to figure out
are there needed changes to policies, procedures, regulation, etc., that would
facilitate greater knowledge about place. Because when we first started our
hearings something that we were trying to ask the question about is what
variables do we need to know other then just race and maybe we haven’t
made the case strong enough about the concept, the importance of place, and
that that’s part of what our need is to make that clearer and then what we
need Privacy to do is to help us figure out ways in which policies and
procedures will look each time at themselves and ensure that place can be
included. Maybe that’s one of the strategies to talk with Privacy about.

DR. LENGERICH: I think it may be not at the forefront of our thinking about
that place is important and so would not, it would benefit by reemphasizing
that to the group.

DR. MAYS: See that’s what I’m said, it’s interesting that
when you said it it was very eloquent about we know this but we don’t say
it like that and I think it’s a simple but powerful statement. We have
data to back it up and we have testimony to back it up but we just haven’t
I think been articulate about that and I think that that’s a way, when you
talk about a business case I think there’s one that’s made for that
but it’s just a matter of actually putting it out there in terms of a
business case for the side of cost and all that, I don’t know what that
would mean but I do think that that should be part of our discussion about
this.

Let’s talk about these issues about accessibility because there’s
lots, I mean technology is getting better and better and there’s lots of
different ways to think about and I want to generate some thinking about ways
in which to increase that because it’s not always that we may need a data
center. Like for example there are now these query systems that will allow you,
we have one on the California Health Interview Survey, CDC has them, where you
have the dataset and I want to know something about American Indians by age, so
you put those two things in and then out pops a table.

It may be that for example some availability of this data, it might be
pushing for greater query systems like that even so that at the level of the
community simple things like that really assist them in writing proposals,
simple things like that, like for example we have a system at UCLA called NKLA
which is Neighborhood Knowledge in LA and what it is is that you put a zip code
in and then up comes what the density is by racial ethnic groups, what the
income, I have my students working on a project of that and they’ve just
been realizing how much more they’ve learned and there’s a lot of
simple information but it allows them to be able to pursue other things by
knowing something about the number of racial and ethnic minorities in their
community, knowing the income people have, so there’s just a variety and
it all comes from the Census. But it may be to think about these query systems
a bit more in terms of some of the datasets that we have.

Does this sound like something that we may also want to think about?

MR. LOCALIO: I guess my current concern is not with our trying to come up
with recommended solutions, I think my current concern is how do we start a
process going whereby we can discuss, we can have people discuss what solutions
might be and then make recommendations on those discussions. For example we
currently don’t know, at least I currently don’t know, exactly what
is the spectrum of solutions that NCHS for example has, what are the technical
solutions? What is the availability of software? What are the terms and
conditions for using that software?

DR. MAYS: This is for data accessibility —

MR. LOCALIO: This is for data accessibility, we don’t know a lot of
things, at least I don’t know them, we haven’t discussed them, all in
the context of what are the implications for looking at variation in race,
ethnicity, and place, we don’t have that, I mean we don’t have that.
The reason I like the idea of place is that there’s a recurring issue that
comes up that I face at least once a week on other projects that I’m
working on and that has to do with this, when you look at, when somebody comes
to me and says well what are the differences by race and ethnicity and I say
what question are you asking, are you asking whether overall there are
differences by race and ethnicity in the country or whether there are
differences by race and ethnicity given that you happen to be in New Mexico.

And the reason I use New Mexico is on this July 2^nd report called
Broken Promises, Evaluating the Native American Health Care System,
there’s a wonderful map on page 48 that indicates all the native,
it’s called Trends in Indian Health and it has a map of all of the
populations across the country that are served by various centers. Well
there’s huge geographic variation among these things so it might be there
are substantial differences in the Billings, Montana population and not nearly
so many in Nashville, in the Nashville population which is Florida, South
Carolina, Pennsylvania, New England and New York, so I mean these are some of
the issues so GEO coding comes into it but what are the current technical
solutions and we don’t even know them.

Because as I said it may be that we collect all these data and at great
expense and then nobody can use them because we can’t ask the questions
that we want to ask. Does it make a difference whether you have a certain
ethnic background or does it make a difference that people who are not doing
well tend to be in this location and the majority of people in this location
are this certain racial ethnic group? And Vickie you and I have talked
extensively about the Cape Verdians and Providence, Rhode Island, and you
kindly sent the email around about the New York Times article of a few days ago
about the differences of not only, well the dramatic differences in small
locations, although this was New York City which is not small, in some
locations about people of different backgrounds when they’re all
denominated, what was it Hispanic? And there’s so many different, I mean
the biggest one is that Cuban Americans versus the Mexican Americans, these are
huge groups and they’re very different. So it matters as to whether
you’re Hispanic in South Florida or whether you’re Hispanic in San
Diego.

DR. LENGERICH: Just a couple of points, one is you mentioned the
technological components there, I think you’re absolutely right about the
technology being an important issue but a lot of it lies between what data is
collected and then how it is made available or not made available. If
we’re discussing this in terms of national surveys where there’s a
one on one interview going on it’s barely, for a couple of thousand
dollars once can have a very precise GPS system so when I’m sitting in the
front room interviewing somebody I can push that and I will know within ten
feet exactly where I’m located. So that technology is there and very easy,
the question is how do you move that information to useful information to the
population and that I think, Vickie, I think you were right on when you were
saying that what we need to do is make the information about place available,
the place and the impact of place on health available to the groups that are
interested in improving health for their groups.

And I think that that is what we heard when we did the testimony, people
want the data so that they can build the case for funding for new health care
clinics for whatever for that health to be improved. So I think it’s a
matter of moving the technology, it’s not not having the technology to
locate the place, it’s a matter of not having the technology or
accessibility or assurance that that information can be used.

MR. LOCALIO: If it can’t be used because it makes the data too
sensitive because it’s potential identifiable then I would ask the
question why collect it if you can’t use it. Of course I would be in favor
of collecting it and then working out systems so that it can be used. Now I
have to, when I was thinking of technology I wasn’t thinking in terms of
GPS, I was thinking in terms of masking or jittering the data so that it cannot
be, so it’s not longer identifiable. But the work that’s been done on
that is relatively new, two, I haven’t seen it connected with the GPS
problem, and three, there’s no software to use that, so who’s doing
what, with whom, when, to resolve this and what’s the process by which
these issues are doing to be dealt with. Is Brenda still on the line?

MS. EVELYN: Yes, I am.

MR. LOCALIO: Brenda, you were very kind to tell us that what the current
status of what was going on in the FDA was and that was wonderful because
it’s the first time I’ve heard it. We also need to know what is going
on in the other agencies to address, or what is not going on in the other
agencies to address this problem. I don’t know, maybe I’m the only
one but I certainly don’t know.

DR. LENGERICH: I don’t think that this committee has heard that
presentation, I think that we have heard presentations which says place matters
but I don’t think we’ve addressed the question, like you said, what
are the methods that exist to overcome that issue or what research and
development needs to take place to over come the issue.

MR. LOCALIO: And please bring the software with you when you come to talk to
us, thank you, because again, there’s a difference between something
called vaporware that I just want to make sure is on the record, vaporware is
oh yeah we can do it but somebody has a FORTRAN program that can only be
complied on some arcane operating system and it really is only available to one
person, that’s not software, this should be software that somebody could
get and use on a laptop like Gene has and ask what is the effect of race on
something in Providence, Rhode Island.

DR. LENGERICH: There are methods, some better then others, to be able to
look at very specific data about place, the data can be shifted in a systematic
or random manner if you’re dealing with actual point data. There’s
issues around having data residing on different servers so that people who see
aggregate data like the example you were giving, Vickie, of zip codes in LA
that they can’t actually go back to the raw data, the individual level
data, that there is an effective blockage in place there. So there are methods
that are being developed, they’re in different states of being developed,
that may be applied at different levels of the issue. But it comes back to if
we’re going to collect the data, it can be collected technologically, it
is important to the groups, how do we make, what are the mechanism to make it
available to support their case.

DR. MAYS: One of the things is I think that the Privacy group may know,
Russell, more about, what are the mechanisms that are in place for dealing with
issues of identifiability, I don’t know if they’re going to know
software and what have you but I think that’s why it will be very good for
us to have a discussion with them because they’ve actually been holding
hearings about not our particular issue but they’ve been concerned about
this relative to a number of other issues. So it may be that in our meeting
with them we might learn more about what is kind of the prevailing activity
around dealing with identifiability, trying to figure out, having mass software
that can be utilized by individuals wanting to work with particular data, that
within the federal government should be open and easy to use. So I have a
feeling that they have some ideas about this, I think it’s good.

I think what we need to be able to do is really explain what the
difficulties are that we have relative to our data needs for these populations.
Probably the more specific we can be about either barriers we encounter or
things that we would like would be useful to them because it’s almost like
their expertise is the privacy part, ours is in the population part, and I
think the better we describe the population part then the better that we can
have a collaboration with them because they’ll describe the privacy stuff
to us.

MR. LOCALIO: I would say that I want to be a little bit broader then that
because the Privacy Subcommittee certainly would have some, if not current
knowledge be able to obtain the information that might help us and provide some
expertise. But the Quality subgroup certainly has some bearing on this, also
the Standards group because again, if we’re collecting this information at
the health plan level or something like that and we’re collecting race and
ethnicity date how is that information going to be used along with other
sources of information to serve the needs of the various groups.

And last but not least, today, this afternoon when we talk about the
National, the NCHS Child Study, I assume that’s the longitudinal study,
well, it’s only relatively recently that we figured out what the study
design is going to be but given that now we know what the study design is I
would really want to know what are the provisions in that study design for
collecting information on race and ethnicity and place, and then what are the
plans for making that information available, because this is a very important
study and who has worried, I love to use the word worried, who has worried
about these issues and what are the plans. Or, as I hear, are the data going to
be collected and held very tightly and not available to the people who need it
and just used by certain designated insiders, that’s not very cost
effective.

DR. MAYS: I think they’ll actually answer that today because I put that
question before them in terms of when they said what are some of the questions
we want to discuss and that was actually one of them so I think we’ll
learn something about that today.

What you’re raising now is a procedural question and that is while
we’re going to meet with Privacy is this a broader full committee issue.
Remember quality was thought to be a full committee issues for some of the
reasons that you’re talking about in terms of this and the question is are
these issues that should be at the full committee or are these issues that stay
in the subcommittee.

MR. LOCALIO: Well, why don’t we start with Privacy —

DR. MAYS: Well, we have to, we’re already set up to do that.

MR. LOCALIO: And start with them and say what does Privacy think. Once we
have them and we can discuss the issue and by the way I’d love to hear
what people’s viewpoints are because they have a different perspective,
and then we can say what should we do about quality and what should we do about
standards and we’ll see where it goes. But I have to say that it would
really help if we were able soon to have a more formal input from NCHS, more
formal then email, to really sit down and chew this out for a couple hours with
NCHS and lay all the issues out, and all the options, and get an idea of
what’s going on, and that is really within, very much within our port I
think. Getting the issues out because right now we’re starting on, really
starting, I know we’ve been working on it for a long time but we’re
really starting on some of this.

DR. MAYS: No, I agree. Let me just raise one thing and that is again in
terms of process, we’ve talked about, talked about, talked about until
I’m kind of blue about it but we’ve talked about the Census people
having a relationship with us and it’s gone from Census having a
relationship with the full committee to Census having a relationship with the
subcommittee, I think that this is a time where having a relationship with
Census and having Census be one of the agencies working with us would be very
good, Russell, in particular on this issue because I think that Census has a
set of rules and procedures that it engages in, Census has managed to have all
these data center, Census has managed, so I think what we might want to do is
also see, and I just think when it’s all said and done as we get near the
end potentially I need to either write a letter or have a conversation with I
think it’s either Jim or Marjorie about all the complements that we need
of resources and things that we’ve talked about that just need to move
forth on.

So how would you all feel about trying to get Census to the table? I
can’t guarantee that they’re going to be at the table for November
4^th, that’s just way to quick, but that we just move ahead with
the Census invitations to be with the subcommittee because I think this is a
topic on which it would be really useful to have them with us.

MR. LOCALIO: I agree, I mean my minimum sentence here has at least two more
years to run so I mean I have plenty of time and maybe our goal is once a
quarter to have another group come or we go to them and discuss things,
certainly Census is an obvious choice but I can tell you right now I don’t
even know the list of agencies that we ought to talk with —

DR. MAYS: Let’s ask the pharmacy people.

MR. LOCALIO: I don’t know, what are people’s plans, that’s
what I would like to know.

MR. MAYS: Also we might want to have a presentation about, and Audrey maybe
you can tell us a little bit more about this but I think is it called CIPSEA,
is that how you say it, CIPSEA, because I’m sure the Data Council has
talked about the passage of CIPSEA.

Can you introduce yourself and then just make a few comments about CIPSEA
because it would benefit us to have a presentation about CIPSEA.

MS. BURWELL: Audrey Burwell, Office of Minority Health and lead staff to the
subcommittee. And CIPSEA legislation that passed, I guess it was about a year
or so ago, is being shepherded by the Office of Management and Budget and has
to do with data sharing and also various levels of data safeguards and access
to data. The Data Council had a presentation on it, probably six months ago,
where the representative from OMB really did sort of put the question back on
the Data Council as to can we think about more innovative ways to allow access
to data so the question is out there. I don’t know what the acronym is and
Russ has the law in front of me, he loves to bring the law with him.

MR. LOCALIO: I like to do my homework. I just brought my file —

MS. BURWELL: But this particular piece of legislation offers a lot of
protection for non-disclosure and such and —

MR. LOCALIO: It’s the Confidential Information Protection and
Statistical Efficiency Act of 2002, and yeah, it would be good for example to
have somebody explain this to us because maybe I shouldn’t say this but
often legislation is not written with the clarity and simplicity that one would
like.

DR. MAYS: We have a problem because he’s an attorney, if he’s
saying this then we don’t even need to read that, we need a presentation.

MR. LOCALIO: I’m sure I can find a thousand judges in the country who
agree with me.

MS. BURWELL: But I think a presentation wouldn’t be hard to arrange at
all if that’s what you would like.

MR. LOCALIO: The reason is I find this particular statute to be, how shall I
say, not inconsistent but somewhat have a different flavor then the statute
under which NCHS operates. And I would say well did the people who drafted this
understand what the 242B says, did the right hand completely understand what
the left hand was doing when this was occurring.

DR. LENGERICH: I would support all of the suggestions made in terms of our
process that NCHS and Privacy and Census and this new law, or new to me anyway,
all be part of our process for being fully informed. I guess I would just ask
that it, or recommend that it be sort of sequenced in to how we approach this.
I think that I guess I still go back to those testimonies that we heard from
groups and we approached them specifically with NCHS data systems in mind that
to me it seems that the first thing to, and that raised the privacy issues, so
I think to me kind of the first or what I would put as a priority is to deal
with the health of those groups, NCHS data systems and privacy kind of together
and then moving forward to the Census and then also to the new regulations
because as you said, Russ, that is maybe a little bit different then what NCHS
is currently working under.

MR. LOCALIO: Yeah, when I read this I thought it was better written then the
NCHS controlling statute but still, how do these work together. Well, I think
the answer is they probably don’t work together because they have
different labors. By the way I did reference all of these things in that few
pages I wrote up so I tried to pull things together but we need much more
detail then this. And I do think the sequence is important, we have to make
sure that we do two before four and that three comes in the middle.

DR. MAYS: Here’s what I’m going to suggest is that we set aside a
day and that in terms of setting aside a day that what we have is a day to
learn about these issues. But I agree with what you just said which is we want
to learn about these issues of privacy and confidentiality and data access
within the context of our populations. We don’t want to just learn,
Russell, we don’t want to just learn about the law, we want to learn about
how to utilize this relative to accomplishing the goal. And so I think rather
then having people come and do a dog and pony show, that’s the law we just
learned about, what I need to know about is how that law either can facilitate
or serves as a barrier for my ability to be able to share and access data on
Native Americans in Oklahoma if that’s data that’s available in some
federal dataset.

MR. LOCALIO: I think what you’re suggesting is very good, let me just
interpret what I think you said. If we started from the end point of we have
identified a certain number of needs, one, two, three, four, five, six. Given
these needs how would one go about getting the data, what would be the
requirements to use the data and what would be the barriers in using the data,
so it will be very ends oriented, we’re not doing this as a thought
experiment or as a theoretical exercise. Russell wants data on this, how do I
get it, how can I use it, and what am I going to end up. I have to interject
this, so much of what I see in terms of results in manuscripts is terribly
flawed, terribly flawed because people don’t have the data that they need
and they can’t answer the question they want to ask. So we should ask the
kind of question that you have in mind and say okay, whose approval do we have
to get to do this. Privacy may say well you have to get IRB approval from your
institution, fine, we can do that. And then what are the other hurdles and we
may want to lay out somewhat of a roadmap and where are the potholes and where
are the washed out bridges for anybody who wants to do this. Would that be
okay? Very applied.

DR. MAYS: Yes, exactly, I mean I think to some extent we can be very liberal
and just deal with these issues at this level of this is not good but
let’s deal with them at the level of this is the path to take to make it
better.

MR. LOCALIO: And so I think our recommendation should focus on we’ve
identified the problems to do these selected number of tasks, which we think
are representative of what needs to be done, and so these are the areas that
need some attention.

DR. MAYS: Here’s the other thing I’d like to do is also turn to
our staff list and see which of the staff can help us in the preparation of
this particular activity. Now Edna is not here today but I’m going to
speak up on behalf of Edna because this is something I know that is near and
dear to her heart so I definitely want Edna involved in this.

MR. LOCALIO: In order to further the tasks or the scenarios that we’ve
just outlined it might be helpful, Vickie, to elicit at least informally some
expression or some, I don’t know, to assess from the people, from
additional people who want to use the data as exactly what they want to do.
We’ve heard some testimony, we know because of the kind of work that the
three of us do what we’d like to have, but I would really like to know
from people who are heavily into this field, other people are heavily into this
field and I’m not by the way, I’m just another statistician among
other things, what they need, what they would want. Does that make sense? Or do
we know that well enough?

DR. MAYS: Well I think it might be better to have it on record and I can
think of off the top of my head quickly even three people who can do an
excellent job. One is Paul Ong(?) who just came off of the Census Committee and
in terms of dealing with the populations of Asian, Native Hawaiians and other
Pacific Islanders actually did an excellent report where what he struggled with
was the data that he needed, I mean he actually tried to come up with some GEO
coding efforts to be able to talk about health needs and health utilization of
American Indians in the Los Angeles area, we have one of the largest urban
populations but the difficulty that he faced, and he was on the Census
Committee so he knew a lot about for example how to get that.

The other person who has recently joined the Board of Scientific Counselors
is Matt Smith, Matt Smith is a sociologist by training, American Indian, and he
also will tell you, and he was on the Census group, so again, part of what we
might be able to get, these are individuals who know these areas very well.
What we might do in this is to start our day for example with some testimony of
if you had a wish list and what is your experience in terms of accessing this
data I think it would be, I agree, that keeps the day in a context. And then to
have the other people who follow after that who want to tell us about these
different rules to have heard what these individuals say.

MR. LOCALIO: Well that would help, I mean it would help if we had, we put
them at the same table and said go at each other until it’s done, I mean
that’s a great —

DR. MAYS: I don’t quite think we can do that but we can have them in
the same room and we can have these individuals testify first and the people
who come after them, if they’re doing just a little dog and pony show that
doesn’t answer any of those questions I think they’re going to have a
hard time.

MR. LOCALIO: It should be very applied because we don’t want some
theoretical argument this is what we have to deal with, we have to have the
people that have the needs to state what the needs are and then we should say
how are we going to resolve these issues. I think we know what some of the
issues are from our own experience and from people we’ve heard but I would
agree we need to get more on the record, otherwise this is just a waste of
time.

DR. MAYS: What we also should get is a couple of people to talk about NCHS
datasets, I mean the people that I actually suggested have used the Census and
I think that that’s important particularly if we’re starting to talk
about GEO coding, but let’s also try and figure out in terms of some of
the NCHS ones, and what we might want to do is to not just get all over the map
with populations, why don’t we take the population of American Indians,
Alaska Natives, and Asian Native Hawaiians and other Pacific Islanders because
I think that’s where the complexity of these issues come. We’ll have
to think about who else, Audrey we should probably talk about that, who else
might be some suggested individuals to present, I mean this may even be a time
to think about bringing Dan Friedman because Dan Friedman will tell you some of
the difficulties of doing this relative to BRFSS and some of those datasets.

MR. LOCALIO: And those rules are different you know, and why, I mean I
can’t understand why the rules are so different and if it’s because
each one is controlled by a separate statute does that make sense.

DR. LENGERICH: Vickie, just to take the idea of the populations a little bit
further is we might even kind of pose a semi-real scenario, we might identify a
population in a geographic area who become concerned about a potential health
outcome issue for their group and that they believe or they are aware of a
disparity that exists there and their attempt to secure data from a national
source to verify, document, to seek support for doing something about it. And
what are the barriers, either real, what are the barriers that might occur
along any process, data don’t exist, data exist but I can’t get to
them, I can’t get to them but I don’t know how to analyze them, I can
analyze them but then I can’t do anything with them, and money, and sort
of try to understand a little bit about what the process, where the barriers
arise, what they are.

DR. MAYS: Can you help us with that? I mean I think that’s a great idea
—

DR. HAYNES: Can I just ask a question?

DR. MAYS: Can you introduce yourself?

DR. HAYNES: I’m Susan Haynes, Office on Women’s Health. Is this a
simple matter of sample size and not being able to analyze data properly
because you don’t have enough power and numbers in these surveys and are
you beating your heads against the wall by trying to use the national surveys
to get your answers versus going and doing a unique focused representative
sample in that area? Which is what we faced when we did Hispanic HANES, the
national surveys didn’t give us enough, the over sampling didn’t give
us enough, and so we actually did a unique survey across the United States of
all Cubans, Puerto Ricans, and Mexican Americans. And are you just beating your
heads, I mean this is all just talk because it’s a really a matter of
power in numbers and you’ll never get it, no matter which survey you look
at.

DR. MAYS: I’m going to make a comment and then I’m sure the
statistician in Russ wants to make a comment —

MR. LOCALIO: I think your point is well taken but at least I for one had not
been thinking in terms of sample size because I don’t think anybody would
expect to be able to study this type of problem with a national survey. The
issue can be better characterized that if NCH collects the data it has to
operate under a statute and a set of rules and guidelines which has certain
implications for the availability of the data to the ultimate user. If the same
survey is done using federal funding under an NIH grant the issues are entirely
different, do the exact same survey of the same people in the same population.
I think that’s the issue, one of the issues, so I don’t think
it’s sample size, it has to do with if there are going to be data
collected what are the issues that people have to face in terms of using it and
the source of those data. For example in the Hispanic NHANES, what is that
called, Hispanic NHANES?

DR. HAYNES: Hispanic HANES.

MR. LOCALIO: Same issues would apply there as to the regular NHANES —

DR. HAYNES: I don’t understand why, I mean what is it that
couldn’t be done with NHANES, Hispanic HANES, that we know could not be
done with regular HANES? Name one issue that you can’t address in a
focused survey versus in the regular representative sample of HANES.

DR. MAYS: Let’s say I want to learn something about the health status
of American Indians in the state of Georgia and no, I shouldn’t do it by
state, I want to do it by population, and what happens is that I can’t get
those because once I do anything other then just say here’s the frequency
other analyses that I may want to do might, they might be a mess, so then what
happens is in order for me to be able to do those analyses I have to come to
Rockville. Or what I have to do —

DR. HAYNES: But then you’re saying then the issue is I want to analyze
data by state, that’s your bottom line.

DR. MAYS: No, no, no, I want to analyze it by a population, that’s why
I changed it because state wasn’t good, but I want to analyze it by a
population and in each year let’s just say there are 40 people and so
it’s like, I might be the kind of statistician that I can deal with 40
people depending on what my outcome is that that’s a big enough in for me
to be able to do the kind of analyses I want to do. But the problem is once I
ask for age, once I ask for certain things —

DR. HAYNES: Age and sex you’re over —

DR. MAYS: That’s it, that’s it, and so what happens is that I
can’t get the data unless what I do is use either remote access or I have
to come to Rockville.

DR. HAYNES: Okay, well let’s say it’s Cherokees and I say to you
let’s do a study of Cherokee Native Americans in the Southeast, just like
we did with Hispanic HANES, we’re going to have all of the Cherokees from
Tennessee to Florida, do it in a ten state region. Then you’ll be able to
answer your questions because it won’t be just 40, it will probably be 400
or 500 or 1,000 and you’ll be able to do your analysis. So you maybe
focusing on Georgia but you’ll get the answer for the Native population.

DR. MAYS: I just want to check and see if the Quality people, wait a minute,
let me just see if the Quality people are online. If you’re online can you
identify yourselves?

MS. EVELYN: This is Brenda at FDA.

DR. MAYS: Okay, anyone else? Okay, go ahead.

MR. LOCALIO: Because it makes a difference, you don’t want to know just
what the overall experience of Cherokees —

DR. HAYNES: But that’s what the question is.

MR. LOCALIO: The study population is Cherokees but you need to know whether
the disparities in health between Cherokees for example has to do with the fact
that these are Cherokees or the fact that they happen to live in a certain part
of Tennessee or a certain part of Florida and are there differences depending
on where you live. One of the big problems I hear about is you can’t get
the PSU, forget about GEO coding, you can’t get the PSU. So for a lot of
purposes the analysis can’t be done because you can’t even get the
sampling unit.

DR. HAYNES: But you could if it were a focused sample on Native Americans,
you’re stuck in the regular representative sample, you’re stuck
there.

DR. MAYS: Just to be clear, we don’t have control over that, we have
put forth a recommendation about doing these Community HANES is the title that
they’re using, so that is a recommendation that’s on the table. But
for the majority of data that’s collected that users actually have access
to there is data on racial and ethnic groups but here are the problems
we’re facing, we’re trying to do the two prong thing and say we all
know we’re only going to be able to have these Community HANES like once
every other year, what we’re trying to do is get the Secretary to put
these things on a cycle, we’re hopeful they’ll be a series of them,
but again, it’s like the question is we’re collecting all this other
data, the information is sitting there in some minority population, if you
amass the data over several years you might be able to use it to get to larger
numbers but again, Russ’s point I think is right on target, sometimes you
can’t get PSU, sometimes once you start trying to display things by ages I
just, we’re hitting our head against the wall and I don’t think that,
and attention to be able potentially to make some changes that would allow us
to figure out ways in which to access more then we’re accessing.

DR. LENGERICH: I would just add a little bit here and say that I guess I
sort of work by the premise of if the problem is in the community then the
solution is probably in the community as well. And so using your example there
of the Hispanic HANES, if there’s a community group of Hispanics in LA
would they be able to get access about Hispanics in LA to support, justify,
monitor their attempts to improve health of Hispanics in LA from the Hispanic
HANES. We expect that there were data collected from that group either
intentionally —

DR. HAYNES: It’s probably half of the sampling unit.

DR. LENGERICH: But can they get those data, can they get a hold of those
data, are the data available, can they get them, can they analyze them, can
they manipulate them and put them into their process to improve the health of
their community.

DR. HAYNES: And NCHS would say no.

DR. LENGERICH: Well, that’s the problem.

MR. LOCALIO: Actually Gene, could I just elaborate, it’s not in Los
Angeles, Hispanics in Los Angeles, it’s the Hispanics in East LA versus
the Hispanics if there are any in Westwood, I think it’s even more
specific then that because LA is a huge diverse community and diverse not only
in terms of race and ethnicity but socioeconomic position. And it may be that
if you look at Hispanics in East Los Angeles they are identical in terms of
health with Native Americans in East Los Angeles, and they’re very
different from the Hispanics in Westwood if there are any. Because the
problems, you can’t, when you look at problems especially of disparities
you have this issue that disparities may be terribly confounded by place and
this is a problem that we see again and again and again, and by other things —

DR. HAYNES: But you have people like George Kaplan who have done analysis by
people living in poverty and where you are in the United States if you live in
poverty, poverty, primary sampling unit or Census unit or whatever, there are
disparities compared to folks in your same town who aren’t in poverty. He
would say that the city is not unique but the circumstances, the income
circumstances and access to medical care and things like that you get in these
poor areas is what drives the disparities across the United States. So you
don’t need to analyze East LA versus Westwood because we know the answer
already from Kaplan’s work.

MR. LOCALIO: No, no, no, I wouldn’t say that, I mean —

DR. HAYNES: In other words I could take people living in poverty in
different cities in Chicago and LA and Atlanta, compare them to people not and
I think the results would be the same.

MR. LOCALIO: I’m not sure that we’ve asked all the questions we
want to ask, I think what I view one of my purposes is when people do come up
with questions they have the data and the tools and the access to answer them.
And by the way, reading this again, just to follow up on your point, reading
this report about the Native Americans, Native American Health Care System,
fascinating stories in here about just how far away the Indian Health Services
are from where people live, but that may be very specific to where people live
so you can’t just look at Native American, you have to look at the
Billings Native Americans because the issues may be very different. And again,
you can’t give people a —

DR. CARR: Hi, this is Justine.

DR. MAYS: Hello, Dr. Carr, how are you?

MR. LOCALIO: — Native American NHANES if there were one.

DR. HAYNES: Why?

MR. LOCALIO: And just say well here it is, we can’t give you PSUs but
you can do with it whatever you want because it won’t answer their
question.

DR. MAYS: Okay, let’s summarize this so we know what we’re
direction we’re going in and move on to our next topic.

We’re going to have a day meeting, and I’m loath to call it a
hearing because it gets done in a different way, but we want some testimony in
the morning let’s just say from some of the difficulty that people are
having accessing data, I think Eugene, we’re going to work with the notion
of even having a case study as opposed to like these questions, but a case to
some extent and have people then describe whether they’re able to access
the information and what problems they would go through, I think that’s a
good idea.

And then in the afternoon what we want to do is inform ourselves so we want
probably Allen Zarati(?), I think I’ve said his name correctly, from NCHS
to come and talk to us, but I think the other thing we want to do is to bring
in a series of people to talk with us. We’ll work that out, Census, NCHS,
and others.

Staff, Audrey you want to work on this one with us? And let’s look on
our staff lists and also have Edna work with us, and we’ll see. Suzanne,
you want to work with us on this? Oh, we’ll hold you off, no there’s
actually something else.

DR. HAYNES: I’m coming from a Hispanic HANES point of view —

DR. MAYS: Believe me, we have several issues so I have another one that I
actually want you to work with us on.

Thank you. Brenda is on the phone, Justine is on the phone, do we have
anyone else —

DR. STEINWACHS: Vickie, this is Don Steinwachs joining you, hi.

DR. MAY: Hi, Don, how are you?

DR. STEINWACHS: Doing well, I got my big proposal done, I’m trying to
come back to life again and be a good worker again.

DR. MAY: Oh, I understand after a big proposal it’s like you go
there’s life.

DR. HOLMES: Julia Holmes call in from NCHS.

DR. MAYS: Oh, hi Julia, how are you?

DR. HOLMES: I’m good.

DR. MAYS: Great, welcome.

Agenda Item: Candidate Recommendations of Quality
Workgroup that Overlap with Populations – Dr. Mays

DR. MAYS: Okay, we’re at a good time then, we just missed our break but
we’re at a good time, if people need to get up I have to tell them they
can but since we have our majority of people on the line let’s see if we
can just keep going here.

Julia, can we actually ask you to give us some kind of a summary of what has
taken place in your two meetings, you had the hearing that was on September
14^th and then you had another meeting, I can’t remember exactly
the date where you also were talking about some of the candidate
recommendations.

DR. HOLMES: Unfortunately I’m probably not the best person to give you
a summary, at the hearing on September 18^th I think that there was
agreement from those who testified that the one candidate measure should go
forth. Then I believe that the Quality Workgroup had a subsequent meeting to
discuss the future of the other candidate measures and there was kind of this
idea that perhaps a good focus for the future of the workgroup would be
continuing to work on promoting and seeing the viability of these functional
status indicators. But as I say I wasn’t a part of this immediate past
meeting of the Quality Workgroup so I don’t think I’m in a good
position to say definitively what happened.

DR. MAYS: Justine?

DR. CARR: Yeah, I’m just pulling up my notes here, from I guess I got
this earlier this month. The Quality Report First Addenda, a Proposal. And
candidate recommendation three was the one we could all agree on and perhaps if
I could read to you from what I have in the summary, about a paragraph or two.

While the testimony sought in the three hearings was focused on the first
eight recommendations the testimony often went beyond the narrow confines. This
broadening look took two directions, in the first two hearings several
testifiers spoke to the importance of all 23 recommendations as important in
the evolving measurement system for the improvement of health and health care.
In those comments no suggestions of omissions in content of the original report
were offered. In the last hearing the testimony broadened to include discussion
of candidate 11, modify existing mechanisms for reporting race and ethnicity of
subscribers and dependents on the HIPAA enrollment transaction and 12,
investigate how best to capture race and ethnicity on a standard provider
transaction.

Just a little bit more here, it says since the first testimony merely
endorses the general agenda which the NCVHS is addressing no further action is
sought. The second, discussion of the capture of culture, defining race and
ethnicity data elements, pointed out the need for further hearings related to
implementation issues. By the nature of the NCVHS organization structure these
hearings should be jointly conducted by Standards and Security, Populations,
Quality, and given the sensitivity around identity, Privacy. In view of that it
seems that this hearing structure would be best to involve the full committee
and be best addressed as soon as the need for this information to address the
reduction of disparities as well established.

Is that helpful?

DR. MAYS: Yes. I think that there is recommendation, your candidate
recommendation three has been sent to the full committee so that we’re
going to deal with. What we’re trying to figure out are a couple of
things, there’s the overlap of the Populations Subcommittee with the
Quality Subcommittee, and where you all are relative to functional status. When
I listened to the September 14^th hearing and I guess it was Ken
Kizer brought up this issue of the importance of functional status I kind of
stayed with that issue and it appears that you all have also picked up the
issue of functional status as important in terms of quality assessment.

DR. CARR: Yes, I think that that’s true, that there are several
recommendations about capturing data on functional status and that it’s
clear that that’s very important in terms of looking at patient level
outcome.

DR. MAYS: The question is whether or not there is an area of overlap in how
Populations can be helpful to the Quality Workgroup. I know earlier we talked
about in terms of the candidate recommendations there are three of them for
race and ethnicity that we would explore what data we already have that can
help push those forth as recommendations as well as whether we should have
hearings and I just don’t know where the group is, where the Quality group
is on that and whether we should be preparing ourselves in Populations to
consider having a set of hearings with Quality.

DR. CARR: I think that that was what folks were thinking when we discussed
this at the last meeting. I guess, well, I think we also talked about this in
the context of sort of the research agenda or focus for this year but compiling
what’s already been done I think is important and then identifying where
are the gaps and what hearings would fill those gaps.

DR. MAYS: Maybe what we should do, and Debbie, can you come to the table so
I can ask you a couple of questions about the research agenda? Each of us,
Debbie, I think had a set of research recommendations that came forth, are they
going to be in the agenda book for the entire, each groups, will they be in the
agenda book?

MS. JACKSON: That’s right, we got some responses from Audrey, Debbie
Jackson, got responses from Audrey for Populations, got an item from Anna Poker
for Quality, and from Mike Fitzmaurice, so those will all be in the agenda
book.

DR. MAYS: So next week everyone should be getting those?

MS. JACKSON: Yeah.

DR. MAYS: So Quality will get ours and we will get Quality’s, so I
think that what we may want to do is also be guided by what we’re seeing
in terms of the research recommendations to figure out as you said Justine
whether there are gaps there as well as making sure that for both of us we put
forth any of the research that we think will help us in terms of the candidate
recommendations in the Quality Workgroup.

Did you all have a sense of when you’re thinking you want to do this in
terms of having hearings? Justine, I’m sorry to put you on the spot, we
haven’t been able to get Bob.

DR. CARR: I guess Bob Hungate is not joining us today?

DR. MAYS: I was just going to say I’m sorry to put you on the spot
because we’ve actually been trying to get Bob Hungate —

DR. STEINWACHS: He’s not going to be with us today so on scheduling I
don’t know, I think Bob would like to move things ahead as quickly as
possible was my general sense of Bob.

DR. CARR: I think what would be helpful is to sort of scope out what’s
known, what work has been done, and what do we need, and then maybe by when we
meet next week begin to identify the dates that would be possible for the
hearing. I’m just kind of deferring to Bob a little bit on these thing,
what I was just reading to you, that was his summary of it and as I look at his
summary I don’t see in this particular summary the reference to functional
status although you’re absolutely correct, that there was a lot of
attention paid in multiple hearings to the functional status. So I guess, I
apologize if I’m out of sync but just starting from ground zero with this
it sounds like there’s certainly we want to move ahead with the functional
status recommendations and more hearings and then what about what Bob had
written about the recommendations on candidate 11 and 12, modify existing
mechanisms for reporting race and ethnicity, is that something that’s a
near term or is that a longer term, or how does that fit with Vickie?

DR. MAYS: Here’s again, we don’t have Bob or Anna so I think
we’re all kind of struggling a little bit. Here’s what I understand
that’s part of why we wanted to be able to discuss it today is in those
two recommendations we can offer you comment because we’ve already had
some testimony but at the same time I think we do need to probably have a
hearing on this to talk about when you’re asking groups to collect this
data how difficult they think it is, what the barriers are, because remember in
the September 14^th hearing I think that the issue, and it was only
touched upon a little bit, is this is not as easy as it seems. Like when you
say let’s collect data on race and ethnicity we already know that how
people do it, we know, and there’s some difficulty in that, we know that
people sometimes have difficulty in knowing the format in which to do it and
that you need some probably, I talked about this before, training, because will
check Hispanic and it’s like oop, you forgot that you also need to —

DR. CARR: This is true, who does it, whether it’s an unknown
registration clerk or the physician and when they do it in the patient’s
encounter and how they ask it and what data definitions they’re using and
what the training is to ask and capture this data are key. And I speak for our
own institution that it’s a formidable, quite formidable task because it
usually falls to the front line registration person and I suspect that’s
pretty much it’s happening most places. So I guess a question for us to
think about is is there, do we believe that there is a direction that we want
to put forward that we decide, that a data definition be decided upon and that
we would research more about how it might be collected or even whether this
could go forward. I mean I think that is kind of the big, even the two kind of
data definition approaches are very different. Am I right?

DR. MAYS: It’s interesting, I think that that’s one —

DR. CARR: So I think we also need to bear in mind as I’m sure you all
know full well that the federal government in terms of reporting race and
ethnicity is bound by the ’97 requirements that suggest a particular way
that race ethnicity is to be gathered and then reported and I suppose the issue
is to the extent to which we can translate that into people gathering data on
race ethnicity in the private sector so that we have comparable data on race
and ethnicity, this is oftentimes been a problem. And for example for the
National Healthcare Quality Report and the National Healthcare Disparities
Report we encounter quite a few problems in terms of the reporting of data by
race and ethnicity because of the fact not all federal agencies for the data
that were reported in the first NHQR and NHDR, not all federal agencies had
made the transition to the ’97 requirements, OMB requirements, in the
private sector and there were a few data systems from the private sector. They
are obviously under not the same set of requirements in terms of how to collect
and report it so that’s a major issue.

DR. HOLMES: Is that something that we can influence? We’ve got an
agenda, so now you’re saying the 1997 OMB definitions are what all federal
agencies use but not private agencies.

DR. CARR: Yes, that’s correct. Private groups are not bound obviously
by the federal OMB requirement but it’s certainly, I would assume,
I’m not that familiar with the federal government but I would certainly
assume we can’t require that but we can certainly suggest that the
importance of having data collected and reported in comparable ways so that
adequate and valid comparisons can be made.

DR. MAYS: Okay, let me just try and recap a couple of things and also
because the people who are recording are coming to me to find out, so when you
want to make a comment online if you’ll just say who’s talking it
will actually help them.

Sounds like what we should cover is who does this assessment of the
collection of data on race and ethnicity, when they do it, and my bet is that
it kind of winds up being the clerk, but I think we ought to find out from
systems when they do it and whether or not that’s the best way, it may be
that we need to come up with a recommendation of when is best to do this. And
then I think the issue of definitions is really what is probably the key issue
here and that also one other aspect that wasn’t talked about is the
different between people in states how they want data versus how it’s then
done at the federal level. Often when we’re talking about the federal data
collection we’re talking about what’s minimal and that’s kind of
like the five categories and that’s it. But at the state they really want
to break it down even more and they may have state reporting that they’re
interested in and again, it’s like Russ and I often tease about when you
start thinking about it at the state level some of the necessities of having
data say for instance we’re talking about Massachusetts, people want to
know about the Cape Verdiens, if we’re talking in Los Angeles they want to
know about the native Hawaiians, I mean so in terms of how things are reported
it depends on who needs that data and I think that that’s going to present
the biggest challenge is what type of advice, training, recommendations, etc.,
can we make.

DR. CARR: You’re fading a little bit in and out —

DR. MAYS: So I think the things that we need to cover would be who, when,
and how do we define, and I think just that in and of itself would be very
helpful.

Now I think the question is who do we want to talk to about these issues and
I think this is where the Quality Workgroup should probably, I mean we can give
you some ideas but in terms of, for the purpose of ending up with
recommendations that actually can get implemented who should we hear from
around these issues?

The question that I was asking is when we make these recommendations we want
to hear from the groups and the industries that actually have to enact them
because that goes a long way in terms of facilitating change. Who are the
groups that you think it’s important that we invite in and ask them about
these who, when, and the definition questions.

DR. HOLMES: Well I think NCHS should be invited, we obviously run a lot of
the large health care surveys that are used to measure quality and gather data
on race and ethnicity and also I would like to point out that to add another
wrinkle to the whole difficulty of measuring race ethnicity is what a project
that NCHS has recently been working on which is the Bridging Project, which
tries to bridge the changes that were necessitated by the OMB 1997 requirement
or new requirement of how race and ethnicity should be reported because
obviously a lot of the data that are reported and are important public policy
issues are reported over time. And that this change in the requirement as of
1997 requires that some bridging mechanisms be instituted particularly for
example for the issue of the new requirements that people could report multiple
race whereas before they had been forced to make a choice. So for example
there’s a lot of work coming out now and so for those reasons I think NCHS
has a lot of expertise and a lot of history in terms of collecting data,
reporting, and the issue of problems with reporting over time when there have
been changes in the definition of race ethnicity.

I also think AHRQ should be involved, particularly those who worked on the
National Healthcare Disparities Report, because they saw first hand the
challenges in terms of trying to report on race and ethnicity across multiple
types of data collection efforts. And in terms of the private industry I
can’t think off the top of my head who should be specifically involved but
it should involve some large employer groups because they oftentimes encounter
difficulties with requiring that race and ethnicity be reported. For example I
used to work at Blue Cross/Blue Shield of Michigan and we had no reporting of
race ethnicity on any of the, I don’t know about the enrollment but the
claims form, and that was simply something that was not considered appropriate
to gather. So I think cases need to be made for the importance of why is it
necessary to gather that information.

DR. MAYS: Justine, do you have any suggestions of who we might want to hear
from particularly in the private industry, private side?

DR. CARR: Yeah, I mean we could certainly round up some, I mean I know folks
in Massachusetts, even perhaps Harry Reynolds might have some thoughts about
who would be on the private side. I guess the question that I had was what
about at the state level, we ought to hear at the state level from where there
are additional or different requirements definitely.

DR. MAY: And I think the best ones are Massachusetts and California.

DR. CARR: We should hear from AHA and AMA as to their perspective in terms
of they represent providers whose offices and institutions are being encouraged
to gather these sorts of data.

MS. EVELYN: I think the pharmaceutical industry would be a good one to have
because they have a lot of resistance, number one, to doing it, and it’s
something that we’ve been trying to get them to do. I think their
perspective might be valuable in this situation.

DR. HOLMES: I think that’s a good idea, too, this is Julia.

DR. CARR: This is Justine, I don’t get the connection exactly, the
pharmaceutical industry, how does that tie, how do they, they don’t
collect this data do they?

MS. EVELYN: Well, we’ve been trying over the years to try to get them
to report to us the enrollment in clinical trials and so it’s not
necessarily related to health outcomes although it might be I guess if you
really thought about it, and by race and ethnicity and the difficulties that
they have reporting according to the OMB categories because many of their
trials are conducted offshore in countries that don’t use our categories.
So it might just be interesting to hear their perspective on that.

DR. MAYS: What about in terms of Phase IV post marketing, would that be
relevant, Julia or Justine?

DR. CARR: Pardon me?

DR. MAYS: In terms of from the pharmaceutical side in terms of Phase IV,
once drugs have hit the market, in terms of kind of the marketing phase. FDA is
actually supposed to monitor drugs in a Phase IV but I think budget
constraints, it doesn’t necessarily do that much and the question is is
there any way to be collecting data on race and ethnicity in order to, well,
just collect data in general but to collect data on race and ethnicity in Phase
IV post marketing surveillance, if you all think that that’s useful to the
work of Quality, if not I mean that’s fine, I was just asking.

DR. CARR: Well, perhaps the, is it pharma that represents the, is the
interest group that represents all the pharmaceutical industry —

MS. EVELYN: Yes, it is.

DR. CARR: And perhaps it would be they who would be the appropriate
representative of the pharmaceutical industry because I mean don’t they
provide data to large databases about the characteristics of drugs that have
been prescribed to certain people which would include socio-demographic data?

MS. EVELYN: I don’t know that they do, they might but I don’t know
that they do, I’m not aware of any databases that they contribute to.

DR. MAYS: Okay, well we’ll do a little more investigation I think —

MS. EVELYN: It’s worth a though, it’s something to think about.

DR. HOLMES: And I think your point about clinical trials is interesting.

DR. CARR: This is Justine, so as I’m listening to this it sounds like
there’s sort of two parts, one is sort of the basic data collection of
each individual and the second part in terms of the pharmaceutical,
potentially, for example a hospital based trial might just report out if they
had to report their race ethnicity data they might just pull it from their
hospital dataset and give you the same information. So one question I would
have is what would we be looking, what new information about, what new
information would we be getting. I guess it see it as a couple of different
ways, one is sort of coming up with a data definition that meets the needs that
are out there and the second one is the implementation of that data collection.
And then the pharmaceutical I see more as sort of the application of those
definitions telling us about the trials but I don’t see it addressing the
data definition and the application so much because I suspect a clinical trial
would just pull from the hospital dataset if it were a hospital based trial.

MS. EVELYN: I’m not quire sure how to respond to that but the most that
I could probably say about it is that as I said we have asked, because
we’ve had constituents to approach us and say you don’t really know
how drugs are behaving in various and racial and ethnic groups because you
haven’t included them enough to study them and does this in some way
contribute to health disparities. And so as we’ve been trying to work with
the pharmaceutical companies to get them to tell us the racial and ethnic
makeup some have been better then others at reporting but again they have
issues with respect to foreign trials that they conduct and how do those
categories used in other countries match up to what we use here in this
country. So it’s not a direct measure of health outcomes per se or health
status but the question lingers in people’s minds sometimes as to the use
of the pharmaceuticals and do they or don’t they in some way contribute to
the health disparities so that’s the context in which I bring it up.

DR. MAYS: Justine, I want to piggyback on what you just were talking about
and then as Brenda was talking that also kind of raised an issue. Once this
data is collected what we don’t know is where it can get shared, so for
example it could get collected in the health care encounter and the question is
whether or not it crosses the barrier for example to go into the pharmacy
record, I mean because remember there was some discussion in the hearing on the
14^th about when it’s collected it may stay in certain places
and not go to others. So it might be worth our while to explore when it’s
collected, how widely, I mean we have a lot of privacy rules here, so we
actually don’t know within a system, let’s just say within a hospital
system, how widely shared could that information be. Because it could be that
it doesn’t necessarily go over to the pharmacy, it could be that it
depends upon you come into an HMO, see your provider, and then for example end
up being hospitalized, whether or not then that data, what data then is
available if you’re not, depending upon where it is you’re
hospitalized.

DR. CARR: Well, I think I would call upon at the last meeting as you were
saying that people will, may change their response depending on whose asking or
why it’s being asked or how it’s being asked and it raises the issue
of the privacy of what information, what do patients know or individuals know
about how their data is shared. I think that operationally and this is
something that we could find out from AHA and so on but I think that something
like race and ethnicity on a trial or whatever would be another field to be
filled out. So if it’s available electronically it can just be an upload
from a hospital dataset, I think that the person conducting the clinical trial
might just have that as an automatic feed. But I don’t know, your point
about how a person is identified in a physician’s office versus when they
come to the hospital, I don’t think that’s information that’s
routinely communicated, I think really the way it’s captured is the person
is asked at each venue and they answer according to what they think should be
the right answer or the answer that they want to give. I mean it raises that
issue that came up the last time that a person speaking with their physician
might give a more extensive history because their background is going to
influence what genetic diseases they are at risk for in contrast when they go
to the registration clerk in the emergency room they’re going to just give
whatever is simple, fast and expedient because they’re not about their
health risks of genetic diseases, they’re there for their acute injury.

DR. MAYS: I think what we should do is to make sure when we look at the
research that’s been put forth by both Quality as well as Populations,
this is one of those that we should make sure is on the research agenda, I
think the context of the changing of race and ethnicity within the health care
encounter is important. And if neither of us have put that forth I think we
should. Eugene?

DR. LENGERICH: I’d like a little bit of clarification here, I’m
not at all used to working with data systems that are specifically within the
hospital and the inpatient system so I am not certain what sort of information
is available about race, ethnicity, or even place. But I have some familiarity
with working with administrative hospital discharge datasets and those are
usually state operated and a number of them do not have any information on race
and ethnicity. So I’m a little bit lost in the conversation about
collecting data, data that is available about race and ethnicity on a hospital
data system, hospital discharge data system, but I know that at the
administrative level, at the state level, it’s frequently just
unavailable. So is there, am I wrong on that?

DR. HOLMES: No, this is Julia, you’re absolutely right that one of the
problems with administrative data from hospitals, and that includes HCUP, the
Hospital Cross Utilization Project, which is administrative data reported up
from states on hospital discharges, and in addition to our NCHS National
Hospital Discharge Survey, because hospitals are under no requirement to report
race and ethnicity many choose not to do so. And I think another part of this
recommendation, I don’t know that it could be mandated but to provide a
very strong case that this should change and that all health care providers
whether they’re institutions or office based physicians or clinics or
whatever should develop a methodology and a way to capture race ethnicity.
Consequently I think part of the hearing should be to make the case for the
importance of collecting data on race and ethnicity in order to talk about
quality of care and changing disparities over time hopefully in terms of access
to quality of care.

DR. LENGERICH: But it’s not just an issue of changing the data
collection form or who’s actually entering that data, it’s a matter
of also allowing that data that is captured to be continued to be transmitted
through the data system so that it can be used for quality improvement at
various levels.

DR. MAYS: That was actually my point in the sense that it gets collected in
the system but because of privacy and confidentiality it doesn’t
necessarily mean that for example let’s just say that my physician is a
part of the UCLA Medical Center and in that first encounter they ask me and I
give it to them, doesn’t mean that even though I’m still going to be
treated by that physician that if I come rolling in the emergency room that
they would know, they would know I was African American but let’s just say
that for example they couldn’t tell whether I was American Indian, Latino,
etc., and it’s like either it’s what I say at that point or they just
operate in the dark until somebody else shows up or somebody gets some
information. That’s the point which is we don’t know how many
barriers there are in terms of the collection of that data to be passed on
throughout the system. I mean the person probably would like you to collect it
once and leave them alone, not every encounter they come in, they go to the
pharmacy, what’s your weight, what’s your race ethnicity, what’s
your, and then they go in the emergency room and they get asked, they think
that you have my records, that’s how some people think, you have my
records like why are you asking all these questions again.

DR. LENGERICH: I guess I was thinking about it in somewhat different
purposes, I had been thinking about this in terms of the health care quality
for a particular plan or hospital and then that, if the data are collected it
can go along that track and be used to monitor and possibly improve health care
quality. If it’s already collected it can also go along the track of being
used more by larger systems, administrative health surveillance data systems, a
little bit further removed from actually improving the health care of it so
I’m sort of seeing it as it should be collected at one place and probably
can be but then it needs to be maintained to both of those purposes.

DR. CARR: Just taking this again to a different level, if the question is
the quality of care the answer will be different in different settings and
I’ll give you a personal example, my husband’s father is Irish
Catholic, his mother is Russian Jewish, and my family is Irish Catholic on both
sides. And so my obstetrician made a point when we were expected our first
child that my husband’s family get tested I think it was for Tay-Sachs or
something like that because he said when your children going forward they might
lose sight of that that genetic heritage.

So for example if one of my kids who more or less identify themselves as
Irish Catholic as sort of the way they’ve been brought up but yet if they
went to an oncologist or a hematologist it would be important for them to say
that that’s part of their background because they might have factor 11
deficiency or a gene for breast cancer or any number of things that would be
associated with that heritage. So a quality encounter would be identify that
part of their genetic makeup or whatever that would put them at risk for those
diseases that someone might not otherwise look for. So that would be one kind
of quality encounter and to the extent that you line that up exactly right and
look for the right risk factors you would have had a quality encounter.

But at another level the question might be something different and their
answer might be something different. If you just stopped them at the front desk
of the emergency room, whatever the question was, they would have just a
different answer or a simpler answer. So from a quality standpoint I guess the
question is define quality.

MR. LOCALIO: Along the lines of quality I’m concerned about something
far worse and that is certain data put in as whatever and what are the
provisions for auditing records to make sure that these tasks have been done
with some degree of accuracy. Sandbagging is all too common in terms of data.
Even in the best of circumstances things get messed up, my specific example is
a study in which gender was collected simultaneously from the medical record
and from an interview in 3,000 and in nine people the genders were, how shall I
say, in disagreement. This is a real problem, luckily we had the names of these
people and it was relatively easy to figure out which was correct but there are
many opportunities for the quality of the data to go wrong even in the best of
circumstances and there are even more opportunities for quality to be bad when
there are incentives to have every blank filled in but not the, not financial
or other repercussions to getting it wrong. At least in the diagnostic data I
think there’s some auditing mechanism is there not under Medicare and
Medicaid?

DR. MAYS: I think so.

MR. LOCALIO: And certainly I think the HMOs or the managed care
organizations have some auditing function but I’m not sure, what are we
going to audit here and if we found out that all this data, or a lot of these
data in certain areas we collected were bad what would do.

DR. HOLMES: I think that your point is well taken and that that again
suggests that part of the hearing that we propose and plan on the gathering of
race and ethnicity data have to deal with establishing a case, a business case,
a quality improvement case for the importance of gathering data on race and
ethnicity and coming to agreement about what exactly should be collected,
whether it’s at a minimum and can vary from state to state or data
collection to data collection. But a key beginning point is simply being all on
board across many different health care organizational types that it’s
very important and we make the business case for why it’s important to
gather these data on race and ethnicity to ensure that they ultimately are good
data.

DR. STEINWACHS: I don’t know whether it’s helpful but as
we’ve talked it seems to me there are at least three levels we’re
talking at, one is race ethnicity as part of actually making patient care
decisions. And the second may be as you think about the quality improvement
processes within organization, like hospitals now I understand are supposed to
all have accredited some system for capturing medical errors and if you want to
address patient safety there’s probably every reason to believe that there
are probably disparities there too, is that people of minority ethnic and
racial background may be at greater risk of errors and how that is actually
captured and looked at. And then there’s what AHRQ has to do is produce
the annual report on health care disparities and quality in the nation and so
some of these like Russell was talking about, the patient care level,
there’s a different level of accuracy that’s necessary then if you
think of statistical applications where you may be still concerned but not
quite as concerned. I guess my question is are we trying to look at these in
hearings and then maybe have different groups address them at different levels
or is there a reason maybe to focus on one or two.

DR. MAYS: What do the Quality people think?

DR. HOLMES: Well, my understanding of the purpose of the Quality Workgroup
is to say what data improvements can we make to support quality improvement and
I think even the reporting at the national level at the level of the Quality
Report and the Disparities Report, there are several different purposes, I mean
one is to report at the national level but another is to be able to report at
lower levels to give guidance to particular regions and localities in terms of
specific quality improvement initiatives. So to me the issue of gathering, of
creating a business case or creating understanding of the importance of
collecting data on race and ethnicity cuts across all those issues, and then
secondly is the issue of how to collect the data and we’re dealing with
different kinds of data collection systems, we’re dealing with surveys,
we’re dealing with administrative data, if you’re talking about
hospital patient safety you’re talking about say surveillance systems like
CDC’s National Nosocomial Infection Surveillance System. And so I think
that the issue of race ethnicity cuts across all those issues that Don has
raised and that we should deal with it at a most general level to establish the
importance of reporting on race and ethnicity as how it should optimally be
collected and what should be reported at a minimum, for example whether
reporting can be commensurate with the ’97 OMB requirements at a minimum.
To me those are the key issues to get started.

DR. CARR: One thought I have, as we struggle with coming up with these
definitions maybe a value added element would be the development of ways to do
accuracy checks, sort of a crosswalk of different datasets that would have, it
might be hard to do but the concept is there, that could identify how often
there’s a change in what’s reported in one setting versus another.
But as I say I realize it’s hard to obviously in the absence of a uniform
identifier, but I think the point about how often is there a meaningful change,
is it five percent of the time, or 25 percent of the time, I think would be a
very important contribution even as we are working with the datasets that we
have today where wrong conclusions potentially could be drawn in the absence of
understanding how reproducible the answers are.

DR. HOLMES: I think we need to remember too that there have been some talks,
and I think, I’m sorry that Marjorie Greenberg is not here today, that as
far as gathering race and ethnicity at the administrative level, administrative
types of records, that probably one efficient way would be to gather at the
point of enrollment in a health plan and that that data should be then linked
to subsequent medical records. Which is fine as far as it goes but obviously
that doesn’t capture people who are uninsured in terms of capturing race
and ethnicity and you’re back at gathering it at the point of contact. But
that was one way to deal with the issue of well is it collected every single
time that the individual has a contact with the health care system, that is to
collect it once at the point of enrollment and link the records with other
administrative data.

DR. MAYS: Again I think where it might be useful is to be guided by either
an audit or some research on it because I have a feeling that when you think
about it, when a person is sitting down with their health care provider and
they’re being asked this question they probably sit down and really think
quite carefully about their ethnic racial lineage versus sometimes on a form
where what they’re worried about enrollment is whether or not they’re
going to get discriminated against. So I think there’s a set of very
interesting questions we need to either recommend that someone collect that
data or we try and figure out if there’s any way to look at some of this.
And Eugene and I were just talking about it earlier today but there’s
actually a program announcement that NIH just put out and Justine I don’t
know if you’ve seen it, if not I’d be happy to send it to you, but
they’re actually asking for applications to look at a lot of these issues
from a research perspective, it’s a PA looking at race and discrimination
in the health care encounter specific to the delivery of health services. So
let’s hope that maybe someone actually pursues some of these issues and
can provide some answers in the next year or two after doing this.

DR. CARR: Yeah, it would be helpful to align what are the questions that
need answering with funding for the studies that are being done.

DR. MAYS: See it might be interesting because people are going to crawl,
individuals that are associated with that particular PA and it might even be
useful for us to share with them our discussion so that as they’re trying
to think of, people call, they have ideas, they discuss the ideas with them,
that they can just know that these are real live problems that are looking for
a home. So that might be something also that we might think about doing is
sharing that, well, we’ll talk with Virginia Cain about that since
she’s our NIH rep and have her —

DR. LENGERICH: She’s one of the technical people who’s not here.

DR. MAY: Exactly, I was just going to say, so to have her actually send a
little memo to them about this might be useful, I think she’ll be here
this afternoon.

Okay, let’s try and recap what we want to do which is it’s very
clear in terms of these candidate recommendations about race and ethnicity that
we should have a hearing. I think that we in that hearing want to try and
discuss, this is actually not a one day hearing it’s very clear, but we
need to make a business case for example about the importance of collecting
data on race and ethnicity and we probably need to make that business case
relative to where and when it’s collected. We need to learn a bit more
about who typically does this and kind of what some of the burdens are
associated with the collection of this data. Critical to all of this is what
definitions are going to be used, we can talk about OMB guidance 1997 but it
doesn’t mean that all groups are doing that, I mean those individuals who
have to report data to the federal government probably do it but it
doesn’t mean that necessarily the privates do.

So those are I think some of the issues that we need to think about, I think
the context in discussing those has to do with kind of quality improvement,
health care disparities, and maybe even patient safety as kind of like well
what’s the broader context of why we’re asking that. And then who we
want to hear from, I think the who we’ve talked about involves both
administrative data, as well as research data, as well as data that’s used
for evaluation so now we’re talking about NCHS, we’re talking about
AHRQ, we’re talking about health plans. We’re also talking about this
at a state level because there are some differences in terms of the state
particularly when we’re looking at the hospital discharge data and we
should probably have major organizations like AHA and AMA participate and
there’s some discussion about maybe from the pharmaceutical side having
pharma come in. So that’s our who side of it.

Anything else that to contextualize, I mean I think this gives us a clear
direction, is here anything else that I’ve missed in terms of like the
broader picture of what we want to do in terms of this particular hearing?

DR. HOLMES: I think you did a great job summarizing the first of our talk
today.

DR. STEINWACHS: This is Don, I think you did a great job too, just one
random thought which may not fit into this hearing is that sometimes we look
for the government to try and play a leadership role in some of these areas so
the VA is the system that has patient records and one of the questions might be
is how are they handling the capture and application for quality improvement of
race and ethnicity information at the clinical model and CMS, both for Medicare
and potentially Medicaid, same sort of issue. And so at some point it would be
good to sort of see is our government agencies really providing leadering.

DR. CARR: Then the other thing, and yes, I add my praise to your excellent
summary, I think this is a really wonderful clarifying, but would we also want
to incorporate in there sort of a research agenda saying what would be some
mechanism for checks and balances. I think the one that was mentioned at the
last meeting was looking at birth certificates versus I don’t know, some
other thing, and finding the correlation there. But I think understanding the
magnitude of the difference and the type of the difference is an important tool
for everyone to have, or at least some frame of reference as folks take the
data as it exists today before the definitions are decided upon but at least to
put that as a research agenda that we understand how far off the mark we might
be in any given situation.

DR. MAYS: Let me ask Julia a question about that and Julia, do you know
whether or not that you all have data, the reason I’m thinking about this
is sometimes we can actually make requests for something to be done of an
agency and as opposed to doing this in the theoretical sense it could be that
there’s intramural funds at NIH, there’s I assume the same thing in
terms of NCHS and AHRQ. Is this something that if we ask, for example NCHS, do
you have a way, it’s interesting because the rest of us can’t get to
this data the way you can, this is internal, would you have a way for example
to look at for example your administrative data and even look at your birth
certificates, etc., and get a sense of what the difference is in the reporting
of race and ethnicity?

DR. HOLMES: Well, that’s a meaty question and one that I would have to
ask about although my initial inclination is NCHS is interested in the quality
of data for surveys say for vital statistics because we’re the repository
for vital statistics. And as I said this work that’s been done on the
bridging project, which looks at, very carefully at the way people report when
they’re allowed to report more then one race how they report and then
bridging back to well, how did they report ten years ago when they were forced
to only choose one race. So in that respect we do have some data looking at the
way people report race and ethnicity under different circumstances.

In terms of checking on the quality of data reported say by health plans or
administrative datasets I’m not sure that we’ve ever done that but
I’ll ask.

DR. STEINWACHS: I don’t know whether or not they would have the
capability or not but AHRQ has I guess in collaboration with NCHS the Medical
Expenditure Panel Survey and I know that as part of that there are follow backs
sometimes to insurers, payers, and so on, and they might have the opportunity
to compare what comes through the interviews with what comes through health
care data sources.

DR. CARR: That’s a good idea, Don.

DR. MAYS: Okay, here’s what I’m going to do because Justine I
think, I kept when you were talking about was it five percent, is it 25
percent, I could kind of see research ideas popping up. I think what we want to
try and do is ask, and if we do this far enough in advance of even the hearing
we might ask whether or not at an intramural level we can get some idea of this
issue. So I’m going to ask, I mean all I can be told is no, I think what
we want to do is make a formal request of NCHS, a formal request of AHRQ, and
ask them whether or not this is something that they from an intramural
perspective, or I mean I should try and micromanage, they may ask one of their
funded people to actually do this, but whether or not for example that they
could provide some information on this because I think that that would be a
good start for a business case and it may be that one just has to ask to see if
they can get it. Justine, I think that was a great suggestion about the
research agenda, I think we should keep it as part of what we want to have as a
research agenda with these ideas. But let us see if we can actually get some of
this information now rather then later.

DR. HOLMES: Is your question specifically on the accuracy of reporting of
race and ethnicity on administrative health records as opposed to the accuracy
of reporting of race and ethnicity across all kinds of data collection, most
specifically in our case, NCHS’s case for surveys?

DR. MAYS: I think that each agency might have a bit different question and
so I think that for NCHS that the question, that’s your question —

DR. HOLMES: About the accuracy of race and ethnicity with respect to
surveys?

DR. MAYS: No, not the broad, I’ll get a note if it’s a little too
broad, but let’s go to the administrative datasets. So within a context of
administrative datasets whether or not there is either knowledge about
circumstances or there’s knowledge about what difference, what is the
magnitude of difference in the change in the specification of race and
ethnicity. See one can be a simple question and that is to merely look in
administrative data, well nothing’s simple, anyway, to look in
administrative data and to figure out if you have the capacity to determine the
difference of change and who tends to do that.

Then the other question I think is of greater magnitude, which would take
more time, resources, and people power, and that would be what are the
circumstances under which those changes occur but you actually may have that
from your bridging project.

DR. HOLMES: Well, as I say the bridging project has to do specifically with
the change that occurred as a result of the OMB 1997 allowing people to report
more then one race because before people were forced to make a choice and when
we went back to look at longitudinal data the question became okay so for
people who reported multiple race beginning in 2000 what had they reported back
in 1990 because obviously as you well know in terms of doing longitudinal
studies of change you want to know whether the change took place because of
real change or whether it was simply the different in the reporting. And so
NCHS has just completed a huge bridging project that gives them some every nice
studies but it only dealt with that issue of the change from multiple race from
only being able to report a single race.

DR. MAYS: And I think that that would be helpful particularly since
you’ve already done the work in terms of maybe sharing that.

DR. HOLMES: That we can certainly give a data presentation on that,
certainly.

DR. MAYS: Okay, let me try and wrap this up and see if there’s anything
else, Russell you have any comments, I wasn’t sure if you were trying to
get my attention. Any other comments, questions, anything else Quality that you
all would like to see so that we kind of know here on Populations what we need
to do and how to work with you?

DR. STEINWACHS: You’re doing good.

DR. MAYS: We’re just getting a very full plate here but that’s
okay.

DR. CARR: I think this is really, we’re grappling with some sort of
nitty gritty sort of concrete elements and a huge problem but I really applaud
your efforts, I think this is going to be very helpful and a step forward in a
long journey but a first step forward, second step forward.

DR. HOLMES: I have one remaining question, I’m a little confused, did
we decide we’re going to do the hearings together or you’re going to
hold them or what?

DR. MAYS: I’m sorry when I said that it meant that I need to go back to
Populations when we have our breakout and have a discussion but my assumption
is that this a joint effort, there are parts of the effort where we will look
at materials that we already have that we think we can bring in as information
and then on the other side there are groups, processes, etc., that definitely
you’re much better poised to comment on then we are. So no, I think this
is a joint effort so now that we have a sensity(?) effort I think the next
thing to do is to work out the process by which to do this, whether we’re
talking about taking members from each of our subcommittees, whether we’re
talking about the full two subcommittees trying to hold these hearings,
we’re talking about which staff on each side should participate. I think I
outlined a little earlier that I think when Quality is meeting the Populations
staff should probably be involved and vice versa so that we can stay organized
around this thing. So I think now the next step is for each of us to have a
discussion with our respective groups and at the breakout meeting I think what
we want to try and do is to see when we’re all here how we can work it
out.

DR. HOLMES: Sounds good.

DR. MAYS: Great. Well, I want to thank you all for your time, we’re
going to actually take a break of about ten minutes and so those of you who
want to stay on for mental health either you can mute or go off and come back
on but I probably should give people at least a ten minute break because
I’ll be the hated person that contributes to bladder problems if I
don’t. So I’d like to thank those of you who’ve been online with
us in terms of the Quality group because this has been very useful and I agree,
I think that as a team we might be able to see if we can bring some light to
these issues and make some change occur. So thank you very much.

[Brief break.]

DR. MAYS: Do we have people on line?

DR. STEINWACHS: Don is online.

MS. EVELYN: Brenda’s still here.

DR. MAYS: Oh, hi. Okay, we’re ready to get started again. Let’s do
a few introductions because we’ve had some additional people join us and
they’re going to be integral to the discussions so we definitely want to
do that. Cille, can we start with you?

DR. KENNEDY: What kind of an introduction would you like?

DR. MAYS: Tell us who you are, what agency you’re from.

DR. KENNEDY: Okay, I’m Cille Kennedy and I’m in the Department of
Health and Human Services in ASPE from the Office of the Assistant Secretary
for Planning and Evaluation.

DR. MAYS: Thank you.

DR. DRABEK: John Drabek, I work in ASPE with Cille.

DR. MAYS: Welcome, thank you.

Agenda Item: Discuss Future Directions for Mental Health
Statistics Portfolio – Dr. Mays

DR. MAYS: Okay, let’s move to our agenda item on mental health and Don
we’re very glad that we have you online for this because we know that this
is near and dear to your heart but also one of your areas of expertise so
we’re very happy to have you online.

DR. STEINWACHS: Vickie I’m happy to be back online here and you can put
me to work.

DR. MAYS: You remember that phrase, be careful what you wish for, but okay,
I love to hear that.

DR. STEINWACHS: You know how some of us are, we’re motivated by guilt
and I’ve accumulated a lot of guilt over the past number of months when
I’ve been feeling less NCVHS and more grant writing so you can cash in the
guilt.

DR. MAYS: Well I’m a big believer in that, I have to respect that
people have to write these grants so it’s like dropping out is okay
because we need you to not be out on the street panhandling or anything,
let’s make sure we take care of our livelihoods.

Okay, just to recap and Miriam I’m going to actually ask you since
you’ve pulled together the notes for some of the cause that we have and I
think Don this will be the best way to kind of bring you up to day. I know
Miriam sent it to you but it will probably be good for all of us to kind of get
a sense of some of the questions that we’ve talked about.

The start point for some of this was the NHANES meeting where I went and I
was a little concerned about what was being dropped and it really led me to
this broader question of what are the ongoing statistics that we collect on
mental health and what should be a consistent set of statistics that we collect
on mental health. And the second big issues, which I think we as a group began
to talk about, is the relationship of mental health to physical health, I think
physical health is quite often very prominent in the federal agenda and mental
health is kind of separate and what can we as a committee do to try and bring
these two things together. So I think I’m going to let Miriam kind of
describe the two things that have taken plays.

MS. GRANTHON: There was a conference call set up for August 16^th
and that’s a hard copy that you have in your folders and Don, I shared
with you electronically just yesterday, there were several people invited
across the agency really to get an understanding of what mental health
statistics we are collecting within the department and what was understood is
that we really don’t have such an inventory list and actually that was
something that was proposed in the conference call of trying to making sure
we’ve got all the partners. In our follow-up calls as you see SAMHSA
wasn’t represented but there were several colleagues from the National
Institute of Mental Health Statistics but were, but possibly creating and that
could be a wonderful resource that could be an output of this workgroup.

Then something that was discussed is on page three, what questions to
develop to guide further thinking, what data is being collected, who’s
actually collecting, how often is it being collected, and what link is it to
making people healthier.

There was also additional discussion at the September subcommittee breakout
meeting, this was one of the agenda items, and just getting other participants
involved but again, to see where we are, people on the call knew perhaps what
their institute or office was doing but didn’t have an understanding
across the agency for what was being done and again of tying this physical
health to mental health wellbeing, making that connection. So we’ve only
had one conference call and we’ve had a small part of the agenda to
discuss a little bit but you can read the details in here.

And also the group wanted to get a concrete understanding, what is the
charge for this workgroup, it was suggested of doing this inventory list but
what are we going to do in these next couple weeks or months and what’s
going to be our output, what are we trying to achieve. And also again the idea
of identifying key partners to invite for our next meetings because sometimes
we would ask questions and we wouldn’t know because we didn’t have
all the key people involved in the conference call.

DR. MAYS: Great, thank you very much. So I think what we want to do is by
the end of our time discussing this issues is to decide what we want to achieve
on a very concrete level. I think we want to decide the process to achieve
that, and who we need to work with us to achieve it because clearly we need to
add some partners to this.

But before we do that I think that this whole concept, oh, Don, do you have
any questions?

DR. STEINWACHS: No, no, I think what you did last time was excellent as a
start.

DR. MAYS: I think that what we should do with some of the time that we have
is we really need to discuss what is mental health, I mean that sounds kind of
silly but it actually is not, it is probably the most critical question. When
we talk about physical health I think people kind of go to very specific
things, they think of specific diseases, but when we talk about mental health,
and this has just been, and again I may be really now talking about a state
perspective, but it’s like sometimes the mental health includes substance
abuse, sometimes it doesn’t, there’s some points at which in terms of
for funding it’s almost like alcohol, drugs would be separate, there are
other times at which they’re pulled together. So when we start dealing
with mental health there are some very I think complex structural issues that I
think have prevented us at times from having a full picture. But I think what
we should do is decide what we’re going to talk about because as I said
depends on who you talk to mental health can include or exclude certain things.

Cille, I’m going to ask you to actually share with us, because I think
the handouts and the quick brief discussion that we’ve had I think can
lead us down some interesting talking points. And there are two presentations,
Don I think they’re being faxed to you so I don’t know if you’re
calling from office or home but Gracie went and faxed you these presentations.

DR. STEINWACHS: I am calling from home, do you want my home fax number?

DR. MAYS: Yes, what is it?

DR. STEINWACHS: It’s 410-321-8642.

DR. MAYS: Great, thank you. And if you could find Gracie and just have her
fax them to that number that would be good.

Cille.

DR. KENNEDY: First I wanted to apologize, Don, do not be alarmed, one of the
sets of handouts has six to a page and it is illegible for us as well, I
didn’t know it and didn’t have time to redo it but the content of the
tables is less important then what the headings say. The point of this on the
other hand if the committee ends up wanting full size, or at least three to a
page, I’m happy to recopy them for you.

I would like to focus first, there are some handouts that are three to a
page that are headed and cognitive and psychological functioning and these are
Power Point slides that I took from a conference that we had at the UN a while
ago actually focusing on disability associated, and our particular group was
associating disability with mental illnesses, mental disorders, mental health
—

DR. MAYS: Can you say why?

DR. KENNEDY: Why?

DR. MAYS: What those things were together?

DR. KENNEDY: Disability and mental health? Well, it was a disability
conference for one and I’m sort of famous for always screaming that
disability is not just associated with physical health and I got my just
desserts, I always have to go and defend that mental disorders cause
disabilities at equal to or greater rate and the fact is that when you know a
diagnosis you don’t necessarily know what disability is associated with
it. And when you know a disability you don’t necessarily know what the
ideology is, even the difference between physical and mental health, you see
somebody who can’t walk and you don’t know whether it’s spinal
chord injury, amputation or catatonic schizophrenia, I mean you just cannot
tell.

So anyhow, that all aside, I had the opportunity to, I really learned a lot,
I chaired the panel, Carol Yaeger or Jaeger, I cannot remember how she
pronounces it, is a British psychiatrist and her paper, and I could not find
copies of her presentation but I perhaps have it somewhere, she delved deeply
into the definition of mental health and we may want to figure out a way to use
her as a resource. And Vince Campbell is down at the CDC and he had done some
work going into national surveys and his area was much more, was on mental
retardation and intellectual functioning.

But it brought to the surface like how we use the term cognitive impairment,
mental impairment, mental health, mental disorders, and in the broad overview
we were looking at mental health indicators and at that point we were looking
at international surveys, what the term cognitive functioning meant, and then
how do you define mental health. And mental health I think the knee jerk
reaction is either to define it in our country according to the DSM and other
countries more the ICD but then we got in this, Vickie’s interest in
positive mental health is what sparked me to think of this, is there are a
group of people more interested in the more positive aspects and how do you
define positive mental health. And it was proposed that it was positive affect
and personality traits, self esteem, sense of self mastery, coherence, self
efficacy and resilience. And what were the things that defined negative mental
health, the mental disorders, symptoms and problems, diagnostic and sub
syndrome and psychological distress. And you will find all these kinds of
things in different surveys, and what’s the absence of mental illness.

At any rate there were proposed the consensus mental health definition, we
were sort of forced into a shoebox and to, I mean I’m reading these things
Don because I don’t know if you’ve received the fax so I apologize to
people here who clearly can read but there are one, two, three, four, five,
six, seven, eight, about 15 or so consensus mental health items that could be
used.

And then we got into a discussion of the global indicators of mental health
and then operational definitions, because we had had Vince with us so
operational definitions of specifically for example mental retardation which is
one specific subset of a mental health.

DR. MAYS: Can I stop because I want to ask you a question. On this slide
that says cognitive and psychological functioning, this is a consensus mental
health definition. So just walk me through, this defines like if you have these
you have a mental what, you don’t have a mental disorder, you have a
mental what? Problem? You have —

DR. KENNEDY: Well, it was more a deficit in functioning, but that they would
fall under the great rubric of mental health and then these were more
operational definitions that people agreed on but your question still goes to
the point that was not resolved, but that they all fall within the scope of a
mental health issue be it positive or negative. And you could either be free of
psychological distress or have psychological distress.

DR. MAYS: It’s the functioning part, cognitive and psychological
functioning is really the big —

DR. KENNEDY: Make or don’t make suicide attempts, but that they would
fall within a rubric and then you could define the health or the lack of health
if you will.

DR. MAYS: Let’s just quickly tell him what’s on here. Sleep
problems, role limitations, social functioning, social isolation, chronic
stress, life events, self mastery, optimism, cognitive functioning or deficit,
dementia or Alzheimer’s, suicide attempt, alcohol and drug dependency,
anxiety and depression, positive mental health/psychological well being or
psychological distress. So I think that would give you a sense of what
we’re looking at.

DR. KENNEDY: Right, and what becomes obvious when you look at this list,
mean this is not a list that I would design or necessarily put out there but it
was more a consensus of what exists. But you had things like role functioning
which are clearly not mental health or lack of health per se, that’s more
an area of disability that your mental health may affect but if you have a
deficit in it it is often associated with mental health with social isolation
being likely, that those are some of the swellire(?) associated problems. So I
am not putting these forward to you as definitions but what people have
discussed or what have been included in surveys if you will.

I don’t want to go through all these slides so I’m going to put
them to the side if you end up having any questions about them. The virtually
illegible mental health and disability slides six to a page come out of a
presentation about a chart book that info used to put out for the National
Institute on Disability and Rehabilitation Research. And the reason that I
photocopied them is this was the survey again of what gets assessed in
different surveys, the point being that here it flips between disability and
mental disorder as if they were synonyms and sometimes disability along with a
diagnosis becomes a measure of severity and I’m just trying to show you
all the different ways that people are using the terms and all the different
terms that get used, often synonyms and often not and why it is very important
to talk not only about the positive and negative aspects of whatever one would
want to consider mental health but what are all the synonyms and what meanings
do they have and my best and favorite example is cognitive impairment, is a
term that for the longest time was a synonym for mental retardation. That is no
longer true and is now in fact a synonym for Alzheimer’s and dementias. On
the other hand a cognitive impairment can be a functional deficit of the
cognitive aspects of your mental function so you want to be very careful as you
think about these things.

And then the third handout I had I pulled out of, it’s a narrative done
that I pulled out of the part of a project where we were going to go and try to
develop a concise set of measures to identify the population in national
surveys of people, of the subset of people that would consider have a mental
disorder. And so again we were trying, the background starts to try to tease
out some of these terminological issues, some of the definitional issues that I
think we’ll get into deeper here. The punch line is is that what measured
best in this concise set of measures is Kessler’s K-6 which is currently
in the National Health Interview Survey after looking at a whole lot of other
ways of measuring it but we were looking at how well defined and how you can
bring, identify that population.

DR. MAYS: I’m going to ask two questions, one is at the level of NIH
since there is an NIMH is there a definition of what is mental health so that
they know, it’s very specific what the mission is? And number two, and
actually I don’t want to credit for this, I’m raising this because
you did, do we have specific mental health surveillance in Healthy People 2010?

DR. KENNEDY: The best I can do is a semi-long winded more articulate way of
saying I don’t know. At the NIH level I do not believe there would be a
definition of mental health. What I would suggest and have not done is to go to
the NIMH website and look at their mission statement. But what I do know is
that NIMH specifically does not have a definition of severe mental disorders,
and this is purposeful. As close as they come the National Advisory Mental
Health Council in 1997 did at the request of the Senate Appropriations
Committee as part of health care reform were looking at sort of the costs of
severe mental disorders and could they be treated and the like and for the
purpose of that study there had to be a definition of the population that you
could operationalize to do the study. And the definition escapes me in its
exact phrasing but it had to do with diagnosis and duration and disability laid
out and then operationalized. And that’s severe mental disorders, or
severe mental illness, excuse me.

Then of course there’s the SAMHSA definition of serious mental
illnesses that is legislated in one of the acts if you will and so the acronym
SMI gets bandied about as either serious mental illness or severe mental
illness and they have two different meanings for different two federal
agencies, one of which is legislated and the other was proposed by the Senate.
That was your first question, your second question —

DR. MAYS: And Miriam may even be able to answer, is Healthy People 2010 —

DR. KENNEDY: To my knowledge there is no trace for, you’re asking about
a single question or a way of capturing mental health?

DR. MAYS: Whether or not Healthy People 2010 there is a requirement to,
I’m sorry, not requirement but target goals in terms of any mental illness
or mental —

DR. KENNEDY: There would not be at the generic level but there may be
something there for suicides and they have some things for children but I have
to tell you it’s been a while, I don’t remember.

MS. GRANTHON: We actually had a, held a progress review on mental health
less then six months ago and actually did give a briefing book to Dr.
Steinwachs during the summer meeting but I would have to look up each one but I
do remember there is an objective specifically for suicide, but I know,
actually one of the things that’s really difficult with that chapter is
that they also have lots of developmental objectives —

DR. MAYS: A lot of what?

MS. GRANTHON: Developmental objectives —

DR. MAYS: So it’s mainly around kids.

MS. GRANTHON: No, there were some adult objectives specifically target, the
one specific in suicide was adult and also there was an age group for
adolescents.

DR. KENNEDY: Developmental means the objective is not being monitored,
it’s not developmental —

DR. MAYS: Sorry, I was being a psychology, sorry, thank you.

MS. GRANTHON: That’s what I meant, there’s not a data source and
what we’re going through right now is through updating for our mid course
and meaning if it’s a developmental objective and there wasn’t a
source already identified that it’s going to get dropped and so that it
wouldn’t be a goal that would continued and continued to be tracked. What
I can provide though for everyone electronically is Dr. Sondik’s
presentation at that last progress review and I do have extra briefing books
from that meeting whoever would like one but I did provide that for Dr.
Steinwachs, but I can do a follow-up for that.

MS. BURWELL: In the suicide track under the injuries, intentional injuries
—

MS. GRANTHON: We have a mental health chapter and I think it is —

DR. MAYS: Wow, just a little sadder, I’ll be blue by the time I leave
today.

DR. KENNEDY: Let me turn it around, the reason not to define from the
government’s perspective, say from NIMH’s perspective, the reason not
to define is the point is not to limit the potential for researchers to
investigate different areas.

DR. MAYS: No, I was more talking in terms of like 2010, I mean that’s
where a lot of funds go in terms of programming and so if we’re really, I
mean the research side I think is probably, it’s good on the research side
but on the other side it probably means that while we research and find all
these issues that’s what happening is there isn’t necessarily
programs being put into place. So that is a little concerning, I guess I never
realized that. Don, what do you think about looking, I guess since you even got
the materials, what do you think about our need to look maybe at Healthy People
2010 in terms of the mental health section?

DR. STEINWACHS: Well, I think that would be good to do, as Miriam was
describing it seemed to me that 2010 isn’t probably collecting much but we
ought to talk about what the objectives are, those for which there are data and
those which there are not as part of the needs for why we need to improve the
collection of information either on mental illness or on mental health.

DR. MAYS: Miriam, in Healthy People 2010 is there, and again, I really feel
like I’m on my own little soapbox, is there anything in there about, do we
care about whether people are happier or people have good well being? It’s
like it’s so significant to health —

MS. GRANTHON: I think that was, that might have been one of the
developmental, well being, but again, because data isn’t being collected
it’s going to get dropped. But one of the things that we can work with the
subcommittee is next year, we’re going through this internal process right
now with all the offices in terms of what memos, what objectives are remaining
and which ones are getting dropped and next year there’s going to be a
public comment period and what this workgroup specifically can also do is look
at which ones as Don is saying are being tapped and which ones are getting
dropped and making that push that we still need to be able within the
department, although there isn’t an objective for it, that needs to be
included in terms of the research of the programs for data collection.

DR. MAYS: Suzanne Haynes just handed me the mental health and mental
disorders Healthy People 2010 objectives. Goal, improve mental health and
ensure access to appropriate quality mental health services. So mental health
status improvements, so it’s suicide, adolescent suicide attempts, serious
mental illness among homeless adults, employment of persons with serious mental
illness, eating disorder relapses, treatment expansion, primary care screening
and assessment, treatment for children with mental health problems, juvenile
justice facility screening, treatment for adults with mental disorders,
treatment for co-occurring disorders, adult jail diversion programs, state
activities, state tracking of consumer satisfaction, state plans addressing
cultural competence, and state plans addressing elderly persons. So that’s
the Healthy People 2010.

So I’m going to suggest that definitely we need to do some surveillance
of that particularly if you could make sure that we understand kind of the
backwards process, like if you have a sense of about when you want public
comment, when you need it by, and we can work backwards to make sure that as
we’re planning that we’re prepared at that point to comment.

DR. HAYNES: I think they’ve already asked for public comment —

MS. GRANTHON: No they haven’t, we haven’t.

DR. HAYNES: For the mid course review?

MS. GRANTHON: No, we haven’t, we are still within the department going
through each agency, we’re having a steering committee meeting tomorrow,
and then we have another in December for appeals. What each agency has gone
forward is to try to retain their development objectives and next year when we
have a draft document then we’ll go out for public comment.

DR. HAYNES: After they’ve been dropped though, or after they’ve
been suggested to be dropped.

MS. GRANTHON: Right, exactly.

DR. HAYNES: Yeah, but this is sort of key time —

DR MAYS: To do it before they’re dropped?

DR. HAYNES: Yeah, I would think so. Can the public get involved at this
stage or not? Is it the Department of Mental Health —

MS. GRANTHON: What each individual can do, each chapter has a workgroup and
each workgroup has outside partners, and so what you do is you need to contact
the liaison for that chapter and all these liaisons, work of the coordinators
are listed on the website and I can provide the names and emails, that I can
do. So you can have conversations with the workgroup coordinators but what our
office is doing right now is working with the workgroup coordinators in the
agencies, we’ve got updates, because we’re also asking specifically
for population template by the rural area, specific geographic area, and also
by the racial and ethnic groups, like women wind up at all our steering
committee meetings, so we’re doing that internal now but the public at
this time, it’s going to be invited to provide public comment exactly what
Suzanne just said is after we’ve dropped things already.

DR. MAYS: Do you have a comment about strategically what’s useful?

DR. KENNEDY: I have to say no, I really am not, I don’t know. My only,
I mean because my only interface in the past has been when there’s been a
particular topic that was being developed for 2010 as I got involved in that
one particular aspect it was when I was at NIMH, I have no sense of the
broader, or when they were developing the disability aspect of it.

DR. STEINWACHS: Vickie, this is Don, there was a few years back a working
group part of NCVHS that was looking specifically at mental health statistics.
Have you ever seen that report and what they did?

MS. BURWELL: That was back in the Ron Manderscheid —

DR. MAYS: I was going to say, wasn’t that like maybe ten years ago —

MS. BURWELL: It’s been a while.

DR. MAYS: I vaguely remember it when I first came on knowing about it but I
think, and that was when we, oh, Cille would know, that would be like when we
had actually a separate committee but that was like maybe ten years ago or
something?

DR. KENNEDY: At the very least.

DR. MAYS: But Don maybe we should try and go in the archives a sneeze a
little bit with all the dust on it and see if we can get it.

DR. STEINWACHS: I thought it might be good, I remember David McCannick(?) I
thought was involved in that and some others, because it might provide a little
bit of a foundation we could refer back to but it does seem to me this
discussion is very useful as you try and separate the issues around mental
illness including maybe we want to talk about behavioral disorders, include
substance abuse, alcoholism, and mental illnesses as separate from mental
health which with overlapping circles, and then there’s disability as
Cille’s talking about which is certainly not just related to mental
illness but many other things. So maybe part of our value would be to try and
talk a little bit about how do we clarify where the most important foci are for
collecting data in the near future.

DR. MAYS: I agree, I think that having these kind of three areas, mental
illness, mental health, and disability might be important to think about.
Eugene?

DR. LENGERICH: I’m new to this but let me ask this out of
naiveté please, there are questions that have been developed which get
at depression, self reported depression, that’s administered through the
telephone survey BRFSS on a national basis and I supposes there’s maybe
comparable things in NHIS, I don’t know. So I’m wondering do those
appear, are those appropriate I guess, asking the experts, are those
appropriate in this discussion? And if they are why are they not included in
some sort of surveillance process or discussion? But I think that BRFSS has a
series of five questions about were you depressed in the last month and did it
keep you from working or doing your normal activities, very simple,
straightforward telephone questions, but should those be included in this kind
of discussion?

DR. STEINWACHS: I would think they’re important to include in the
discussion. Part of the challenge is when you look at specific conditions which
ones, but that may be part of what we have to discuss and it may not be
possible to look more broadly all the time.

DR. HAYNES: The question is whether or not it’s a core question or
optional for the states, I suspect it didn’t go into Healthy People 2010
if it’s an optional set of items because then you wouldn’t get it
uniformly in all the states that are doing BRFSS.

DR. LENGERICH: If the goal of this activity is to make recommendations about
surveillance then that might be a candidate one to make.

DR. MAYS: See that’s what I’m saying, I think those are the kinds
of things we definitely want around the table. There’s another level to
that because that’s interesting because that’s probably a mental
health question and not necessarily a mental illness question because when you
just ask from a lay perspective about depression it’s not, I could be
depressed but it was like momentarily versus a person who has a diagnosed
depression and requires treatment. So part of the problem is that some of these
things have really lay terms versus if you say were you ever diagnosed with
cancer, I don’t think well I had cancer today and next week I don’t,
it’s almost like, it’s usually something in which a health care
provider has told me I have.

But in terms of this, do you feel anxious, do you have anxiety, it’s
almost like there’s too much TV now with the drugs, I mean it’s like
people think they have depression and what they have may be symptoms of
depressions, stress, and so then it becomes what’s the important outcome
we want. Is the outcome that you’re worried about the service utilization
so therefore you really want to know if it’s a disorder, is it that
you’re worried about days lost at work so you really are kind of in a
mood, so I think it really, these are just some interesting issues that may be
a little different in the health arena then necessarily in the mental health
arena about those are okay but they’re really screens and if it’s
just surveillance and low level screening we’re okay.

I mean it’s like you bring Ron Kessler in here you’d hear a whole
different conversation about what he thought should be put in which is why
he’s argued for his K-6, he was using the K-10 because as Cille started
talking about the K-10 initially was used because it’s not enough to know
that you have a disorder, I mean we could be sitting, if we do the number how
many out of ten one of us sitting here has a mental disorder probably but it
hasn’t disabled it I don’t think, disabled any of us, so Ron is very
clear about you need to know severity and disability because people could be
very functional in terms of mental disorders, even though people in health
would say the same thing. So we have some really complex, as we’re talking
this is really complex what this task is.

But task number one, which you can just see from just what Suzanne is doing
now, we don’t know what’s where and we don’t, I mean I
don’t know, is anybody the keeper of mental health on a federal level? How
would we go about finding out what’s where?

DR. HAYNES: Well, SAMHSA is the biggest producer of mental health statistics
so you need somebody to come and testify here from SAMHSA really. Because they
have the household survey that asks dozens if not hundreds of questions.

DR. STEINWACHS: Is that Ron Manderscheid then?

MS. GRANTHON: He was recommended at one of the workgroup meetings.

DR. MAYS: Is he at SAMHSA?

DR. STEINWACHS: To my knowledge, yes.

DR. MAYS: Well, see again, there’s the difference between surveillance
and prevalence, and if you say prevalence then you got NIH with these huge
surveys that they’ve done about every ten years, ECA and NCS and now the
re-interview and now the minority ones, but then at the same time it’s
like I agree about SAMHSA.

There’s one task, it’s like that’s a task that ought to be
commissioned out or given or something, is to see if ASPE would be interested
in commissioning somebody to pull together what all the different statistics
are and I think it’s this, I think the issue of mental disorder, mental
health and disability and whether or not it’s surveillance versus
prevalence. They’re conferencing, Don you can’t see, they’re
conferencing now when I said that.

DR. STEINWACHS: Thank you for describing it to me.

DR. MAYS: I know, you had to see the dynamic, they’re shaking heads and
things like that over there, so I must have asked —

DR. STEINWACHS: Vickie, they’re concerned that we won’t live long
enough to see the report.

DR. MAYS: Can we have comments? Russell, what do you think? We need to know
something about what we have I think.

MR. LOCALIO: I have a concern, I don’t know what the answer to this is
—

DR. MAYS: Whether we’ll live to see the report? I’m sorry, go on.

MR. LOCALIO: What do people do about getting information on mental health
for populations that are not usually covered by surveys? Institutionalized
populations, what comes to mind of course is prisoners, a large number of
people in prison today and very large prevalence of severe mental health
problems among prisoners, and where does that information come from in terms
of, it doesn’t come from our surveys, does it or doesn’t it?

DR. MAYS: I actually think that data is collected, I may be wrong, because I
actually started my career in forensics so —

MR. LOCALIO: How does that fit in? One more thing, looking at information as
we talked earlier today about place how does that fit in if some places which
are very rural have large prison populations and according to Census they are
counted at the location of the institution, how do people deal with that when
they’re talking about the prevalence —

DR. STEINWACHS: To the best of my knowledge there isn’t very good
information about mental illness status of people in jails and prisons. There
could be, National Alliance for the Mentally Ill a number of years back did a
survey of people who run jails about the occupancy related to mental illness to
try and get some indicators and there are, potentially much like managed care
there are contract providers who provide lots of the medical and mental health
services in prisons across the country because I had some discussions with one
that’s the biggest one and so there’s probably some potential there
but you’re right, it’s a largely ignored population yet we’ve
talked about the largest mental institution in this country being the Los
Angeles jail, which it is.

DR. HAYNES: There is a Healthy People objective, 18A that proposes
increasing the proportion of juvenile justice facilities that screen new
admissions for mental health problems and there is a reference here that
problems of suicide, self injurious behavior and other disorders are
significant among youths in the juvenile justice system. So I guess the Healthy
People group was considered, at least they could tackle one thing because
apparently there’s an inventory of mental health services in juvenile
justice facilities that SAMHSA monitors. I really think you need to get
somebody from SAMHSA to talk about what they have going on.

DR. MAYS: I agree.

DR. KENNEDY: Don, help me with this one, Hank Stedman and Tom McGuire if
I’m not mistaken have done a lot of work perhaps with prevalence in either
jails or prisons or both, I’m not sure which, but there is something
called the Consensus Project, which is a project out of the Council of State
Governments and I know Hank and his shop have been very heavily involved. Those
are the names that come to mind in terms of potential for prevalence of mental
disorders in these settings but another person whose done a lot of work if
I’m not mistaken in jails, maybe in prisons too, whose gone in and I
believe done diagnostic interview work, is Linda Templin(?) in Chicago. Is this
ringing bells with you, Don

DR. STEINWACHS: Those are the names, particularly Hank Stedman, the other
one I didn’t catch when you said it —

DR. KENNEDY: Tom McGuire?

DR. STEINWACHS: — who would know what the status of information is and I
hadn’t heard about the survey that you were talking about, that sounds
promising.

DR. MAYS: Here’s what I think we should do, maybe this is a better use
of our time because I think part of the problem we’re having is how much
we don’t know. Let’s turn our attention to who all should be at this
table and become a part of a workgroup to help us in terms of mental health
statistics, and I think we may need to do the same thing which is to have a day
where we have the agencies come in because I mean some of these are questions
that I think we may want to elicit some information on, it’s like is there
a guiding definition of mental illness or mental health or whatever at various
agencies. I don’t even have a sense of who has what part of these
portfolio. I mean it’s interesting because you’re right, Eugene,
BRFSS is collecting some stuff but maybe as Suzanne says it’s not
consistent, it’s not mandated, I don’t mean consistent, it’s not
mandated so therefore we don’t have a picture across. I think we need to
figure out in terms of Healthy People 2010 whether not or later is a time to at
least have a place card holder, we need to look at NHANES, I mean I think we
need to have a day of education of what’s there and who should be a part,
so we need to talk about who should be a part of this workgroup. Do you have
any other suggestions, Don?

DR. STEINWACHS: You mean in terms of collecting information or part of the
workgroup?

DR. MAYS: In terms of the direction. I guess I was suggesting that I think
we don’t —

DR. STEINWACHS: I like the idea of including substance abuse and alcoholism
and NIDA does do surveys themselves and collect information so it would be good
it seems to me to try and at least initially in a workgroup broaden out to
include those two in addition to the narrow definition of mental illness.

DR. KENNEDY: NIAAA also does a major survey and I know that they are often
including the confluence of alcoholism with depression, Bridgett Grant was the
person, Bill Cartwright, is he at NIDA?

DR. MAYS: I don’t know about Bridgett Grant, and actually we have two
NIDA people and we probably, one is an epidemiologist, Lesley Cooper, and so it
may be that we just need to have them here. So so far we’ve talked about
having a day where we can talk about definition, the distinctions, and then to
get some idea of the scope of what’s done kind of within those. So we
definitely want to hear and bring to the table, I think the first way we ought
to do this is with the staff from these agencies, we don’t need outside
users yet but let’s just start with the staff. So we should have NIDA, we
should have SAMHSA, we should have NIAAA, we should have NIMH, BRFSS, what
else? CDC, I’m sorry, BRFSS so CDC, who else? Does HRSA do this?

DR. STEINWACHS: I don’t think HRSA is doing much except they certainly
have a focus on primary care if you’re trying to ask questions about
primary care and community health centers and so on.

DR. MAYS: Well see the reason I said HRSA because I thought that there was
some initiative for that that the federal government was pushing for screening
to occur in primary care, so I don’t know whether HRSA has developed an
inventory, I would imagine it’s something that if you’re going to
push this that you ought to have some sense of what everybody should be doing
but maybe we can just ask at HRSA. As a matter of fact the person that we can
ask is Catore Felix-Aaron(?) who just went from AHRQ over there to do quality
stuff so she might know.

DR. STEINWACHS: Another thought in that kind of session is the National
Association of State Mental Health Program Directors, NASHPD(?), might be a
good place to ask someone to come and talk about what are the states doing
trying to get their own information —

DR. MAYS: What’s the organization again, Don, could you just repeat it
so we have it, National Association of State what?

DR. STEINWACHS: The National Association of State Mental Health Program
Directors, NASHPD.

DR. HAYNES: In doing our big women’s health database that’s about
ready to be launched, this was the worst area for us to get data because we
went to the states and tried to collect data by gender and race and the problem
that we encountered that all the data systems are different, there are no
standard definitions of collection, the money goes out through some block
grants at SAMHSA for the services and there’s no standard data collection
form. They do what they want because we really wanted to enhance, this was a
big request that we had for our database to have this information so we ended
up putting stuff up like the BRFSS data and some other things that we could
find but we couldn’t get SAMHSA data because they have it by age but not
by gender for public access and it’s really tough so I’ve been
invited to go talk to the women’s committee at SAMHSA to try to get them
to put out a pubic database so it has men and women by race which is what our
dataset has.

DR. MAYS: You reminded me that we should ask the Surgeon General’s
Office because they have two projects underway, one is the one that started in
your office on women’s mental health, and the other is and I actually
heard Al Noonan refer to this, they’re also trying to do one on mental
health services for jails and prisons and they’re very concerned about the
number of people that come out and that need services. So I would imagine that
they’ve had to do some kind of estimates or something to figure out
what’s needed so let’s ask somebody from the Surgeon General’s
Office, it would probably be Al Noonan because I think he’s the one
assigned to work on this.

DR. KENNEDY: I’m going to bring up something, Vickie, that I think
probably should be rejected but should at least be rejected rather then
unstated, and that is the Centers for Medicare and Medicaid Services obviously
for their administrative data collect information, have information. It’s
not surveillance and it’s not epidemiologic surveys, it’s just claims
data.

DR. MAYS: It’s administrative data.

DR. KENNEDY: Right. But the reason to think about it is D has brought to my,
reminded me that NASHPD was working closely with them and I believe with SAMHSA
to come up with the HIPAA codes that would define mental health use and within
that of course would have to be some way of identifying the mental health
population were they just coding diagnoses. That’s something that may be
too outside the scope but better to have raised it and said no then to have
later on said oh, we should have considered it.

DR. HAYNES: I think it’s a good idea, you only get sort of the elderly
population with Medicare but with Medicaid you get obviously folks who have
been classified in poverty —

DR. STEINWACHS: And you get disability on both.

DR. MAYS: Here’s what I’m going to actually suggest which is
rather then us discuss it and reject it is we discuss it and we make a request
of Standards that they pursue looking at this because its Standards that’s
been working on the HIPAA code, no, who is it? Is it Standards or NHII? Debbie?
Who does HIPAA? Mostly Standards so maybe we should ask Standards to update us
on what has gone on —

DR. HAYNES: Maybe in terms of the definition.

DR. MAYS: That’s what I mean, in terms of what has gone on in terms of
development of the definition, we can give that to Standards as a question and
then I think they can pursue whether there is an issue there, I think that
would be good, so that goes to Standards.

DR. KENNEDY: But one of the things I think is reasonable because we’ve
talked as if there’s a definition of mental health that would come out of
this and I think that we may have to operate on the notion that there are
several —

DR. MAYS: And I think there should be, I think there should be, I mean
we’d have our agencies really upset with us if we tried to come up with a
definition because it would really I think be a problem. Yeah, we’re just
going to take a couple other comments and make sure you all get lunch because
we have somebody coming.

DR. LENGERICH: Just a comment, they probably would address it anyway and you
may ask them specifically but if you’re having these people come it may be
helpful if they identified what are the top two or three items that they would
like to have in improved data collection for mental health to help focus the
discussion or focus our perception so that they can tell us what they think
they need.

DR. MAYS: Okay, I mean I think that’s a good point, I did that kind of
on the phone is to ask people for their wish list and to see what they really
wanted and I think that was, I found that to be a useful exercise. And then I
think the other thing that Eugene in terms of like what else to ask them, in
addition to kind of what would they like to see is is whether or not they
product materials, etc., that have anything to do with mental health or
physical health because I don’t think people are doing that so it would
give us some insight as to whether or not the mental health statistics ever get
combined in any way with physical health.

DR. STEINWACHS: I think that’s very important.

DR. LENGERICH: I think it is too because I think that most people are more
used to dealing with physical health and so that is a way to, a hook to bring
them in.

DR. MAYS: Cille, are there any, I want to just kind of use your guidance,
are there any particular issues that are coming up because where a lot of work
is being done is really internationally right now with WHO and particularly
since WHO has collected all these datasets and they’ve been struggling for
example with kind of how to define say depression over several countries,
there’s a lot of different things they’re struggling with and
I’m just wondering is there any takeaway kind of message that we should
have so that we can work on it, that would be relevant to our charge.

DR. KENNEDY: At this point the government is not working so closely with WHO
—

DR. MAYS: Oh, something must have happened.

DR. STEINWACHS: I shouldn’t laugh.

DR. KENNEDY: At any rate it’s not to say they’re not all working
—

DR. MAYS: Now I remember why, I’m sorry, go on.

DR. KENNEDY: It’s just not as closely if you will and that’s not
to say that a lot isn’t being done, so that just as a tidbit of
information. And the U.S. did not participate I believe in the World Mental
Health Survey but we have Ron’s work which is very similar.

DR. MAYS: Okay, read between the lines.

DR. KENNEDY: Check back next Wednesday.

DR. MAYS: I forgot how close that is but anyway. Okay, anything else?
It’s interesting, I didn’t realize, I thought we could just kind of
launch on this one but I realize how much we kind of need to find out first.

Don, is there anything else that you would want to do in terms of leading
this effort? Anything else you want to put on the plate?

DR. STEINWACHS: No, I think this is a great first step in this direction and
as you said we have to sort through some of the complexity and decide how we
can focus and contribute so I think what has come out of the discussion is
excellent. I’m happy to lead it in any way I can.

DR. MAYS: Okay, great. Alright, so I think what we’re looking at is we
need a day or two days or something like that on this. Cille, you’ll work
with us?

DR. KENNEDY: I’d be happy to.

DR. MAYS: Miriam? Are you going to work with us?

MS. GRANTHON: I’d be happy to.

DR. MAYS: Suzanne, do you want to work with us since this is also kind of an
offshoot of what you are going to be doing?

DR. HAYNES: [Comment off microphone.]

DR. MAYS: And I’m also talking about your women’s mental health so
I think that that would be good, so okay, so we’re well covered in terms
of staff for this one.

Okay, thank you very much, I think what we need to do here is break for
lunch before people die because we do have a speaker coming from the National
Child Study at 1:00 so let’s see if we can get some lunch and then get
back here.

[Whereupon at 12:28 p.m. the meeting was recessed, to reconvene at 1:05
p.m., the same afternoon, October 26, 2004.]

A F T E R N
O O N S E S S I
O N [1:05 p.m.]

Agenda Item: The National Child Study – Dr. Knox

DR. MAYS: — I told everyone we have to behave, we have a guest coming, so
we have a guest and I’m very happy to have Sara Knox with us, I just have
to put a plug in, Sara is a psychologist, she’s with NICHD and has been
working throughout the process so she’s very aware of the different levels
of the process and the updates of what have occurred over time.

Sara, just to bring you up to date, what happened was that this, our full
committee, what you’re seeing now is just part of the Subcommittee on
Populations, but the full committee actually received a report in person from
Peter, I think it was March 2003, and so it’s been a while and kind of
after Peter’s discussion there’s a letter which was up on the website
that we sent Peter about things that we wanted to get a sense of and update and
what have you and I guess after a while when we didn’t get updates and
things what we did was we bugged them to see what the response was and so we
got a response back from Dwayne Alexander which is who we actually sent our
letter to. And so the Subcommittee on Populations is actually the group that
has been tracking this so we thought it would be good rather then starting at
the full committee is for us to get ourselves informed as to what’s gone
on, because we did see the announcement kind of after all the little small
groups were disbanded about kind of the direction that the group wants to take,
and so we thought this would be a good time when we as a subcommittee were
meeting to actually hear the update so that’s what Sara is here to do.

The discussion that we had, Sara actually would like to have a discussion
with us and not spend the time talking at us so she is going to update us but
from her side would really like our input so if there are things that she can
take back to the group and things that would help the National Children’s
Study that we have thought about she’s more then open to hearing those. So
Sara I really welcome you and thank you for being here today.

DR. KNOX: Well I appreciate the opportunity because it’s much better
for us to find out up front if there’s something that’s a real gap in
our thinking then three years down the line when it’s a little too late to
do something about it. So what I did is I tried to put together a shorter
presentation, maybe maximum a half an hour and I like the fact that this is a
small group because you can ask questions as I go along on these slide, you
don’t have to wait until the end. But what I did, some of these slides you
probably have seen before but I wanted to make sure that everyone was on the
same page with the basis because it contributes to the discussion about
sampling strategy, that kind of thing. So forgive the redundancy if you’ve
seen it, it won’t take that long.

Okay, this is our requisite slide on the agencies and so forth involved. Let
me just get started with the rationale and the reason I know you’ve
probably seen this before but it is relevant to the discussion of exposures and
outcomes so I’m going to go through it again. The whole rationale for this
study came from the President’s Task Force on Health Risks and Safety
Risks to Children in 2002 and there are some issues with children that are
different then adults. One of them is that compared to adults children are
especially vulnerable to environmental exposures, now it’s written there
as metabolism and behavior but if you think about the fact that children crawl
around on the floor, if you push then in a stroller they’re right on the
level of pollutants coming out of cars, so there really are differences.

And the exposures to some agents demonstrate potential for serious
developmental effects. We know about lead and prenatal alcohol but there are a
whole slew, there 80,000 different chemicals in use and most of them we
don’t know about toxic effect and only about 28 percent of the ones that
are over 3,000 pounds a year do we have any developmental data at all.

So what we wanted to do, what we were tasked with doing, is looking at known
exposures of high frequency, like pesticides, like violence, like media, where
we really don’t know a lot about the outcomes and then also numerous high
burdened conditions with suspected environmental contributions like learning
disabilities, like autism, diabetes, asthma, birth defects, and premature
birth. And the existing research is simply too limited, it’s not powered
to do that, you can’t look at mixtures and you can’t look at
developmental effects, which is the life course approach.

So basically what the Children’s Health Act of 2000 did was authorize
NICHD to get together with some other agencies and develop a plan for a
prospective study and to look at basic mechanisms of developmental disorders
and environmental factors.

DR. MAYS: Sara, can I just ask you, when it was authorized did it come with
a budget?

DR. KNOX: No. And that’s still under debate, we’re not fully
funded.

So the basic conceptualization that is the study is going to be
longitudinal, it’s going to include not just the children but their
families and environment. It’s national in scope, it’s hypothesis
driven, and these are all things that are important for the sampling strategy.
Environment is defined broadly, chemical, physical, behavioral, psychosocial,
and cultural, cultural could have gone under psychosocial but I thought it was
important to put it separately. And also studying common range of environmental
exposures in less common outcomes, autism for instance is less then one percent
and if you’re going to look at a gene environment interaction you have to
have a huge sample or you just can’t do it.

So it’s also, the exposure period begins during pregnancy for most of
the cohort, however we are going to try and get a preconception cohort
hopefully of about 25 percent where we can look at environmental influences on
fertility and also on early neural development. So it also should involve state
of the art technology and a consortium of multiple agencies, we have an
interagency coordinating committee that’s composed of the CDC, the EPA,
NIEHS, and NICHD. It should involve extensive public/private partnerships and
also a national resource for future studies and that means that the data to the
study should be available as soon as possible after each wave.

So okay, what are the study population issues? Generalizability to the U.S.
population obviously, should include specific high risk populations like
agricultural, industrial, and economically disadvantaged people, and also
non-pregnant women of child bearing age due to the possible effects on
fertility and pregnancy.

Now you’ve probably seen this slide before, the reason I’m showing
it again is because it has to do with power and as you’ll see from these
different outcomes listed this slide was basically put together assuming
dichotomous exposure, which is real simple, high and low exposure, so it gets
more complex the more you look at it. But I want to show you is that if you
take something like autism, which is really rare, obviously you need a larger
sample size then for something like ADHD, so we have a whole range of outcomes
requiring different power.

So the basic concepts of this sampling strategy are that it should be a
national probability sample, that is a center base structure and what that
means is that there will be academic centers involved and the reason for that
is that measures and observation requiring expertise in facilities are best
suited that way. And also because we think that there’s a lot of expertise
and productivity in academic centers.

Another concept is it should have multiple dense clusters and the reason for
this is that we want to look at neighborhoods, we want to look at neighborhoods
from a physical perspective, what kind of buildings there are, how much
dilapidation, we want to look at neighborhood cohesiveness, chemical exposures,
and in order to do that you have to have dense clusters. We also want to be
able to look at what’s going on in schools and if you have just one child
per school you can’t really do that. And then there’s a selection of
clusters, I’m going to get to the hierarchy in a minute, which is
probability or quota by characteristics, so within these primary sampling units
there will be smaller clusters and then obviously there has to be feasibility
tested.

But the sampling strategy is basically a three stage process, first
we’re going to select primary sampling units and these correspond
approximately to counties but if you’ve got a number of counties which are
low birth rates they’re going to be put together so we’re saying
approximately, this won’t be exact, some counties will be combined and
there will also be very, very large things that may be divided, self selecting
clusters.

Then selecting these dense neighborhood clusters is really an issue because
you can’t just decide that you’re going to go through and say X
number of blocks, X number of square miles, because that may have nothing to do
with the neighborhood. And so one of the requirements for funding for this
study of people who come in is going to be that they have a documented track
record of actually working with these communities on a longitudinal basis,
that’s a requirement for eligibility. And so the stage of selecting these
denser neighborhood clusters is going to be done together with the people who
are awarded, who actually know these communities.

DR. MAYS: Can I ask a question? Are you going to specify first where you
want —

DR. KNOX: Yes.

DR. MAYS: And then people from those areas come in —

DR. LENGERICH: Sort of along that line then, so I’m from Pennsylvania
and there’s 67 counties in the state of Pennsylvania and probably about
four or five counties that actually have an academic center, so what about, are
the other counties eligible and included in your sample or are they going to be
—

DR. KNOX: Basically they’re all included and what’s happening at
the first level before the RFP goes out is that statistically primary sampling
units are selected for the whole study. Now the first RFP is going to be
vanguard sites so three to six, something like that, a small number of vanguard
sites. And so we will say these are the areas that are eligible for vanguard
sites and there’s a lot of debate about this, if we should restrict the
people who come in, if you get too many people in people are going to put a lot
of work into it and then there’s so many that half of them won’t get
awarded, and we thought we want the best, absolutely the best people so
it’s difficult for us to imagine that somebody sitting in Pennsylvania
could do Los Angeles but if they can prove that they’re better at Los
Angeles then the people in Los Angeles fine. But basically at the first level
in order to make this a national probability sample we have to select all of
the primary sampling units and then we had to select from that in different
areas of the country where the vanguard sites are going to be so that will be
in the RFP. Then the clusters of neighborhoods won’t be selected until
later.

DR. LENGERICH: I’m thinking of the 60 or so counties in Pennsylvania
that don’t have an academic center in them so will they —

DR. KNOX: This was one of the arguments against the national probability
sample and it didn’t win out and so we’re going to really try and
make the national probability center work and that means that whatever the
areas in Pennsylvania are, whoever qualifies is going to have to be able to
get, recruit within that area. So you’re giving an argument that a lot of
people gave and we had workshops and white papers and a year of discussion on
this and we decided to do a national probability sample.

DR. MAYS: But it’s not that those counties are not able to participate,
it’s that if those three or four academic centers have not worked with
those counties, because they haven’t worked with those counties the
counties are screwed. So if you can demonstrate that you have broad reach,
which in areas, certain areas I think might be the case then the counties
actually get to participate.

DR. KNOX: Well, the counties are definitely going to get to participate, the
issue is if you’re going to get somebody whose been involved in these
national samples before who’s going to come in and say we can do that
better then you can so it’s a competition issue, it’s not a question
of excluding those counties, it’s a question of who can best show that
they’re best able to recruit from them. So it’s difficult, I mean
you’re raising the right questions.

DR. LENGERICH: I’m sure you’ve heard this before but I mean the
communities have to participate, willingly participate, and so you’re
selecting them and then kind of putting them out to the best bidder.

DR. KNOX: We’re selecting geographic regions, that’s right. And as
I say that was the argument for the people who said you’ll never be able
to do this, don’t do a national probability base sample, I mean a lot of
people said that so you’re right, I’ve heard it before, it’s a
legitimate argument.

Okay, at the last stage we select the individual households and so again
this depends on the size of the county whether we screen everybody in that area
or in those small clusters or whether we first do a probability of those
clusters depending on the size.

You have seen this so I’m going to go quickly through it but this is
just to show that environmental exposures are defined broadly. Physical, if
you’re looking at a neighborhood of dilapidation or what kind of
structures there are, chemical exposures, it’s going to differ if
you’re living in an industrial area or agricultural area with pesticides.
And one of the issues in this study that we can deal with with a large cohort
that’s really, we spent a lot of time discussing, is these complex
ubiquitous low level exposures and things also intermittent exposures. So if
you have to be exposed to a pesticide at a time where you’re not being
tested for biological samples and it’s not bio-accumulative we’re
going to miss it and so there’s issues of validation and all kinds of
things.

The biological environment obviously inter-uterine, inflammatory issues are
really big, metabolic responses. This is not a genetic study but genes are
involved because they interact continually with psychosocial factors, with
physical factors, and chemical factors, and this is going to be a big issue.
And with a cohort this size we can use the candidate gene approach and look at
psychosocial exposures. And then of course psychosocial factors.

Outcomes, I’m sure you’ve seen this before. Pregnancy outcomes
will be the first ones and things like premature birth which are influenced not
only by chemical exposures but also by stress, a whole lot of socioeconomic
status issues, neuro development and behavior and here you see we have rare
diagnoses, but I also want you to notice that there is normal variation so we
also want to look at resiliency in healthy development, this is not just a
study of illness, it’s a study of the things that also optimize
development, so injury, asthma, and obesity and physical development.

Okay, so even though no single hypothesis can possibly do justice to this
study there are hypotheses which drive it and the reason for that is that
they’re required one for the costly element and also scientifically to
justify the cost. We have to look at how important the outcome is in prevalence
and severity and morbidity, disability, public health significance, and there
has to be reasonable scientific rationale, in other words why are we doing this
in a sample of 100,000, is it really required. Then it has to also be
measurable with a study of this size, now cancer is not included and the reason
for that is because it’s too rare, I can go back and show you that slide,
we’re trying to get together a sort of a consortium of cohort studies,
there’s one going on in Denmark, there’s some other ones going on in
Scandinavia, so that we can perhaps combine them and get some outcomes but this
study is too small to do that.

And also they have to things that require longitudinal follow-up and the
thing that excites me about this study is that a lot of the exposures that
we’re looking at, whether they’re chemical or psychosocial,
don’t hit until many years later so you can’t expect to see very
often some of the gene environment interactions for instance with psychosocial
factors you don’t see until adolescents, you see it with alcoholism or
substance abuse and so it’s really important this whole justifying it from
a longitudinal perspective. I hesitated —

DR. CAIN: I had a question along the lines with regard to the hypotheses.
One of the concerns that I have heard expressed particularly with regard to
some of the psychosocial hypotheses is that many of them would not necessarily
require a sample size of 100,000, influence of family pick whatever aspect you
want, so there’s a concern that some of those which are hugely important
would be left out because you don’t really have to have a 100,000, but it
would be a shame to leave them out. And also along with that when you start
looking at these various subgroups, whether it be racial and ethnic or some of
the others that you’ve identified such as socio-economically
disadvantaged, you do quickly get down to some very small sample size and
there’s concern that some of those are not going to be included because
they require the large sample size to test issue.

DR. KNOX: Well, there are two questions, the first one I think is a very
good question because there’s been a lot of concern about that. The
hypotheses in this study of course have to be approved by the interagency
coordinating committee and there are a lot of chemical people on that
committee. What has happened since the last time you got a presentation is that
there are psychosocial hypotheses which are now being considered for the core,
in other words they’re being written up, one of them which I did the power
analyses on myself is psychosocial influences on gene expression, stress,
nurturance, parenting, and that is justifiable from a power perspective. Other
ones concern media use, one is healthy development, and so pretty much we think
these are going to be accepted as core hypotheses.

If you think about the fact that you want to look at a chemical exposure,
you want to also be able to rule out other covarying factors so we have to have
a lot of the psychosocial stuff in there because it’s confounded by low
socioeconomic status for instance, physical characteristics of the
neighborhood, so as things now stand there’s a huge psychosocial component
and we’ll see where that goes. You can keep your eye on it but I think as
far as the core I think they’re going to be in there and then secondary
analyses, if we have things like family and neighborhood cohesiveness and all
of those things, day care, school issues, discrimination, racism in the
schools, those kinds of things, then I think the number of secondary analyses
you can do even that don’t qualify for inclusion in core hypotheses are
there. So I think that’s the concern and I’ve had the same concern so
we can see how it goes but I’m feeling a lot better about it now then I
did a year ago.

DR. COOPER: Sara, this is Leslie Cooper on the line, did you say that cancer
would not be included due to the rare event?

DR. KNOX: Yes.

DR. COOPER: Thank you.

DR. KNOX: Okay, I don’t know if we need to go through these because
you’ve probably seen some of this before but this is just examples, low
level exposure to non-persistent pesticides in utero or post natally increases
risk of poor performance on neuro, behavioral, and cognitive examinations
during infancy and later in childhood, among those with genetically decreased
pyroxenes activity. Asthma incidence and severity is associated with early life
experience with infections. Nurturing, stress, and other psychosocial exposures
can differentially influence behavioral and developmental disorders associated
with a serotonin transporter gene link polymorphic, region, the 5HTTLPR, those
are just a couple. And there are more maternal infection inflammation, social
environment, gene interactions, again there’s a whole lot of those.

There’s specific hypotheses, material hypothyroidism, the built
environment and obesity and health, that has to do with whether there’s
places for people to exercise, whether children live in an environment where
they even dare go outside. If it’s too dangerous you stay cooped up in
front of the television because it’s too dangerous to go out. Again,
family structure and function as its influenced child health and psychological
development. We’re looking at healthy development, we have to look at
these issues, endocrine active compounds in sexual development and many, many
others.

I’m just going to show you this really quickly but this also relates to
analyses and that is as you can see there are a lot of interactions. So
basically I’ll say this out loud for the people who are listening, you
have the biological environment, psychosocial environment, physical
environment, and chemical environment all influencing outcomes like asthma,
neurological and developmental, growth and obesity, injury and birth, and they
have direct influences and they also modify gene expression to effect these
outcomes and of course they’re moderated by issues such as health care.
But as you can see you really do require a cohort of 100,000 to look at
interactions here and even if you’re only looking at interactions with
chemical exposures one of the things people probably don’t think about is
when you look at a food additive or a chemical exposure the toxic levels are
really calculated on a single exposure, they’re not calculated on
mixtures.

When I worked in Sweden, I started my career at the Karlinsky(?) Institute,
I was at the Institute for Environmental Medicine and they had what they called
sick building syndrome in their day care centers and they went in and did very
technical analyses of all the air, everything there, and none of the chemicals
reached threshold value. What we were going to do before I left and came to the
United States was actually look at the interaction between stress and these low
level physical exposures. But the issue was they were beginning to look at
patterns of exposure so that if you have more then one chemical exposure forget
all of these other things, is your threshold for vulnerability going to be
lowered by the fact that you’ve got a synergistic event here. So these are
the kinds of things that I think are really exciting in this study that you
cannot do in a smaller study.

So how will the study produce results not otherwise available? And it’s
by looking at big issue, low frequency outcomes that you can’t do in
smaller studies like autism, diabetes, birth defects, etc., and it’s
subgroups and multifactor interactions, obviously obesity as enumerable
interactions with other factors. And so when we actually get the study started
we’re going to be doing environmental samples that include air, water,
dust, a lot of things in the air, biomarkers, we’ll have blood, breast
milk, hair tissue, etc. We’re going to do structured interviews,
histories, serology and medical data, housing and living characteristics,
family and social experiences, and neighborhood and community characteristics.
And again, I won’t go through these outcomes, fetal growth, cognitive and
emotional, developmental, growth, nutrition, birth defects, I mean there’s
a great many more, these are just sort of large.

Now I hesitate to show you this slide because this is not the list of
visits, I couldn’t show you that because this is a procurement process.
The only reason I’m showing you this is this was an original thinking for
the budget so don’t ask me why we’re doing X, why we’re doing
this visit because this is not necessarily what’s going to be in the RFP.
But we had to do, we had to try and budget for a longitudinal developmental
study and this is kind of the way we thought about it.

The use of data to maximize output, results will be available beginning
2010, I’m going to show you the timetable in a minute. Targeted hypotheses
and testing of analysis, so it’s going to be successive public use
datasets and we are going to try to de-identify as much as we can of everything
so that really just about everything is available, this is really supposed to
be a national resource and we expect RO1s to come in, we expect it really to be
a resource so this will happen successively. Again, investigator funding and so
forth.

Okay, here’s the timeline. Pilot studies as you know have been ongoing,
white papers, input from a lot of different groups. By mid 2004 finalizing
specific hypotheses and developing the study design. We expect to have the
vanguard sites funded by the end of fiscal year 2005 so I can’t tell you
any more then that but you can figure out sort of where we are with it.

In ’06 we intend to begin the full study with the initial centers, and
’06 and ’07 depending on budgeting of course because we’re not
fully funded, enrolling additional centers. And so the first preliminary
results of the recruitment period is thought to last about four years, for the
vanguard centers five years. And so as data continues, analyzing the data
continues the results will be published and the datasets will be made available
for public use.

So that, obviously you can always call me, half of you know me, or you can
go to our website. So what I really wanted to do today after this short
presentation is get your suggestions on anything that are gaps in the way
we’re thinking, things we should be thinking about up front that we
haven’t thought about, this is for me a golden opportunity.

DR. MAYS: Thank you, okay, let me open it up. Russell?

MR. LOCALIO: I’m Russ Localio. This is a wonderful presentation but I
just have some questions. This morning we had an hour of discussion concerning
the limitations on availability of data gathered through NCHS because of
limitations on the identifiability of data. Now you’re talking about, and
I don’t want to dissuade you from this, you’re talking about making
these data widely available to users and you’ve got lots of detail on
individuals including genetic information. I can understand how you can
de-identify it but how are you going to guard against identifiability?

DR. KNOX: Well, that’s a good question and we’re having huge
discussions on it and it’s going to vary. One of the issues is for
instance if you have a very small population, right now our power is calculated
to be able to look at African Americans and Hispanics separately but we
don’t know about other ethnic groups, so for instance if we take Native
Americans we’re now talking to the Indian Health Department and the
Department of Education about how to over sample to get that group in. One
issue would be if they’re from one reservation which wouldn’t make
much sense because the minute you have that location you can identify a lot of
what this is.

The issue with genetic data is a really good one, for a lot of the issues
that we’re looking at we’ll take for instance, we’ll genotype
them so we’ll have, I use the example of the serotonin polymorphisms, so
there’s a heterozygote and there’s a homozygote short and homozygous
long form, so you can de-identify that by simply using the ID number and since
there’s only three polymorphisms, on that level it’s okay but when
you start getting into more issues, we just had a workshop on gene expression
and the fact that you can’t get tissues for human studies but there are
other things you can do like RNA expression, these are things that we’re
talking about not only having a DSMB but also an ethics advisory board. And I
don’t have straight answers about genetic information right now but we are
really going to do everything we can not to make this a dataset that people
have to come to us to use, in other words when I was at the NHLBI we had a lot
of genetic data and most of those datasets are public use available. So this
one is going to be much more complex, de-identifying these datasets are going
to be much more complex then the ones, then any of us have ever done before.
But it’s our goal so if you’ve got specific things that you think we
should do please tell me, or methods that we should use please tell us.

MR. LOCALIO: No, I’m just wondering, the issue that I just talked about
is something that faces, that we have been discussed today because it faces the
use of other datasets and it would be wonderful, Vickie, if the discussions
that we’re going to have in various hearings somehow at least include
whatever discussions you’re going to have because I think the solution
ought to be consistent.

DR. KNOX: I think that’s a very good point.

MR. LOCALIO: Doesn’t make much sense for the right hand not to know
what the left hand is doing and have one set of rules for the right hand and
one set of rules for the left hand.

DR. KNOX: So do you have a subcommittee that’s looking at this or a
group that’s looking at this?

DR. MAYS: We’re going to and we’re going to have some hearings, I
was going to say what we can do, this is also what we should do because
we’ve done this for other federal agencies, is if you actually have some
questions you can formally send them to us and as we’re doing our, we can
actually ask that so we can help. So we can work kind of with you in that way.

DR. KNOX: That would be very helpful.

MR. LOCALIO: It might be nice if you could, somebody could let us know a
process and people involved in these deliberations on how you’re going to
de-identify the data and how you’re going to release data that have the
risk of being linked and therefore being made identifiable, I mean if we knew
you were talking with A, B, and C we might be able to talk with them as well

DR. MAYS: We have a Subcommittee on Privacy and Confidentiality and
they’re the ones that hold tons of hearings, so again it might be that if
you have a set of questions or issues that we can talk with them and we’re
going to be working with them around this issue of small populations so again,
that may be the group, remember our recommendations are ones, just so you
understand the recommends, I mean we would share them with you but
recommendations also would go to the Secretary, so you’d have a broader
kind of —

DR. KNOX: Well, that would be very helpful to have specific guidelines
because that’s what we’re struggling with right now. So what
we’re doing is we have Allen Fleischman who’s from the New York
Academy is agreed, thank goodness, he used to be on the Advisory Council, to
come to the program office and totally devote himself to things like ethics and
retention and things like that, but the statisticians that we’re using are
also from NCHS and from NHANES and so these are the people that are sort of
struggling with these issues.

But the other issue that I didn’t bring up that we’re having a
problem with is we have a white paper so to speak being developed now on other
existing databases that we can hook up with, for instance state data, county
data, criminal data, social service data, things that could give us more
information. We had the same problem, we had a presentation by the Census
Committee and in order to use their data you have to go to them and so
we’re trying to avoid this as much as possible but we need all the help we
can get trying to figure this out. So what I’m wondering is if, I mean the
hearings would definitely be a help, we’ll put together questions, but if
there’s any other level where our statisticians should be talking to you
or, do you know what I’m saying? Just so that everybody is on the same
page?

DR. MAYS: I think that part of what we should do is when we have, this
should go on the agenda for the November 4^th breakout meeting and we
should have a discussion about ways in which we can be helpful. Did you have
your hand up?

DR. LENGERICH: Yeah, I did, another part of the discussion that we had this
morning was around place and its impact upon health and you’re very
consciously choosing based upon place. Now you’ll have genetic
identifiable information but there’s going to be a number of other
information which would be specific to that place as well, rainfall, solar
radiation, environmental exposures, all those sorts of things as well and that
a lot of people would be able to gather that kind of information probably more
easily then they can identify an individual’s genetic information. So I
think you have a different level of identifiability that comes into place into
the discussion here and that is around place and its impact upon those
potential health outcomes.

DR. KNOX: I’m wondering about that, I’d like some feedback because
obviously if you’ve got a county in there that’s got to go,
that’s an identifier. But let’s say, let’s just talk about for
instance level of dilapidation, I mean we have someone walking around the
neighborhood making structured observations about the neighborhood, you’re
going to have multiple places where you’re going to have certain levels of
dilapidation or you’re going to have certain levels of chemicals. And so
I’m really wondering if that’s an identifier, I mean these are the
kind of sort of gray issues that we need to resolve but in my first thinking
that wouldn’t be an identifier simply because I’d be able to find it
in multiple places. Can you give me some feedback on that?

DR. LENGERICH: Well, yeah, I think that there would be a lot of ways, I mean
you’re going to want to explore that for example in SIDS, and SIDS being
more occurring at higher latitudes and so there’s a geographic occurrence
at a very large scale but it’s also going to be implicit within your
assumption is that it’s taking place at the very local community household
level as well.

DR. KNOX: Are these things that you’re going to be discussing, because
these are things that we really would like in put on, I mean these,
they’re quite the same questions we’re asking, how do we do this and
trying to work through the best way to do it.

DR. MAYS: Our interest overlap in the sense that we’re interested in
specific small populations and yours are not even racial and ethnic but are
still populations, I see we’re focused on racial and ethnic so to some
extent some of the questions will be the same so these are the kinds of
discussions we’re hoping to have and then to let that drive the kind of
hearing that we’re going to have in terms of who we want to bring in.
We’re doing this jointly with Privacy and Confidentiality Subcommittee so
that we can try and figure out, what we’re trying to figure out is much
what you’re trying to figure out which is in data that’s collected by
other parts of the federal government such as NCHS is there a way to be able to
allow more access to this. And it may is there a rule, is there regulations, is
there a process, is there some great de-identifier method or what, but those
are the same things that we’re trying to figure out for other things. And
part of why we were trying to talk with your group is that we don’t want
you three, four, five years down the line to repeat the same things that
we’re concerned about already so I think there’s real compatibility
there.

DR. LENGERICH: I’d also like to go back to that community issue, I mean
I’ve worked with groups that have been developing the relationships with
particular communities in rural areas for 15 years now and have pretty good
relationships not only in delivering health, disease prevention, health
promotion and initiatives but also in conducting community based participatory
research. And I guess I would, I guess it would go back to, I get the sense
that this is going to be a very federal government academic center driven
initiative and that I think the role for the community wanting and willingly
and being a part of the study is very important and I think that that is going,
I mean you can choose counties based upon a probability sample but getting
their participation short of paying them, and I’m not sure that
that’s even going to work, is really, I mean if it’s, I’m in
Dauphin County, Hershey, I mean we would be an academic center but if part of
you are trying to choose a rural area that’s 100 miles away I’m not
sure how us being the closest center would ever apply and successfully do that
let alone somebody in Philadelphia or LA or anywhere else but yet that is from
my understand your design a critical part of getting it. So I guess I’m
sensing a very kind of almost top down driven selection and there’s not a
whole lot of talk about the community’s willingness and participation
assuring high compliance and retention in your study.

DR. MAYS: Can I ask this question a different way because it’s the same
thing, I want to know how in the RFA process, what will you require in order to
be able to demonstrate what he’s talking about? I mean because we all know
it’s an issue but what I want to know is like for example the typical
process for NIH to do this is getting letters —

DR. KNOX: No, no, no, no, that isn’t going to cut it.

DR. MAYS: Thank you. Okay, can you talk a little bit about, I mean I have a
whole series of questions but one is how do you define —

DR. KNOX: I asked exactly the same question you’re asking —

DR. MAYS: — how do you define expertise and collaboration?

DR. KNOX: I even formulated it the way you’re formulating it, letters,
we’ve all seen that, so no. There are certain things I can’t tell you
because they’re procurement issues but let me put it this way. First of
all they have to show a history of funded research, productive research,
longitudinal, that has been involved with communities, they have to have done
it and shown that they’ve done it. That doesn’t mean every single
county within that thing but do you understand what I’m saying, this is a
competitive, if you’ve got four out of seven counties and one investigator
and zero in another guess who’s going to get it.

The other thing is that we are, and I can’t tell you anymore because of
procurement, we are making it strong incentives to keep the community involved
and that’s all I can tell you but it’s not just lip service, it
isn’t going to fly any other way. So still we could have problems but this
is something we thought very, very seriously about and not allowing them to
just fall in and say we’re going to do it, there are other —

DR. LENGERICH: Your criteria are very logical, I guess I’m just
wondering if you were to look at a map of centers that have that kind of
history I think it’s going to be a pretty sparse map.

DR. KNOX: And again, I can’t tell you how many people have said that
and said you’re going to fail.

DR. MAYS: Maybe I’m being very West Coast but I don’t —

DR. KNOX: No, people say you’re going to fail, they say you can’t
do —

DR. MAYS: I don’t agree, I think the coast will be fine, that’s
what I think, I think the coast —

DR. HAYNES: I don’t know it necessarily has to be an academic center to
win this. If I were looking at it I’d look at groups like Westat coming in
and folks who do surveys all over the country and do follow-up surveys anywhere
in the country if you think about it. So there are some strong competitors out
there that could come in and Westat does the HANES survey —

DR. MAYS: She’s right.

DR. KNOX: They’re going to have to be able to show that they can’t
just take a van and take one blood sample and move their van to someplace else,
they’re going to have to be able to do it continually —

DR. MAYS: Westat can show you that.

DR. HAYNES: They can show you that, I don’t know it’s necessarily
going to end up at academic centers —

DR. KNOX: It’s going to be a very objective competition so they’re
going to have to —

DR. MAYS: This is frightening, the nice part about it is if you’ve been
doing it you’re in the running, if you try and create it to be in the
running they’re going to see through it. See letters, thank you, Sara,
very much, that —

DR. KNOX: The beginning part of the Childhood Study they brought in several
investigators around the country who have done longitudinal studies in kids and
there were probably eight cohorts over the last 20 years that have been
followed up so I mean there is also a track record of doing that, following
kids in cohorts.

DR. MAYS: I want to ask just some questions and then I’m going to turn
it back over to my colleagues, is this going to be a U(?) mechanism?

DR. KNOX: No, it’s going to be a contract mechanism.

DR. MAYS: When you talk about the hypothesis one of the question I have is
are there models though that are going to drive this. What I’m trying to
understand is there can be a hypothesis but I can come from a model that
emphasizes one thing and somebody can have a model, like is the work to be that
there is going to be a driving theoretical —

DR. KNOX: No.

DR. MAYS: Okay, great, this sounds better and better. Can you talk a little
bit about the measurement of things like social economic position and
education? I mean is everybody going to do it the same way and in whatever the
mechanism is, let’s just call it an RFA, any RFA, do you have to
demonstrate that you have the expertise, not just to do the study but to do the
study in a culturally competitive manner?

DR. KNOX: I have to be careful about what I say now but let’s put it
this way. When you write an RFP you have to give enough information for people
to be able to budget and so you have to be able to say at this home visit you
will need this type of expertise. What we’re doing because we have to
standardize this across centers, we have to be able to do that, that’s one
of the reasons the contract mechanism, it’s also one of the reasons why
we’ve been doing an enormous amount of work ourselves selecting, trying to
get down to specific measures, that will be completed with the investigators
but we have really gone through, we’ve had workgroups working on specific
measures making recommendations and we’ve had workshops and so forth so
we’ve really tried to synthesize a lot of what’s been done out there,
and Audrey has been involved in a little of this, what’s been done out
there on an epidemiologic level so it’s got to be standardized across
centers.

DR. MAYS: Can you talk about the biological samples issues in terms of this
might fall out now given the populations that are the ones that you would think
would participate but there’s some groups for whom there are real issues
about taking biological samples and there are groups by which, at least what
one of our centers at UCLA is finding is that we keep kind of getting into
these problematic places when we do this genetics because there are kids who
think so and so is their father, and it’s not even that the mother
didn’t tell them, the mother didn’t know, she was with people at the
same time, there’s these complexities that are going to go on and I
can’t imagine what it is that you’re required to tell someone. Have
you all gone through a lot of this?

DR. KNOX: We had discussed that extensively and one of the biggest problem
is that when you get people together from different institutes they’ve all
got different policies. It’s not even the same policy within NIH. So
that’s been a humungous problem. When I did one of the studies with the
NHLBI what we did is we got the mother in one room and we said to her, and it
was a confidential thing, please tell us who the father is genetically and that
was strictly confidential and so on and so forth, we just had a workshop where
these issues were discussed. And we haven’t come to any specific,
we’re still synthesizing down the issues from that workshop so I
can’t give you a direct answer to that right now. But the issue that
we’re trying to resolve is the fact that there’s different policies
in different places and we have to resolve how we’re going to do this.

DR. MAYS: One of the suggestions I would make is that, I mean I know
you’re going to bring some ethicists on but I really suggest that what you
think about is going to some of the other institutes and seeing if they can
fund a study of some of these things while you’re in the process of it.
For example I forgot the group but I think its NIGMS or something, they do all
this ethics stuff, is that the right one, I forget, but there are some issues
that we really could use some leadership on and it’s like you’ll have
the perfect living lab for it and if you have a study that’s going on it
will allow you to be able to change. Because otherwise —

DR. KNOX: We have been in touch with genome —

DR. MAYS: Yeah, but what I’m saying is see if they will give you money
to actually have some —

DR. KNOX: It’s been a big issue, they’re competing with us
actually for government funds —

DR. MAYS: I was going to say then if you all could do this because I think
that you really need to be able to change in the middle of this and
unfortunately changing ICFs and all this other stuff isn’t easy unless you
have some very compelling reason and I have a feeling you’re going to find
yourself in the middle, particularly with poorer people, more complex kind of
issues, that you’re going to need to change and your best way to do this
if you’re driven by science and not an incident.

DR. KNOX: So what you’re saying is one of the biggest issues is going
to be ethics issues with respect to genetics, I think you’re right.

DR. MAYS: Oh yeah, and I think you don’t know what they’re all
going to be and I think that’s different then having an ethicist panel.
What I’m suggesting is that you actually have some studies of these to
determine what the issues are, what the method will be for handling those
issues, and that you do it not for people sitting around and with
pronouncements about it but that instead you do some studies about what
difference the ICS makes. Because I’m thinking about racial and ethnic
minorities, why they don’t participate in some things, the kinds of things
that they are fearful of and that we may need to have a different kind of
certificate of confidentiality for example. Confidentiality really is designed
to protect people in terms of legal issues, we have not thought about a
certificate of confidentiality that kind of helps people who if there is data
discovered about parenthood and all this other kind of stuff that one person
doesn’t sue the other person, I mean there’s drama that I think will
evolve.

DR. KNOX: Well, that’s why we just had this workshop and that’s
why we included people from different agencies to discuss this, and the reason
I’m not giving you any answers is because that report is still being
written, it’s not because I can’t give them —

DR. HAYNES: We had an example earlier this morning on Tay-Sachs and say for
example there’s no Jewish person in the family tree but then the kid ends
up with Tay-Sachs gene then you know, and you have to report that to the
family, then you’re automatically caught in a situation of where did it
come from.

DR. MAYS: See I think you’re going to break ground on some of this
stuff and I just, I have a feeling that you’ll just do much better if we
have a new kind of certificate of protection.

DR. KNOX: I can tell you what the consensus was sort of around this issue
and that was the basic question from our ethical standpoint was how much
feedback do you give to an individual about their genetic makeup, what if
there’s something that may be a risk. This was one of the major issues
that came up. And so the consensus seemed to be in this group that if it’s
not interpretable from a medical standpoint, in other words if you specifically
say you’ve got a gene for sickle cell that’s one thing and you know
about it. But most of these genes, most of the illnesses that we’re
dealing with are very complex so there are either multiple gene or there are
gene environment interactions and having the gene one way or the other
doesn’t tell you anything.

I can give you specific examples of why we’re looking at the
psychosocial factors on gene expression and that is if you look at the
non-human primate research you take one and the same, polymorphism, the
homozygous short form, you rear those monkeys together without a mother and
that particular genotype becomes alcoholic if given alcohol in adolescence. You
take the exact same genotype and put it with a nurturing mother and not only
does it not become alcoholic it has a protected effect against alcohol.

So to give somebody the feedback that your kid is going to become an
alcoholic would be off the way so that is why the consensus was that if there,
and this is why a separate ethics committee is being created in addition to a
DSMB, because there are also issues of what if we get into a community and we
see a rather high level of, not like lead, that’s obvious, we know what
the thresholds are, but something else that we suspect is injurious but we
don’t know it, what do we do, go to the DSMB first, ratifies the data and
say it’s real, this isn’t just an artifact, that goes to the ethics
committee that says okay, we need to up the number of measurements here, we do
to X for this, we need to do X for that, or we need to call.

So the consensus with this issue of giving information back to individuals
is that you basically don’t just simply because you can’t interpret
it and all you’re doing is causing fear or upsetting them when you
can’t give them anything so that’s kind of where the discussion was
going with this.

DR. MAYS: Let me open it up, Russ?

MR. LOCALIO: I have just one other question, you’ve described a three
stage sampling, probably stratified, over sampling so it’s weighted, you
described a longitudinal study, repeated measures one exposures, probably
repeated outcomes with an individual. I don’t know a software that will
handle problems of that complexity.

DR. KNOX: Let me add another level of complexity, it’s even worse, and
that is that there are going to be dependent measures —

MR. LOCALIO: Define dependent.

DR. KNOX: What that means is that if one women has two pregnancies, a woman
who’s recruited has two pregnancies within the recruiting period so the
kind of software I use for that personally is GENMOD(?) but I have already said
that in the data analyses discussions that this is going to be so much more
complicated and hierarchical and so this is one of the reasons we’ve got a
whole separate data module to this to deal with precisely those issues.

MR. LOCALIO: To make the data available to users are there plans for making
software available to users or how —

DR. KNOX: Again, I’m speaking with two mouths here, being an
investigator myself and having been on the outside there’s the issue of
somebody works their tail off for three months, 18 to 20 hours a day to get
this contract and they work their tail off to get all this data in and then
before they have time to even analyze it you’re giving it away to somebody
who hasn’t done jack on it. So I don’t know how easy we need to make
it for these people, maybe I’m speaking out of turn here but the issue is
we do want it to be used and we do want it to be analyzed but I don’t know
whether we’re in the business of spending our government money to create
programs for people, maybe they can come in with an RFA, that level has not
been satisfied yet.

DR. HAYNES: I think the NIH rule isn’t it, Sara, that you can’t
release it until have the first, it’s like three months after the first
publication, aren’t there NIH wide rules about that —

DR. KNOX: If you look at the NIH rules I think they talk about something
like three years after the first wave, it’s after the first publication
but the first publication takes a long time but now there are a lot of people
talking about doing it earlier and I think these are people who have never been
involved in a study like this and don’t realize how long it takes to clear
it through the publications committee and how long it takes, so there are both
sides to this issue about the rights of the investigators here and the rights
of the public to access.

DR. CAIN: But there isn’t an NIH rule that you can’t release it
until that point, I think there are norms in different communities, certainly
in the demographic community the norm —

DR. HAYNES: We just found, because I’m working on an infant feeding
practices survey with FDA and some other folks at NIH and we actually looked up
the rule, there is a rule, I think it’s three months after the first
publication.

DR. CAIN: The demographic community tends to release their datasets
immediately, as soon as they are available.

DR. HAYNES: I can send you the website we were sent.

DR. CAIN: It may be that that’s the first point where NIH could force
the release of it but anybody can release their data sooner then that. The
norms of different communities are different.

DR. KNOX: They give the investigators five years, so it does, it varies.

DR. HAYNES: I was surprised there was an NIH rule and they actually go in on
the NIH website and found that there is a rule now.

DR. LENGERICH: Sorry to be hung up on this community thing but just a couple
of comments, or one question, maybe pretty straightforward, but that question
is what’s the length of that contract that you would be awarding because I
think the longer —

DR. KNOX: I don’t think I can tell you that —

DR. LENGERICH: Well, I hope it’s very long and I hope the ones you
award are good because that’s going to be part of it. And the second is so
if you find an exposure which is related to that community and not to the
individual how will you handle that in the ethics process?

DR. KNOX: Well, that’s exactly the kind of example that’s come up
because it depends on what the risk is but these people are going to be very
carefully selected so that they’re very senior experience wise and wise
people, and it’s going to vary from case to case. For instance if
it’s in the water is it something we can go to the community and say you
need to change this, or is it something that we don’t know what the result
is but we suspect it’s dangerous then it might be instead we have to make
more frequent measurements of certain outcomes to see whether it’s having,
I mean so it will vary from case to case and that’s why the ethics
committee is going to have to make formal recommendations.

DR. LENGERICH: So if it is something, if there is something discovered in
the water that does have an adverse health effect will DSMB or this other group
be willing to go to the water authority and ask for them, and require that they
shut down?

DR. KNOX: Well, whether or not, they can’t require anything, they
don’t have that power, what they can do is they have an ethical obligation
to make it known, to make it known to the community number one so they can
grassroots but also if there’s a certain authority to make it known to
them. So that’s why we need a committee because these aren’t simple
answers.

DR. LENGERICH: And that’s why the people in the community are the ones
that need to make that, will be there to make or not make that happen.

DR. KNOX: That’s why they need to be informed and that’s the
issue. So the ethics committee needs, first of all DSMB has to look at it to
find out if it’s real, you don’t want to go scaring people about
something that’s an artifact, so you have to first establish that the data
are reliable but then you need to decide about what to do about it.

DR. HAYNES: Is there a representative committee planned of community
representatives in this study?

DR. KNOX: They’re supposed to be involved at every single site.

DR. HAYNES: Like seating X number of community representatives on the boards
—

DR. KNOX: Well, they’re already on our advisory committee but it’s
more specific then that because they have to be involved at a local level and I
can’t tell you anymore for procurement reasons what —

DR. HAYNES: You’d use those people probably to help make that decision
—

DR. KNOX: But they have to be networked to the study.

DR. MAY: Let me ask a question and then make a comment. Are any of the white
papers available?

DR. KNOX: Yes.

DR. MAYS: Okay, how would we get them?

DR. KNOX: I think they’re on the website, now unfortunately
there’s a huge latency between the time these things get finished and the
time they get on the website but they should be on the website available to
anybody who goes to it.

DR. MAYS: And then my comment would be, because again one of the things we
keep dealing with is this access to data and I’m wondering if it could be
part of the RFP, RFA, whatever, is that those academic centers also should be
places where people can go and access the data. If they have the capacity for
example to be able to do this you could, and again it’s a budget issue, is
to help them to develop, it’s not that expensive it’s just a matter
of how long you fund them for but they could come back in as an incentive to
actually get money to have a secure data site so that once the data is in
people from the local area who’ve now worked on this would have the
capacity to be able to use it —

DR. KNOX: Local researchers or —

DR. MAYS: Well, I mean even the community groups, I mean it depends, we were
talking about this earlier, you could do things with the data, have query
systems, you could do things with the data where there are community groups
that are able to come in and researchers help them for example be able to use
that data. Or you could, they could be a secure data center where a researcher,
like it would be easier for somebody to take the train to a local area and go
back and forth to work on data then it is to come and camp out in Washington.

DR. KNOX: So you’re bringing up something very interesting because my
first thought when you started talking was we want to make sure that what goes
out to researchers is standardized, what I’m hearing is a completely
different thing and that is that people in the community can have access to
data about their community and we haven’t talked about that —

DR. MAYS: One of the things we’re talking about is that quite often
local people want the local and not necessarily the whole national. What
I’m saying is that some of the best people to do that might be that
particular academic center —

DR. KNOX: Now that’s making sense.

DR. MAYS: — and that what you would do is you would incentivize that after
this is over that they could come in, or somebody else could come in in the
local area but that would you would do is these are the requirements. And what
we should suggest that you look at because we’ve looked at it is like what
the Census is doing in terms of local access to data, they actually fund people
and those people go out and give it away and make it easier for local people to
be able to use the data in their area. For instance UCLA has one and it’s
funded for Asian and Pacific Islanders, so they work with all the community
groups in getting the Census data out about the Asian and Pacific Islander
community. They also make —

DR. KNOX: They didn’t mention that when they came to talk to us,
that’s interesting.

DR. MAYS: Oh, okay, see these are the little local, because it’s more
local, it’s not a national —

DR. KNOX: That’s very interesting, I think that would also be an
incentive to the community to have that available.

DR. MAYS: Exactly, exactly, so that would be a suggestion of access.

Okay, final comments, questions for Sara?

DR. HAYNES: I just had a question about your breast feeding questions on
there, are they well developed and will you be asking them multiple times
through the first year how they’re feeding their baby?

DR. KNOX: We are asking about breast feeding and we haven’t decided
specifically which measures to use yet. Can you be specific about what you
think is most important?

DR. HAYNES: Exclusive breast feeding at six months is a real important
variable to access because it’s been shown, we’re doing a national
breast feeding awareness campaign here in the Office on Women’s Health and
we have extensive literature on a number of diseases that you have listed there
that are protected among children who were breast fed exclusively for six
months.

DR. KNOX: Don’t American pediatricians recommend, I had my baby in
Sweden, that’s why I’m asking this question, at four months —

DR. HAYNES: No it’s six months, it’s six, and ACOG is six, family
practitioners is six —

DR. KNOX: For solid food, don’t they start on solid food at four
months?

DR. HAYNES: No, six. I mean there’s a fight within AAP but ACOG and the
other organizations are at six because there’s a nutrition committee
that’s fighting with the pediatricians but the actual breast feeding
section of AAP recommends six months.

DR. KNOX: That’s good to know —

DR. MAYS: I was going to say, I think she’s a good contact person for
that because of what they’ve been doing in the Office of Women’s
Health.

Let me just summarize where I think we are before we move on to our next one
and that is we will look forward with dialoguing either directly with you,
Sara, or whoever from the study in terms of keeping you abreast of our hearings
and activities and you may want for example to ask us to receive testimony on
certain issues —

DR. KNOX: I want to follow-up on this issue so definitely, please do keep
us, because I think our statistician should be talking to you and trying to get
these issues solved, that’s very important.

DR. MAYS: And as we’re having these discussions, they’re public,
so we will make sure that, again, the contact person, just tell us if it’s
you or if it’s someone else, and then what we can do is make sure that you
are invited to these, you’re aware, can being whoever, and then listen to
the discussion. Russ, was there anything else? Okay. And then I think the other
thing is that in our letter one of the things that we had asked is to be kept
on an update and a periodic presentation so I guess what we’d like to do
because I think there may be more that we can assist each other with as opposed
to each being off in our corners, that that might be a useful thing to do so
we’d actually like to invite you to consider kind of periodic updates to
us.

DR. KNOX: What would be easier actually is a presentation then trying to
remember to write letters because we usually say all of this stuff is available
on our website but we’d be more then happy to make periodic presentations.
Would that be helpful, or you want a letter, or —

DR. MAYS: No, I think probably, particularly in the beginning, I can see
that a presentation is probably better then just getting a letter from you and
you say something and we say well that didn’t tell us very much, so I
think we should probably, we should do that and I think what we need to do on
our end is to figure out whether your updates sometimes might be to the full
committee and not just to the subcommittee but this has been important for the
subcommittee at this point to get a sense of where things are headed and then
we can talk to the full committee.

Virginia, is there anything else you think we should, that you can think of
would be useful to do since this is kind of NIH?

DR. CAIN: No, I mean I think the important updates that will be coming along
or when the RFP is released and then also when the projects are funded, get an
update on where that’s —

DR. MAYS: Particularly after the first vanguard centers are to tell us a
little bit about that and what have you. Okay, alright, I think that
that’s, this has been, I just want to tell you this has been very helpful.

DR. KNOX: It’s been very helpful to me too especially this issue with
the genetic data and hearings, I would really like to stay in contact about
that because we need all the help we can get, and for all this other
information so thank you very much.

DR. MAYS: Thank you very much.

Okay, I think that worked well, I think it has us in a different place and
we don’t need to send off any letters. Leslie, did you have any comments
that you wanted to make before Sara goes?

DR. COOPER: No, no, I thought it was a very, very good presentation, very
informative, and it definitely helps to identify some of the potential roles
that we have in terms of working together to accomplish all of our goals.

Agenda Item: Discuss Potential Role for NCHS BSC on the
Vision of Health Statistics for the 21^st Committee –
Subcommittee

DR. MAYS: Okay, that was good. Let’s move on to our next task which is
to begin to talk about now the potential role for NCHS’s ownership, their
Board of Scientific Counselors on the Vision of Health Statistics for the
21^st Century.

As you know our two groups, oh no, I shouldn’t say two groups, that
NCHS because it wasn’t the BSC, that NCHS and Subcommittee on Populations
worked jointly to produce this report which is called the Vision of Health
Statistics for the 21^st Century. What happened was that we were kind
of left to monitor that and one of the ways we’ve been
“monitoring” it is making sure that for example we’ve had
population health presentations, we have introduced our newer people as part of
what they get when they first come on board, we’ve introduced them to some
of the models that appear in this vision.

But I don’t think that that’s enough and now that there is a Board
of Scientific Counselors I think that maybe it would be good for us to have a
discussion of how do we officially put that on their plate for them to actually
look at because NCHS is supposed to be one of the premier statistical groups
and we’re supposed to be advisory to some extent to Tommy Thompson, and I
think that receiving the report and actually trying to do something about it
within NCHS would be a big, big step forward.

So maybe we should talk just a little bit about well what can we do to put
it at the top of the BSC’s list, they’re right at the point of
beginning to form committees, they want to look at specific surveys, this would
be kind of a big overview to look at in terms of thinking about the Vision of
Health Statistics for the 21^st Century.

And we have Aldona here with us today, who is our liaison from the BSC so do
you have any thoughts about this report?

DR. ROBBINS: I guess let me, of course I’m familiar to some extent with
the report and I guess what I’m not clear on is what you, you’re
saying to try, taking the existing report and somehow having, see if it can
somehow have a greater impact or greater effect?

DR. MAYS: I guess what I’m saying is that on our side part of what
we’re trying to do is make sure that people have a population health
perspective and so kind of the movements that we’ve made in terms of the
report is making sure that committees other Population understand a population
health perspective, that they integrate a population health perspective into
their thinking and their recommendations, and where appropriate I think
we’ve even used the model to make a case for how it is statistics maybe
should be looked at, what the importance is of population health statistics to
a particular problem. And I guess what I’m saying is I can see in the
formation that’s going on for the Board of Scientific Counselors that
they’re about to start looking at specific datasets and maybe some issues.

But again this whole notion of kind of our vision of where health statistics
should go, how do we integrate that in our perspectives on health at the level
of the department. For us I would think one entity that’s very important
to do that would be NCHS and so I’m just trying to figure out how to put
it on the agenda of NCHS Board of Scientific Counselors so that you actually,
see it was a joint project so we have a sense of a process by which we’re
keeping it alive and I guess it’s a matter of well how do we start a
process to get you all to keep it alive.

DR. ROBBINS: And of course as you know the board wasn’t really around
when the report was developed so to us I guess we’ve seen the report and I
think at some, one of our early, and sometime last year at one of the meetings
we did have a presentation I believe on it, and I guess I would say, certainly
from my perspective how we can better use health statistics in some sort of a
modeling fashion, I guess I see that there are a lot of questions that can be
answered, there’s a lot of rich data out there already that have not even,
we’re not even approaching utilizing it in some kind, we’re not
utilizing it to its full capability. And the board at this point is really
struggling with what direction that we want to go in.

I think we had done a survey of the members in terms of how do you want to
organize yourselves into subcommittees and what kind of topics do you want to
do, and I think there’s kind of a split, I don’t know whether
I’d say it’s 50/50 or 60/40 at this point, but some of the board
members really want to get to overview the datasets because that has never
really to my understanding been done, just to take the surveys on a survey by
survey basis and sort of look at a whole host of questions in terms of the
sampling and methods, to how relevant are the questions that are being asked
and could some questions, could they be dropped and should we add others and so
forth and so on. And within NCHS there is going to be an evaluation that is
coming up sometime during 2005 and I think that Dr. Sondik would like to have
some kind, the systematic review let’s put it that way would be extremely
helpful.

The other part of the board is really kind of more, I think more interested
in getting at issues and population health and the issues and how do you model
it would be certainly one of these, would be certainly a candidate. At the
September meeting there was discussion, we didn’t come to any consensus
really other then I think we’re going to revisit the issue in January at
our next meeting.

I guess if there is some specific suggestion that you can make, that this
subcommittee can make, about what might be one thing that the board could
consider, one I certainly can bring it before the board and talk to our
executive director and whatnot and I think get it on our agenda. So I hate to
throw it back to you but this is your, you live and breath this so what do you
think would be helpful, and you sit in on the meetings as well, Vickie.

DR. MAYS: Eugene?

DR. LENGERICH: I guess I believe that this a good document and offers some
guidance and principle values and principles and recommendations and I’m
not sure to what degree the board has looked at it but there may be a place
where the board can get behind some of these specific recommendations or areas
and kind of pick those up, I mean one place is that this tries to integrate
well with the National Health Information Infrastructure, NHII, and I think is
in some sense this document is a large part of the population health piece of
that. So I guess I would look at going back to the document and talking about
kind of the specific principles or recommendations and seeing what people or
maybe the board would pick those up or work on those.

DR. ROBBINS: So you would actually go through and say alright here’s
some recommendations, do we want to take, is this something that would be a
good idea for us to take up or have a subcommittee take up and look at more
thoroughly or whether it’s a good idea, bad idea, and then if it’s a
good idea try to implement it. Is that the sort of thing you’re talking
about?

DR. LENGERICH: Yes.

DR. MAYS: Well, let me just make a comment because I don’t know if you,
if for the board it’s a matter of a good or bad idea because the agency
has signed on to it —

DR. ROBBINS: So it would be more a matter of —

DR. MAYS: It’s a matter particularly I guess if you’re about to do
an evaluation, it’s a matter of how does this fit into your framework and
I know that you all became interested and that was going to be kind of my
comment, I know that you all are somewhat interested in NHII and it’s
almost like this is your, one of your ways to connect with NHII. It may be a
guiding principle of thinking about the datasets, it may be for example that
this is a way of thinking about if there is going to be kind of a coordination
and you want to look at how your datasets do that, I think the question of the
models that are in there, actually I wish I could put it up, could you turn to
the page with the model because I think it’s actually going to talk about
whether or not if you look at all your surveys, for example do you have like
contextual factors that are there, do you have, but yeah, and so I just think
that at this juncture of when you’re about to start that might be useful.

Another suggestion that I would make which actually was brought up, I’m
sorry that Marjorie isn’t here, I remember when we were at the retreat
Marjorie brought up that they had been having some discussions with Dan
Friedman who is actually one of the major people on that report about doing
some work and I think that, I think we’ve discovered a process here for us
and you see how we’re getting into more and more discussion about it in
terms of the kinds of issues that are a part of that what type of statistics to
collect and what have you. But I think that if there is a need for consultancy
it may be with the BSC or NCHS to figure out how to move those things.

Like to me a concrete example would be if you’re going to look across
all the datasets and it would be with those datasets can you formulate a
population health perspective and what’s missing, or what can be shored
up, if you’re talking about changing questions, it may be that if you look
at the question in and of itself it does look like it doesn’t matter that
much in a particular survey but it may be the only place that you learn
something that really helps and to remember that sometimes people do kind of
brick by brick approach here in terms of what the various surveys and they take
something from NHANES and then they take something from NHIS, and then it helps
you to get this picture. And the question would be are the datasets constructed
in a way in which NCHS really, since its signed on to this vision of health
statistics for the 21^st century, it’s there with it.

DR. ROBBINS: As Eugene was talking I could see, what came to my mind was
using this as a guide if you will, or framework, for going about, at least as
part of the framework for evaluating the datasets if that’s indeed the
road that we end up going down, that would seem to make a lot of sense to me.
But then on the other hand, and if we don’t go down that road, the issue,
going down that issue road is still, this is still a useful document as well at
least for those people that would be concerned with issues having to do with
population health. And the board, well, as is your committee, has a lot of
various interests and disciplines and so I think that, I think population
health is one of them but there are others as well and as I said I’m not
sure how much emphasis, especially early on, will go to population health. It
may be a great deal but just be aware that we’ve got others, I mean there
are people that are very much interested with the vital statistics program and
working with the states and all that —

DR. MAYS: It all fits though, it all fits. I know that Dr. Sondik is
interested in this so do we need to formally send you anything or can you have
this discussion?

DR. ROBBINS: I can have the discussion and let me do this, let me have a
discussion and see what, at least with the chair and the director and maybe the
executive director of the board and possibly Dr. Sondik because the agenda for
the January meeting is still to be set.

DR. MAYS: Or you may want a presentation from inside, the individuals at
NCHS who actually worked on the report which is Rob Weisemer(?) —

DR. ROBBINS: He’s our executive director so I’m sure —

DR. MAYS: I’m just staying that that may be actually a good start is to
have a discussion because he knows the report inside and out and he may be a
useful ally to help you figure out how to put it into the fabric and even how
to start the discussion.

DR. ROBBINS: And then depending on how successful I might be that maybe your
subcommittee might want to, we can talk again and maybe formulate some other
suggestions —

DR. HAYNES: Is your feeling that the report has been put on the shelf? And
then okay well there was that report and we went it around to everybody but
it’s just another government report?

DR. ROBBINS: Well, my feeling is, again, the board is dealing with NCHS and
there are just, first of all they’re primarily a producer of data, there
are a lot of other things I guess that are sort of overtaking, the short answer
is probably yes but we’re also caught up in a reorganization of CDC. I
think a lot of things may be in flux at this point —

DR. HAINES: You mean they would reorganize NCHS itself or just —

DR. ROBBINS: No, CDC is being reorganized —

DR. HAINES: NCHS has always been off by itself.

DR. ROBBINS: No longer, it’s going to be a coordinating center because
the CDC is going to be regrouped into these coordinating centers and they are
going to be in a center that includes really I would say essentially two new
entities, one is informatics group, and that makes sense, and the other is
right now for a lack of a better term they’re calling it a marketing group
the idea being to have the information that CDC produces, to have a better
outreach with various affected communities and with the public in general but
the term marketing is bringing up a lot of people short because we have
objective statistics on one hand and NCHS has got a pretty pristine record
among government statistical agencies for really trying to do objective, put
out objective data, and now we’re going to have this coordinating center
which is got marketing in it. So I think this reorganization may have eclipsed
a lot of other, put a lot of other things on hold as sort of people are kind of
groping with what do we do. And this program evaluation is coming up in 2005
and it seems to be pressured at kind of we’re going to look at all of the
things that NCHS does to see where does it fit —

DR. HAYNES: It does seem like though you would have much more power in that
evaluation or strength in what you were doing if it tied into the vision, in
other words that if you organized around the vision and sort of tied of
everything to that rather then nitpicking every survey and say well I
don’t like question number 20A or things like that that it would fit into
the broader scope —

DR. LENGERICH: There are ten guiding principles in here, some overlap with
some of the things that the NCVHS does such as standards but there are two or
three which I don’t think NCVHS has really talked about, one is
flexibility, at the back, page 56, one is flexibility of the data systems,
another is the timeliness of the data systems, and the final one actually is
evaluation. So there are three that are called for here but I don’t think
there’s a lot of work being done in those three areas.

DR. ROBBINS: And maybe since the survey evaluation has already been put on
the table that that maybe is one of the roads we may go down, that this in some
sense gives added, both does two things, it gives added weight to perhaps that
that would be a good idea to do and also gives you a framework for doing it.

DR. HAYNES: Yeah, look at those three issues in terms, and yeah, the
evaluation probably includes what some of your people on the board want to do
is nitpick the surveys but you don’t want to get caught in that because it
could take you ten years to get through that debate.

DR. ROBBINS: And I think also at the end of the day they want to be able to,
we don’t want to consider each survey in isolation which this kind of
survey, base survey approach can lead you down that road but to somehow manage
to we want to relate it all to surveys to each other and also from my
perspective the big thing is what are the questions out there that need to be
addressed and are the surveys designed to at least produce data that will,
could be used to address those questions and that’s another —

DR. HAINES: This is where the Populations Committee comes in because
let’s say the pressing issues that are in minority populations, is there a
way to use this data that’s already being collected to get teased more out
of the datasets for Asian and Pacific Islanders or Native Americans or other
populations, that’s what this committee has been concerned about.

DR. MAYS: I want to build on what Eugene is saying because I think that in
those ten recommendations, even if we discussed that, we may have to discuss
them at a much broader level and this is part of what I think we’ve had
the discussion at the board is yes, we do it at a national level and we have to
be concerned about some broad issues but almost, many of those issues are still
your issues also to some extent and even if we talk about technology, if we
talk about health informatics, these are things that do, I don’t want to
say come down to but they are central to some of what you’re doing even
though you’re concerned about them not only for you but for a lot of other
groups.

Again, I’m just going to urge that you all think about putting this on
your agenda because I do think within this context of just, of what you’re
getting ready to start that it would probably be useful to have this either as
a framework or useful to have it, for you all to pick it up. See we now have a
process and I guess what I’m trying to do is to make sure that you all
have a process —

DR. ROBBINS: And in fact I can see that on the agenda it’s not
necessarily to talk about the report because we’ve supposed been
introduced to the report but the idea would be to have, and I’ll talk to
Rob about this, how about is there some way we can tie using this document to
the kinds of things that the board seems to be interested in doing anyway and
then kind of presenting it to them because that may help the board with its
kind of groping —

DR. MAYS: Or you introduce that report into your structure of operating, see
that’s what we’re trying to do is to introduce it into our structure
of operating so that we always know that we should be paying attention to
population health and it may be the same way with you all.

DR. LENGERICH: And as you look at whatever structure of those ten
recommendations or whatever, use the ten, I think some coordinate with what
NCVHS is doing, for example there is one on standards there and we have a
Standards Subcommittee, so that may not be as high a priority for what you do
but there are other ones here that I don’t think we’re doing.

DR. MAYS: No, see in an interesting way it is, this is what we run into like
when Russ is interested in privacy and confidentiality, they have committees
and their committees will deal with standards but they’re not the national
standards, they’re the standards that have to do with for example policies
and procedures for the datasets and all that. So they have almost, almost all
of those things apply to them but it’s just that we deal with it at the
level of establishing the code and they’re interpreting the code and so
it’s kind of like, what’s helpful is for there to be I agree some
coordination.

DR. ROBBINS: In fact one of the crosscutting issues that they’re
talking about is confidentiality, which is your privacy —

DR. LENGERICH: I think I’ve heard that before today haven’t I?

DR. MAYS: We’ve been talking about it too. Okay, I feel like our job is
done there, that we don’t even have to write letters or do anything, it
was just a discussion, thank you very much, we like that.

Agenda Item: Poplations Report Status – Agenda for the
November 4^th Breakout Session – Dr. Mays

Alright, as we kind of get to the end here let’s start with one thing
first and that us Audrey let’s kind of walk through the things that
we’ve said that we want to put on our agenda, because I thought it would
be good to do the agenda by talking here about what kinds of things we should
carry over to the November board meeting. So what do we have so far?

MS. BURWELL: Well, I have three main things, the first is discuss possible
joint hearings with Quality based on our discussions this morning —

DR. MAYS: Exactly, that’s a definite.

MS. BURWELL: The second one is I heard some discussions around requesting
HHS to conduct special studies on changes in race ethnicity, whether or not we
might talk with ASPE or someone about that.

DR. MAYS: Oh, that’s the mental health piece. And the mental health
piece, what that is is that we were going to ask whether or not it’s
possible for internal studies to be done and that was asking NIH, AHRQ and NCHS
as to whether or not they have data in which they could look at this issue of
the extent to which people are changing their race and ethnicity. So I think
that will be kind of a, you know what we ought to try and do, Virginia, do we
need, would it be better like if we actually sent letters or is this something
that’s better that we ask unofficially?

DR. CAIN: Well, I think it would be better to ask unofficially until we
figure out where.

DR. MAYS: Alright, okay, so what we’ll do is we’ll discuss it at
the meeting and we’ll kind of figure it out rather then sending some
formal letter because it might take three months to get a response whereas if
we just talk to each of, somebody at each of those agencies. Can we figure out
who at NCHS to talk to? Well, Julia, we’ll let Julia continue but Julia
actually brought it up about the bridging process so I think Julia will come
back, we’ll need to figure out for AHRQ and I don’t know at AHRQ who
we should talk to but that’s the Quality group so I guess we should talk
to, it’s Anna Poker and who’s the other staff person, Eduardo —

PARTICIPANT: Eduardo is now —

MS. GRANTHON: Harvey Schwartz —

DR. MAYS: Harvey Schwartz is our person so why don’t we call Harvey and
ask him and then see if Harvey can come to the breakout, that’s a good
one.

MS. BURWELL: The third one was the whole discussion on the National
Children’s Study and how to help them better flesh out the issues,
de-identifier methods and inventories of linkable data.

DR. MAYS: I don’t know if we can do that for the 4^th because
I think, don’t you think they need to come back to us or can we talk about
what we can do for them, I think we have whole committee/subcommittee buy-in —

MS. BURWELL: I thought this was just discussion —

MR. LOCALIO: When are we going to talk with the Privacy Subcommittee?

DR. MAYS: Oh, that has to be done and we actually will have a separate
meeting so we’ll have our breakout and then what they’ve done is
scheduled it so that, wait a minute, Gracie, do we have —

MS. WHITE: [Comment off microphone.]

DR. MAYS: That’s what I thought, I think we already have the meeting,
here it is. And then do we go, what it is is that we’re going to their
meeting? Do you know, somebody figure out, we did these times so that we
wouldn’t meet and overlap but it isn’t clear —

MS. JACKSON: We were trying to get some overlap discussion with Privacy —

DR. MAYS: Oh, it’s an hour, somebody is coming to us, thank you Debbie.

MS. JACKSON: — change the schedule to accommodate attendance with Dr. Cohn.

DR. MAYS: So they’re coming to us our last hour, is that it? Okay,
thank you.

MR. LOCALIO: What we should do then, Vickie, or at least I recommend that we
should do, is that at that time we should report that we had this interesting
conversation and say here are the issues, here’s what’s going on, and
ask for their advice as to what we should be asking of a longitudinal survey or
what we should be doing or how we should handle that. Because some of these
issues may be new to Privacy Subcommittee.

DR. MAYS: Okay, here’s what I’m going to suggest is that from 4:00
to 5:00 that’s Privacy time and that, because then we’re going to
come for the last hour and first what we do is introduce kind of the things
that we’ve been hearing, the issues that we have, and then use this as a
great case example and then make sure that we wind up with enough time to have
a discussion with them about how to proceed, what we can do jointly. So how
about that? So we only have two hours —

MS. BURWELL: Let me just understand something, was the Standards
Subcommittee coming to our meeting, or did I just misunderstand? I probably
misunderstood something.

MS. JACKSON: As far as I aware it was the Privacy contingent you wanted and
at one point we had them scheduled at the same time —

DR. MAYS: Usually they are.

MS. JACKSON: — and we just couldn’t do that side of the schedule. So I
think we worked it out in the last telephone conference call that Mr. Rothstein
was going to come or someone from Privacy for the Populations portion and then
discuss those issues, as you did last meeting, I think you did that before. And
it’s just to make that clear to them in an email communication some of the
items that you want them to start considering, things like that.

DR. MAYS: Debbie, are we still going to distribute the report that we have?
Stay here for a minute because I’m looking at the whole agenda now and
we’ve left the pop reports on, should we be distributing that to people or
should we be talking with people about it?

MS. JACKSON: For the meeting, for the full committee meeting? It’s
really what you want to do, it’s just up to you, it’s totally your
decision.

DR. MAYS: Okay, it doesn’t have an action item on it, alright.

Okay, so I think that we can’t have a whole lot of agenda items for the
breakout because instead of three hours we have two and we have an hour that is
very clear that, it’s very clear what we need to do for that hour so I
honestly think that those three items are it and I think that’s it.

Okay, anything else about Privacy that we want to be prepared to do?

Alright, Debbie don’t go yet, sit down, I was about just to ask you
about the writer, but I was about to ask you, you had a discussion with her —

MS. JACKSON: We’re in discussions.

DR. MAYS: So, okay, it’s like the procurement thing, we can’t say
anymore. Then I’m not prohibited from talking.

Okay, where we are with this report and I’ve kind of heard
people’s struggles and frustrations, etc., and really wanting to move this
along and feeling like we keep having it as within our grasp and then we all
get kind of revved up and with short deadlines and then it kind of falls. So we
have pieces to the report right now which is we have the comments that we need
from NCHS perspective, we have comments, we have new sections written which is
the privacy section, and I’m trying to think if there was any other, no,
that’s it. Eugene put in a bit about GEO coding but that was like some
editing in terms of adding some brief comments in terms to beef up the GEO
coding section.

So what we need now would be the following steps is that we need to be able
to have assistance, and I think our assistance is going to come from a writer
to do I think three tasks. One is to take things, all these different edits and
pull them together and to bring up a working document which has a suggested new
outline for it if possible. Two is to look at what in the past we called
recommendations which then moved to being strategies and getting rid of
duplications and trying to reorganize how those strategies should appear. And
then three, I think that what we’re looking for from the writer is an
overall sense of an organization for the entire document because I means she
had some ideas about even how visually should this thing should look so that it
can go out to our community groups so I think that part of what we want to do
is to also make sure we have time to do that. So those are I think three big
tasks for her, for the resource that we’re going to use to be able to do.

Is there anything else that anyone can think of that you think is necessary
to take into consideration for the report? Russ, in terms of your sense of your
privacy, the privacy section, do you feel that you need to have, like how close
to feeling like it can be integrated in or do you feel you need to have any
other discussions —

MR. LOCALIO: Oh yeah, I mean my feeling is that that piece will not fly
until we have some type of buy-in from NCHS, it’s as simple as that.

DR. MAYS: We actually do in the sense of I think what Jennifer says is that
rather then it looking as if well this can be done, that instead it is looking
at it like these are areas we need to look at.

MR. LOCALIO: One of the things they can do is I think if we’re giving
her email I can make some edits and then —

DR. MAYS: We actually thought we originally were going to give it to the
resource person to just shape it in that direction.

MR. LOCALIO: Well, whatever you prefer but I have the capability of making
edits.

DR. MAYS: It’s your piece so why don’t we let you start by doing
it and then we’ll give it to her.

MR. LOCALIO: I think what we really need is to get the benefit of the
Privacy Subcommittee and get a reaction as to what they, how it fits them, also
this information from the childhood survey, I mean it’s very important and
of course there was no mention of that at least as I recall, I didn’t
mention anything about that. But again, the theme should be at least as I see,
we want to have consistent policies for all these things to make data available
as widely as possible. That’s one sentence and so if the childhood survey
has a bearing on it we should include that as well, and whatever they’re
going to dream up. I don’t know, I have to see what they’re coming up
with. They’ve got some plans I’m sure, it’s the same issue so
they’ve got to have some plans. Well, we would hope they would have some
plans.

DR. MAYS: Well, they’ve been pretty bombarded and have heard a lot of
testimony and I know Sara really well as a straight shooter so I mean
she’s worried about a lot of these things and I think that worry has been
translated into workshops, etc., so it may be useful for us to try and go
online and even see some of the papers, I just don’t know if Audrey was
kind of signaling that the papers may have been done but if they’re on the
website that may be a different story so we’ll do that.

Okay, let me ask for a little bit advice in terms of the time that we have
on the agenda for the full committee on the populations side. One of the things
I had thought about doing and the question is whether this is good or bad but
unfortunately I really thought that by now the writer would have been able to
at least look, the resource person would have been able to look at something
and reorganize it but we’re at a point where that hasn’t happened so
the question is should we take the report that we have and at least walk
through it with the group, or is that going to be a worse decision for them
when they see it and it’s not organized that we’re going to end up,
it’s almost like a grant, you put it in and if you really didn’t do
the work well enough you get a bad score and it’s harder to recover from
the bad score. So you have to be very careful about these things because the
scrutiny you get the second time around makes it somewhat harder so I think
given that through the best of efforts it just didn’t move what should we
do? You know the group, I’m wiling to get up there and do whatever so you
just tell me. If I wasn’t a successful investigator I wouldn’t
because otherwise I would say please don’t judge me on this, judge me on
my own body of work.

MR. LOCALIO: If you submit to them a document that is not obviously polished
you’ll get hammered, you’ll just get hammered. What you can do is you
can present the issues and say well there’s new developments here as
we’ve heard today and that we’re including those but here are the
issues, what we’re doing is we’re discussing them with Privacy
Subcommittee, and then say what guidance do you have rather then give them
something that they’ll try to wordsmith at the same time they think about.
When they get a polished document then they’ll hammer it with wordsmithing
issues.

DR. MAYS: Okay, so what then I will take from this is that maybe what I
should do is to try and pull together an outline —

DR. LENGERICH: I was just going to say I agree with Russ’s comment to
not present it to them, I’m not sure what exactly to say but I would agree
with him about not presenting it.

DR. HAYNES: Why not put together though a good Power Point presentation of
the points made in the report so that it looks like you’re organized, or
at least you have something concrete —

MR. LOCALIO: Or at least an outline, it doesn’t have to be a movie.

DR. MAYS: Let’s put it this way, if I think I can get it done given I
don’t even go until Friday, well, I get home Thursday night, but if I can
do that I’m happy to do it. I mean there’s a level at which I know
these issues but there’s a level at which I have to think through them as
to well, what’s important for this group, I mean I could just stand up and
talk through the issues but it’s almost like I don’t want to create
false expectations, you get up and do a presentation and then they have, you
know this is a group that has a lot of opinions —

MR. LOCALIO: — just giving them headings of the report, Vickie, give them
the headings of the report, I don’t think it has to be a separate
presentation, it just has to be the outline as you mentioned. This is where
we’re going, this is what’s been developed —

DR. MAYS: That might be good, we’ll do it as this is what you should
receive next time and kind of in the spirit of having had discussions about
population health we want to talk about what some of the issues are that
surround it.

Now I will do that if I have a backup crew, it’s like being a singer,
you’ve got a backup crew, can I have some backup here?

MR. LOCALIO: Who arranges the seating? I’ll be next to Vickie on one
side and we’ll have Gene on the other side.

MS. JACKSON: Not only do you have a backup crew, Vickie, you’ve got
your commentary, you’ve got that and it hasn’t been done –

DR. MAYS: Thank you, Debbie, save me. But just remember if it doesn’t
go somewhere you just sing loud in the background here.

DR. LENGERICH: I was just remembering about the commentary, I think
that’s what it was, we’ve got a document or something back there that
provides some of that outline and I wouldn’t stray into too new a
territory with this presentation, or make too big of promises.

DR. MAYS: Gracie, can you make sure that there are copies of —

MS. WHITE: Will it fit in the book or what?

DR. MAYS: That’s right, it hasn’t gone out.

MS. WHITE: It will go in the books tomorrow.

DR. LENGERICH: Didn’t that already go out though? The commentary —

PARTICIPANT: The commentary has been completed, we’re talking about
putting a copy in the book.

DR. MAYS: Just put a copy in the book because yeah, I can’t give you an
outline, just put a copy in the book, yes, that would be good.

Okay, we have one thing we’re bad about, that I was bad about and I
said that before which is I didn’t put on here the research agenda and
it’s for the —

MS. BURWELL: Marjorie requested us to put a little bit of time on our
breakout session at the next meeting to discuss our research ideas so I see on
the following day we’re going to have an hours discussion on it, so we
need to factor in a little bit of time on our agenda, our breakout agenda —

MS. JACKSON: To discuss what you prepared, it’s a comprehensive
document so —

MS. BURWELL: In the interim people can feel free to look at it and make
comments and we’ll whittle it down.

DR. MAYS: I have a suggestion. Can you send an email out now, like today or
tomorrow or something and say we need, because we were going to have a phone
conversation about it and I had asked people to respond by email and they
didn’t so can you send an email out and say this is the materials that we
have, what we’d like to get your sense of any that you would like to see
some enhancements to or areas that you think should be added. And then the
discussion that we can have is only on things that are added so it will be
shorter, the rest of these are things we already voted on so let’s not
like try and undo reports so we really only need enhancements or additions. And
then from the email what that will help bring about is how to focus the
discussion.

Okay, anything else people? You’ve been great to be here today, I mean
we’ve made a lot of progress but again it’s like that’s never
our issue is in terms of generating the work, the issue is the resources needed
to get the work done. And I don’t know quite, no, I do, I will raise these
issues as to the resources that we’ve needed, I mean some of you have
talked about comparison of resources, some of you have talked about needing
things to come to fruition, and I agree, so I will think about the best way to
see that this happens but at least those things that were decided today we have
very specific people. Is that right for each of those? Can we do a quick review
with the staff and associated with Privacy and Confidentiality?

MS. BURWELL: Myself and Edna.

DR. MAYS: Okay, these are on our side. Quality Workgroup?

MS. BURWELL: We did not discuss that.

DR. MAYS: Yeah, we haven’t discussed our side for Quality, we know they
have a whole list on their side and I think this is maybe where we go back to a
discussion with Harvey, with Catore, who else, and Stan, those are three people
on our side who are Quality people so let’s ask them. And mental health

MS. GRANTHON: Cille, myself and I think Suzanne.

DR. MAYS: So we did fine on that. The National Child Study? We did fine on
that, we don’t do need to do anything in terms of the BSC, and in terms of
the Population Report that that’s Jackie Lucas has been kind of designated
as key and Nancy Breen —

MS. LUCAS: I’m sorry Vickie, this is Jackie, I didn’t hear what
you just said.

DR. MAYS: That the Populations Report that you are considered one of the key
staff for the Populations Report along with Nancy Breen.

MS. LUCAS: Okay.

DR. MAYS: And you’ve already been playing that role.

MS. LUCAS: Yes, I just didn’t hear what you said.

DR. MAYS: I just said you were going to have it finished next week,
that’s all. Listen to her, okay. While we have you online, I didn’t
realize it, just one quick thing, do you have any information or was Randy able
to come on, do you have any information about the Child Study that NCHS is
doing?

MS. LUCAS: No, I emailed Randy about it yesterday and I didn’t get a
response from him and I was in training this morning so I didn’t even get
to find out if he’s here this week, he may just not have checked his email
but I didn’t get a response back from him. If I do I’ll certainly
forward that information to you.

DR. MAYS: Okay, if you could just follow up with Randy and see just so that
when we have the meeting by the 4^th we just have some idea that
would be useful.

MS. LUCAS: Okay.

DR. MAYS: Thank you. Do we have anybody else online? Okay, thank you folks,
seeing that we have completed our agenda I want to thank you very much and say
it’s the end of the meeting.

[Whereupon at 3:05 p.m. the meeting was adjourned.]

Transcript of the October 26, 2004 NCVHS Subcommittee on Populations Meeting

DEPARTMENT OF HEALTH AND HUMAN SERVICES

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS

SUBCOMMITTEE ON POPULATIONS

October 26, 2004

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 [8:45 a.m.]

A F T E R N O O N S E S S I O N [1:05 p.m.]

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

A F T E R N
O O N S E S S I
O N [1:05 p.m.]