Multilevel Interventions in Health Care Conference: Q&A session moderated by Russell Glasgow, PhD

Uploaded by NIHOD on 05.05.2011

>>>DR. STEPHEN TAPLIN: We want to get your feedback on the
papers and thoughts from the morning. It was a pretty rich
set of papers. We've heard a lot of good comments about it.
It's pretty clear the world is changing and that has a lot of
implications for what we're doing. And I think that was a
good little shot of energy for folks to think that there is a
world out there that's going to be interested in some of the
ways we think. So gather around. Russ Glasgow is going
to be the moderator for the feedback from the tables.
And Russ is a Deputy Director, now at NCI. You all know Russ
from RE-AIM. And as I said in another forum, he's a straight
shooter. That's a joke, a little joke, a little postprandial
joke. Russell Glasgow is Deputy Director of
Dissemination Implementation Science in the Division of
Cancer Control and Population Sciences at NCI where he
provides leadership on numerous research projects to close the
gap between research discovery and program delivery and public
health, clinical practice and health policies. Dr. Glasgow
specializes in the design and evaluation of practical and
generalizable behavior change interventions, especially using
interactive technologies for use in health care, worksite
and community settings. He earned his Ph.D. and master's
degrees in clinical psychology from the University of Oregon
in Eugene. And as I said, we're happy to have him with his
experience at NCI and look forward to his moderation.
DR. RUSSELL GLASGOW: Thank you so much, Steve.
But just for the record, two things -- Moderation is usually
not one of the things that I get charged with. And second, just
for the record, even though I am a straight shooter, I was
not down at the thing where you shoot all the automatic weapons
that they have out in town. Okay. This is your chance now
folks, and this is our last public forum anywhere here.
We'll love to have your written feedback all throughout.
But to counteract the inevitable slump and postprandial
depression that Steve mentioned, I want you to start
exercising by lining up now to get up to the various
microphones around here, this is not only your discussion
point and opportunity from the Session III papers, but also
from cross-cutting things. You know, those things where you've
just had the "ah-hah," where you're having lunch about that
what this really -- how it all came together for you, we
need to hear those. We do have a couple of ground rules
though. Rule number one is we really need to hear from
everybody. So at some time during this next forty, next
fifty minutes now that we have, I'd really like to hear from
each table, at least one thing from each table. In order for
that to happen, if you do the math, you'll be pretty able to
figure out that we're going to be brief. So I'm going to ask
you to keep your question or your comment to no more than
one minute if you will when you come up. So we can hear from
the others and see if there is emerging consensus or not.
Two, as you're starting to line up now, so this is any comments
and key things that emerged, evolved from your table in
Session III or any cross-cutting things you have.
So please start lining up. But just to warm you up, I
anticipated this might happen. So please we need you to come
on down now folks. We don't have any bingo things.
I'm going to just put up a couple of things, while you're
lining up at the various mics here, that I heard throughout
here that I think might be questions or issues that you
might want to weigh in on. And I'm not going to read these.
But what I am going to do is subliminally...I'm going to
cycle through these during the next forty minutes just to see
if there's anything that catches with you. In order to
provoke a little bit, I might point your attention to maybe
the issue or question about policy and the one about the
consumer/citizen/patient level. Second, I thought some things
that I heard this morning were provocative, were
unanticipated, unpredictable outcomes. We haven't heard much
at all about framing of issues, even though a lot of speakers
imply that this is critical issue when we start dealing
with the political and the larger spheres. But how do we
frame issues, health communication science issues?
And finally, a quick warning, I'm going to call out people if
you don't volunteer. So here's among the ones. We reserve the
right up here to call others of you out. But here's a little
early warning if you see your name here, you might be
prepared to gather some thoughts. And with that, let's
just go back and open the floor. Shin-Ping.
>>>DR. SHIN-PING TU: Shin-Ping [inaud.] Washington, Seattle.
And I'm representing table 17 which was really excited
about the opportunities for potential fixed science with
health care reform, et cetera. And I think in particular,
table 17 was really looking forward to a paradigm shift,
especially with respect to opportunities for national
experiments, as well as existing and hopefully additional
information platforms to collect data where you can then have a
information of small studies. So that was our first comment.
And we would love to hear more about that. The second is
regarding the genomic component, I think our table had a lot of
health care providers as well as a lot of public health
providers. And we really felt that it would be important to
have the population health benefit incorporated into that
consideration regarding how to move forward on the genomic
front of health care. And how would that one piece work
regarding the element disentanglement of your
roles and how to be able to have funding mechanisms
that incorporate and further enable [inaud.]
>>>DR. RUSSELL GLASGOW: Thank you, very much.
Again, just under a minute. So that was great.
I'm also going to ask the panelists up here to
also please be very concise. And I will try to exercise or
let you know if you're going on to also keep comments. And I'm
also going to -- unless there's something that I'll let you
override me -- but unless it's really, really burning, I'd
like to try and reserve the responses to either one or at
most two people. I don't know if we need four talking heads
to say I agree with everything. So is there anybody that would
like to take that one? And in the future, if you do have your
question, if you could begin by saying if it's directed to one
or more of the authors or the presenters, if you could start
out with them just to give them a heads up. But is there
anybody... I thank you for those comments and they're duly
noted. Is there anybody that would like to reply to any of
Shin-Ping's comments? Of course, we'll do that. Muin.
>>>DR. MUIN KHOURY: So since genomics was mentioned,
I'm happy to kick off. So I agree with the comments that
were made. And especially, vis-à-vis the idea of
conceptualizing genomics from a population health benefits
angle. Because I think that's been not too much in this
course. It has been traditionally driven by
technology, sort of the more sequencing we do,
the assumption is that the more we do,
the cheaper it gets, the better it will be. Rather than
starting from the pull forces of whether or not this
actually can work to improve health outcomes.
And that's a population perspective. So I agree with
that being somehow integrated into the discourse.
So, thank you for your comments.
>>>DR. RUSSELL GLASGOW: Okay. And thank you. One thing
that came up at our table I know was that Muin's paper I
think pointed out more sharply maybe than some of the others
the potential for harm and waste also which is a huge
issue that somebody might want to follow-up on. Deb?
>>>DR. DEBORAH BOWEN: Deb Bowen from Boston University.
Perhaps this is a little bit yesterday's conversations as
well, but I'll try it now. One tension that we noticed was
between the idea of health and the idea of health care.
So health care is in the title of our conference. And yet,
several of the presenters brought examples as data models
that were from other than health care. And so we struggle
a little bit with the role of health care in say an onion
or even a pyramid perhaps of interventions.
I mean, is health care a level? Or is it a goal?
Is it both, depending on how you look at it? And are me
or you or folks who are writing these papers looking at health
care as how exactly? It ties into the definition of "level"
that sometimes over the papers perhaps got away from us a
little bit that there might be some need to have a little bit
of consensus about it, and a very specific comment about
level. When one defines MLI, why does an individual
intervention... why is it required that individual
intervention and more individual outcome be a part of
an MLI? Why is that? What was the thinking behind that
decision? Many folks at the table yesterday and today came
up with examples from the literature from their own work
on MLIs, things that might qualify as an MLI, that had
neither an individual intervention component nor an
individual intervention outcome, because they leaned on
the data that had been collected before. So just a
thought. You so clearly define individual as part.
Certainly, it is. But why the definition of MLI?
>>>DR. RUSSELL GLASGOW: Okay. I'm going to let Steve take
the first shot at that and then Dick.
So if we can, we'll keep it to two. But just to try
and paraphrase... Deb, you might want to stay there.
You might have a chance just for a very, very quick back and
forth depending on what happens here. We're all going to try
and be quiet. I'd say this is the first kind of provocative
issue that we've had since I'm here which is great.
But I heard "is health care a level" or is it the definition
of the whole thing there? I also heard maybe a challenge to
the definition of the onion or a friendly comment or a question
for clarification on that. So, Steve first and then Dick.
>>>DR. STEVE CLAUSER: Well, I just wanted to comment on the
individual intervention outcome thing, because I think one
thing I mentioned in my presentation, and also Steve
mentioned early on is that there was a real focus in terms
of our interest in doing this that the ultimate thing we're
trying to drive is patient outcomes. And it's not
necessarily that every single element of that knowledge
stream has to have an outcome as part of that intervention
stream. But there must be some kind of framework in which
there's an expectation that it's going to be tied to
ultimately deliver something that's going to improve the
ultimate experience of care. And it doesn't necessarily
always mean in cancer terms survival. There's other kinds
of outcomes that we've tried to acknowledge in terms of the
thing. And I think you're right. Some of those outcomes
that get into the quality of life domains and some of those
other things you have to think about whether you're talking
about health or health care. But largely, we are trying to
focus our attention right now on those things that drive
patient outcomes as part of those care processes.
>>>DR. RICHARD WARNECKE: Two things. I would call health
care a mechanism under Steve Taplin's definition.
It's something that happens between levels.
I may have taken his point wrong.
But that's what I would... that's what I would call it.
It's something that it's an action rather than...
>>>DR. RUSSELL GLASGOW: That's why I asked Deb to stay here.
And as Steve's coming up,
go ahead and finish your point, Dick.
>>>DR. RICHARD WARNECKE: I just have another thing.
>>>DR. RUSSELL GLASGOW: I want to give Deb a quick chance
to respond. Then we're going to have Steve. But I also don't
want to lose one other thing that Deb alluded to when she
talked about a triangle as supposed to an onion or maybe
compatible with an onion, I think what she's alluding
to... Deb, I don't want to put words in your mouth...
was the Friedan type triangle, the CDC triangle and that.
So somebody else might want to address that.
>>>DR. RICHARD WARNECKE: I just want to finish.
>>>DR. RUSSELL GLASGOW: So finish yours and
then Steve and then Muin.
>>>DR. RICHARD WARNECKE: The other thing is that, of course,
from the perspective that I come from, which is disparities,
there are no individual issues. There are differences between
groups and populations that affect individuals. But the
focus really is on resolving the differences at that level.
>>>DR. STEPHEN TAPLIN: I'll just reinforce Steve's comment that
the focus of this effort has really been about health care.
And that's why health care was in the title. The issue of the
process of care is individuals are there and providers
are there. And so process is simply a way of manifesting the
effects of these levels on the health care team or the health
care individual or the health care provider. So we'll see the
process as just a location in which the effects are manifest.
So I don't see process as a separate level in the way that
I'm thinking. We did talk about individual patients. We did
talk about providers and provider teams, those are all
involved in different parts of the process of care.
>>>DR. DEBORAH BOWEN: I think it's interesting that...
I'm forgetting which person brought this comment up,
maybe a couple of folks... that much of our evidence for any
sort of multi-level effort, it comes from primary and
secondary prevention as a topic. And in that sense, again,
we're not always talking about patients, people, individuals
or entities or groups of people. So maybe at least one thing
that the papers could do is as they bring examples up outside
of the patient provider health care system arena to help the
readers kind of grasp what those differences are.
>>>MS: That's a good point.
>>>DR. DEBORAH BOWEN: I did not really in these other
papers, our table talked about this and
didn't maybe follow everyone as well as they could.
>>>DR. STEPHEN TAPLIN: And it's certainly true that you can
have three levels anywhere and they don't have to end up with
the individual. So that's perfectly fine. We just chose
to make sure that we tried to drive it to the individual.
>>>DR. DEBORAH BOWEN: When we look at state tobacco
sales as an outcome, we could probably... folks have
generated multi-level interventions to change state
tobacco consumption levels, not measuring anything
about an individual's cessation habits.
So we count that as a multi-level intervention
that gives us great ideas for the future.
>>>DR. STEPHEN TAPLIN: I think that's all completely
appropriately. But as Dick was saying, we focused
on the individual patient outcomes.
>>>DR. RUSSELL GLASGOW: Muin in just a second,
and I'd like to take the prerogative of the
chair to invite if there's anybody from the audience...
I want to let this go just a little longer because we're
starting to have fun now. So if there's anybody that's directly
going to comment on this, I'm going to give you priority to
come to the microphone directly on this particular issue. But I
will comment there's one other thing up here that kind of
relates to that is the third one down here if anybody wants
to talk to that about the terminology used about
consumers, citizens, patients and how that interacts,
particularly with electronic health information
issues and things as well. But with that, Muin.
>>>DR. MUIN KHOURY: So Deb, thank you for your comment.
The issue of the onion versus the pyramid also came
up yesterday in group seven when I was at that table.
So for those of you who may... who don't know this,
the CDC director, last year published a paper in
the American Journal of Public Health about the health impact
pyramid which he postulated there are five levels of
intervention. And at the lowest level are interventions that
fight poverty and education, increased education which we
normally don't think of them as clinical or even public health
intervention. The second level is the policy work which is
what he calls changing the content, meaning tobacco excise
tax or putting fluoride in the water or whatever. The third
level is the clinical intervention which is the ...
no, actually the third level is public health programs, like
newborn screening when you think about genetics. You know,
it's something that public health departments do.
The fourth level is clinical interventions. That's where the
process of the health care goes. And the last, the top of
the pyramid is health education. And it reflects the
bias of the CDC director that with as you go to the top of
the pyramid, the impact becomes lower. As we go down to the
bottom of the pyramid, the impact would be larger on the
number of people affected. And therefore, policy change for
him actually supersedes health education. And he postulated
that you can attack any problem from multiple levels of
intervention, both from a clinical and a public health
perspective, including health care delivery.
So just to put a bit more salt and pepper on this.
When the recommendation came out last year from the...
>>>DR. RUSSELL GLASGOW: Quickly, Muin.
>>>DR. MUIN KHOURY: So we had an implementation panel
that discussed how we're going to implement it from multiple
levels. And I forced them to think through the onion. And I
forced them to think through the pyramid. They liked the
pyramid a little bit more than the onion. Because it provided
different tools that are not all in the clinical domain, but
outside the clinical domain. But it doesn't matter. So I
agree health care is part of the health continuum and you
can attack the problem from multiple other perspectives.
>>>DR. RUSSELL GLASGOW: Okay. So I heard a couple of cross
mega level type issues that authors of the paper might want
to keep in mind. I believe these two are both directly related
to the issue. So go ahead, Mike. And then over here.
>>>MS: It's good you spoke. Because I was going to
speak to your genome presentation. But I was at the
table with them and another person here. And I came here
as a person interested in rural health care assistance.
My focus as of late has been networks, rural health
networks. And so, I was very happy to be here.
And I have all the focus on individual patients.
But my stance on multi-level analysis is more
like making the rural health network the level of interest.
And I might look at two or more levels above or one above and
one below. So I was extremely interested in that. And I
thought, well, it was good the message was brought. Well, this
morning in looking at the genomic medicine presentation,
two things came to mind. A colleague from the table here
yesterday said that the levels that we started out with didn't
have... and these are my words... biochemical,
biocellular, physical. So you fit that in. But what struck me
from Dr. Scott's presentation is that you showed a lot of steps,
a series of steps to kind of put it in the place. And it
occurred to me that on a multi-level intervention
approach might be a good one to help along those steps.
>>>DR. RUSSELL GLASGOW: Okay. One person respond. And again,
we're having great discussion, but we're not following the one
minute rule. Muin, it was kind of mostly directed at you,
but if somebody else is more passionate,
I'll let them. Go ahead, you got it.
>>>DR. MUIN KHOURY: I think you're referring to the T1 to
T4 framework. This was not designed by me or anything.
This is sort of the two phases of translation by the IOM.
And it applies to any field. It doesn't really apply to
genomics in particular. I was trying to understand the
landscape of translation when it came to genomics.
So that's why I wrote those papers. But it can also apply
to everything else we do. And I think, Russ you have been
thinking along the same lines as well.
>>>DR. RUSSELL GLASGOW: We have been scheming on that.
I'm going to just without any explanation do a little
subliminal image her as we take the next comment
on this, but maybe an additional way.
But the meta issue that I heard here is are these levels
fixed? Is there a right number, six or seven or whatever?
Or is that an adaptive learning process too? And that when
we're focusing on, for example, rural health or maybe we're
talking about indigenous communities or something.
Do we adapt the levels, keep the same principals, or not?
And we got a response from Martin.
And then I will get to you, sir.
>>>DR. MARTIN CHARNS: So I think that the example that I said
in the Yano et al paper would argue that the levels are not
fixed. And in that study, they had to look internally within
the organization for multiple levels. And in other places,
you may not. So at least that's my take on it within a minute.
>>>DR. RUSSELL GLASGOW: And is Dr. Scott still here today or
not from his talk yesterday? I was just wondering about his
take on that. I'm thinking from what I heard, I'm not familiar,
the work that he might concur with that and that
you look for the political points and things.
>>>MS: I actually wanted to comment on that third bullet
related to the terminology we used, whether it's a patient
or a citizen. And I actually want to push it even a bit
further, based on the discussions that we had at
our table. And I was prompted by Dr. Clauser's message
that we really hadn't talked about the role of individuals
except as sort of the targeted of all of our work.
And I, as a person who's studied community psychology
and empowerment models a great deal, I actually think we need
to think about one outcome of our work, not only as
healthier people, but as empowered people.
and as empowered consumers. And I don't mean that just as
a rhetoric. And I actually think that by empowering the
patients, consumers, citizens, that we work with, to be very
influential in their care, either by shopping with their
feet, by directing where their dollars go, by sharing
information with one another about where they can get good
health care, where the health care's not so good. That we can
begin to help foster organizational and policy level
changes. In our discussion, I said, I wonder how many people
really understand the statistics that were presented yesterday
about the quality health care in the United States.
If we were to empower consumers to know that if you're shopping
for cancer care or whatever, that 50 percent of the
time roughly your life would run into substandard care in your
community with a doctor that seems very nice at the hospital
that seems very friendly. So I sort of facetiously,
half facetiously, said that we're running interventions
medical mistrust. You know, at least empower people
to know what they're getting into. And I will say, the last
thing, that this seventeen year cycle of getting innovation
into practice is that we can counter that by saying that
every contact that we have in community based interventions
or organizational interventions empowers consumers,
that we can do that without FDA approval and without
years of churning through a system that we've always
been about helping people take more control of their care.
>>>DR. RUSSELL GLASGOW: Okay. Thank you.
This is the kind of great discussion that we want.
It also brought up some issues related to the second point
here, maybe about framing issues and how we understand or don't
understand issues and framing them at the different levels.
And I'm going to allow one more response here from the
panel if anybody wants to say anything about that.
And then we'll move onto I think a new... are we kind
of reaching 90 percent on this anyway?
>>>FS: I'm going to go way back here.
>>>DR. RUSSELL GLASGOW: You're going to go way back.
Okay. So we're moving on, new topics.
So come on up for new topics, different areas.
>>>DR. MARY STEP: Hi, I'm Mary Step from Case Western Reserve
University, representing Table 15. And we had a series of
wonderful conversations. And I think we're all very
enthusiastic about exploring this idea and applying these
ideas to our own work. However, with each conversation, we kept
coming back to the idea of what really is a multi-level
intervention, and this fundamental idea of what are
the levels and what kind of interventions we have.
And although we all agree that this is not the time to be
proscriptive about these things, we wanted to express
our hope that this learned community, that literally
brings to the table many perspectives on this work, that
we can capture a little bit of lightning in a bottle, and with
the upcoming publications, offer some guidance to those of
us who are setting out and want to do these designs.
We've talked about what makes interventions go.
Why these MLIs fail? What's a proper MLI?
What are the strategies for MLI? Some of the fundamental
criteria are very important to us. And what really clued us
into this was perhaps a throw-away comment that many of
the posters weren't really multi-level interventions.
And that really stuck with us. And that although we'd like to
talk about all these different things, we're hoping that you
could provide some guidance as to what it is. Because if we
want to respond to calls or [inaud.] whatever is going
to exist after this, we need to define this.
>>>DR. STEVEN CLAUSER Part of the issue is, again, trying to
get clarity over interventions, more than one, right?
At least more than one level, whether it be patient, provider
organization. And I think it is a conundrum because within each
one of those bands is a lot of space. At the organizational
level, you probably could make a case that there were many
levels. One could intervene within the context of the
organization. And this is the step for us to begin this
dialogue to try to clarify how these sorts of things play out
in different types of studies. But one of the things we
talked about that it was very clear, at least from our
literature search, that we're still... it'd be great if we
could argue whether studies that had two or three or four
interventions were actually multi-level or not. We're still
single level, single target. So we've got a long way to go I
think to even have that kind of conversation. And that's what
we're trying to engage here in this work.
>>>DR. MARY STEP: Oh, yeah, we did that.
I'm just wondering why those posters were
not considered multi-level intervention.
>>>DR. RICHARD WARNECKE: Well, I didn't judge the posters.
But on the more general issue of multi-level, I think that's why
so much emphasis in this conference has been on
conceptualization of the problem. And it's one thing
to answer a direct hypothesis about individuals. It's another
thing as it is in disparities... again, I'll go back to my own
model... where you know that there are different things that
intervene before the patient even gets into the system.
And when the patient gets into the system, may end up
in the wrong system or getting the wrong care. So it's the
conceptual framework you have to work with. And if you want
multi-level research, then you have to consider what are the
levels of intervening variables, not just intervening variables
at one level, but intervening variables at other levels.
>>>DR. RUSSELL GLASGOW: Okay. I don't know if Becky's still
here or not. I don't see her right now. But I thought she
made an important point related to this too which was the fact
that many interventions... and we've had the minor
sub-discussion of this about patient-provider intervention.
Is that one or two or three or how many? But I thought Becky
made a pretty profound point. And it might illustrate or
benefit from an illustration that many things that we even
call single level are frankly, probably intentionally or
unintentionally multiple level interventions and often can
produce outcomes at other levels, again intended or
unintended, positive or negative, too. I don't see
Becky here. But we're over half our time. And I do want to have
a specific call out here. And if some of you note here,
I particularly want to call out... I'm going to let these
two individuals go. But I want to invite those whose names are
here to come up and comment on some of these issues. But I
also do want to throw out the combination of the Don Berwick
challenge to ask what are you going to do?
In addition to telling us at NCI what can we do for you,
what can you do by next Tuesday to start making this happen,
making it real to advancing the science. And if your table has
not commented yet, I really want your table, I want to hear
from each of the tables so come up here.
>>>FS: Hi, I'm [inaud.] Cancer Center. Our esteemed leader,
Ronald Myers, left. So I get to speak for our Table 10.
And we said many of the things that have already
been said. And I think one of the issues is around are these
papers really focused on health care delivery systems as
opposed to the broader issue of health? And I think that's
where the patient/individual issue keeps coming up. And I
think some of the examples are more public health focused.
Yet, it seems the focus is health care delivery. So it was
just one of the comments as you look at these papers to really
be clear about that and maybe even organize them that way.
Another point was really I think for us we really got
stuck on the issue around health disparities. And I think
having one paper -- we actually like to open this up to the
authors to really think in each paper what the issues are for
diverse populations. Because I think that's so important.
And I think again sometimes we sort single it out and put it
in one place as opposed to thinking about it broadly.
And that's really a consensus.
>>>DR. RUSSELL GLASGOW: In fact I'm sorry,
but in the interest of time.
>>>FS: Is that my minute?
>>>DR. RUSSELL GLASGOW: It is I'm afraid, a little over.
But stay there because we may give you a chance for a
response. Obviously, the applause, the interest meter
went up on the health disparities issue.
So I'm first going to give Dick a chance for a first
brief response. And then I'm going to invite anybody else
in the audience again to follow this thread just a little
bit, to drill down this interest. So if you come,
Beti, don't strangle me afterwards, but you may have
something to say about health disparities. But I'm going to
once again allow people. And then we are going to get to
Beti next after that. But if there's anybody else that has
something specifically to recommend or state or they want
to get on the record about health disparities. Dick.
>>>DR. RICHARD WARNECKE: Well, I think... and this probably
goes back to why some of the posters aren't multi-level,
if they're dealing with racial or ethnic groups, is that most
people don't understand the difference between population
health and determinants of disparities. In one case,
you're talking about mammography's a good example.
It improved the health. Everybody went up.
But it didn't correct the differences between the
outcomes for some people versus other people.
And to think about that, you have to think about it
differently to begin with. Ernie's comments, just to
respond, it's not zero sum game in this. So when you get a
coalition together, you're going to focus on one thing and
you're going to address that. Another coalition may focus on
something else. And if they have the political clout,
they're going to get it. So I think it's a mistake to think
about if you get a change in breast cancer, you're going to
lose the chance to get someplace else. That's not
addressing disparities. Addressing disparities is,
I think and most of the writers that talk about it, taking care
of people who are at different levels in a hierarchy, social
status hierarchy, because it's really socioeconomic.
Race falls away, as I pointed out. So I think that's the
issue and understanding that you're always talking about
comparative effects basically and what causes the differences
between those groups. If you focus on that, then you're
talking about disparities. If you're talking about why
African Americans don't do something or why Hispanics
don't do something, you're maybe talking about population
health. Or you may be talking about cancer control.
But you're not talking about disparities.
>>>DR. RUSSELL GLASGOW: Okay. Quick response if you'd like.
>>>FS: I think we agreed. I think the point is that on
each of the papers as you look at those issues,
to address that. As opposed to only one paper.
>>>DR. RUSSELL GLASGOW: Point taken.
The committee definitely has it out?
>>>MS: And I may be a bit opinionated on that issue.
at last, believe it or not.
>>>DR. BETI THOMPSON: I'm Beti Thompson from the
Hutchinson Cancer Research Center, and really what
I have to say has a lot to do with my work and our work
on health disparities, but it was triggered by the comment
that I heard from Steve Clauser which was let's not start or we
shouldn't start this work from scratch. And I want to talk
just very, very briefly about the whole public health model.
The public health model has been oriented all the time to
working in a multi-level intervention format.
We've always looked upstream to find reasons why people do what
they do. We've always looked at social determinants of health.
We've always looked at health disparities. And I think one of
the things that we need to do coming away from this
conference is recognize that we have a good bedrock on which to
rest our activities. And I would personally like for the
supplement to make note of that. I think it's important
to recognize that we aren't starting a whole new thing.
This is something that's been going on for a long time.
And at the risk of dating myself and not somebody who can
respond to this, the old days of COMMIT, when we were talking
late '80s, early '90s, was a classic example how we do
multi-intervention research. Today we have the CPHHD of
which Dick is a part. And those are all multi-intervention
components. So my point is really just to say it's not all
something new. There are many aspects of this. But I want us
to remember the history of which we're doing.
>>>DR. RUSSELL GLASGOW: Excellent point.
Since you called me out among others,
I'll see if anybody else wants to respond.
But I'll point to the next to last comment here about
international issues. And I think this relates to something
Dr. Scott said yesterday, Beti. And it has to do with policy
level interventions. And for me, one of the implications of
what we heard him say is that as informative and important as
it was, today Kelly Devers' talk, I think about the context
of health care reform. That is critical. That's the dominant
issue driving us. But we are one country. We're at one point
in time, at one point in history. There are a whole lot
of lessons from other people in the world that have other types
of health policies and other contextual issues and lessons
to learn from that as well. I don't want to put words in your
mouth, Beti. But, and I'm trying to keep to my own one
minute rule here, but I think some public health people might
want to say or have a greater articulation of the thing about
the patient requiring and saying the mandatory patient
level outcome. Even if you're patient-centered, I think some
public health people might say, wouldn't building a healthier
community and outcome and index of that be an equally valid,
or at least address the crosswalk between those two?
So do you want to reply to that, Beti? No?
Okay. Anybody else on the panel? Steve.
>>>DR. STEVE CLAUSER: I just want to say a brief thing.
I appreciate those comments on not starting from
scratch. I think you said it much more eloquently than I
did. But I was trying to convey some of the earlier things.
I think even though we may think that there's a small number of
studies, those studies are very, very important in helping
us really get into granularity about how these things work,
even in community health settings that can help us think
about this. And one of the things I think we've seen in
some of the literature review is that the way these studies
were presented in the literature, there's multiple
publications that you've got to pull together to really get the
story. And I think we've got to be a little creative to try to
paint that picture a little more clearly in the way we go
about trying to assemble that literature. And it's something
I think we're going to take back for our journal article.
>>>DR. RUSSELL GLASGOW: Okay. We have fifteen minutes left.
And we're going to get to these two individuals here who have
been quite patient. Then I'm going to call out Tom Vogt as
well as anybody else that would like to respond to the issue
about or anybody from the VA. Becky isn't here. Maybe Martin,
you'd like to take it from the VA. But I'm going to ask what
in particular can these organizations that have these
state of the art, at least for the U.S., electronic health
care systems? What can we do? And in particular, this slide
here on the right hand side, how can that relate to the
ubiquitous media and technology things, the social media, the
patient engagement, and how they're getting their things?
What are the implications of that revolution and phenomenon?
That's the other huge contextual factor that we're
facing with our society today. So I'd like to call out Tom and
Martin on that. I'll give you a couple of minutes to think
about that while we take these. And I forget who was first,
Mark or Deb. Why don't we let you go first?
>>>MS: [inaud.] University. So I'm representing Table 14.
One comment I believe in this public health discussion and
the people at our table came back to a few times was the
idea that these multi-level models, that we are somewhat
drawing upon a couple of different traditions.
And so with one multi-level model going from the patient
to the provider to the organizations is more health
services driven. And then another model we're taking
from [inaud.] individual to the families to the community,
more a public health model. And so there's some integration
or thinking how those relate that will be a helpful outcome
of this conference. And I think one thing -- because
the [inaud.] is related to -- perhaps it could be some of
this issue of incentives, in terms of health care
organizations aren't necessarily incentivized to address
the community in which they reside [inaud.] quality of
the people who come in through their door. So there's a little
bit of disconnect there between population health and
incentives of health care organizations. Maybe ACS will
get at that some. But maybe this is why these population
health statistics could be one of the causes for those poor
health statistics that help bring this conference.
That's one comment. Looking also...
>>>DR. RUSSELL GLASGOW: Very quickly.
>>>MS: Got it. This will be a short comment. From the VA
perspective, one question that I have coming away from this
conference, not necessarily an answer, but a question I'll ask
myself more in the future is how implementation science,
which is something we made a commitment to in the VA
that relates to multi-level interventions and how these
are similar and different conceptually.
>>>DR. RUSSELL GLASGOW: Thank you, very much.
Anybody in response? Or just duly noted. Thank you.
Appreciate your sharing that. David, Deb.
>>>DR. DEBORAH BOWEN: You asked when [inaud.]
>>>DR. RUSSELL GLASGOW: I did. By Tuesday.
>>>DR. DEBORAH BOWEN: Okay, by Tuesday I'm going to email
the head of our CTSI. And I'm going to bring sort of
a synthesis of what we've covered here along with
some dissemination thinking as well. And I'm going to
try to get some of this considered within our CTSI.
Because my sense about the CTSA's or I's or whatever
they're called at our institutions is that they
hit us with a roar in some places, in many places.
And they are really in a good way or bad way monopolizing the
conversation about this topic in academic medical centers.
It's sometimes good, but sometimes bad and we know
things, people here know things, that are not known by
the CTSI or [inaud.]. And so I was intrigued that I think the
only speaker that mentioned this current system of delivery
of funding and support at an institution was Muin, and he's
written a lot about this. But I think that maybe all of us
should at least understand how the team work or team force
systems work in our field and where they don't work and sort
of push that ahead a little bit where they don't work.
And that's what I plan to do by Tuesday.
>>>DR. RUSSELL GLASGOW: Thank you, very much.
I would love to hear thirty seconds ...
we're now down to thirty second talks. I've got you Muin here
in just a second. But thirty second comments about what
others of you are going to do, and if I can find that, by
Tuesday as well. And, in fact, what I'll do is I think the CTSA
is a pretty important issue and opportunity. I agree with you.
It's a train that's out of the station. But it still could be
guided down one track versus another. So I'd love to hear
if anybody has thoughts about multi-level intervention
implications. I'm going to let Muin respond to this. And then
I want to engage Tom on this EHR, PHR electronic health
record. What's the key thing or two we can do with regards to
electronic, Tom and Martin, health things? And then I
particularly invite if anybody has anything else on CTSAs. But
we're going to finish this by allowing, my math says
forty-five seconds now, for one key take-away point from each
of our panelists up here. So I'm giving you an early warning.
So Martin, you get two. You get your response to
Tom's on the EHR, and then the one other key thing that each
of you heard in forty-five seconds or less. Tom?
>>>DR. THOMAS VOGT: Well, to do this quickly, I'd like to
spend a lot more time. But HMOs, particularly those with
professionally autonomous research centers [inaud.]
research effort have spent many years trying to do
research in that setting. They haven't always been honored
or respected by their parent organizations, however.
And the wonderful things they do have often not been
implemented. However, in the process, the HMOs,
particularly those in the network, have begun to develop
EHR systems that look beyond the use of those systems for
the provision of care that began to use those systems to learn
from what we do. And that's resulted in a national virtual
data warehouse, as I mentioned yesterday, it has over
15 million people in it. And we have a long way to go to make
that easy and cheap to use, but it's usable. And that's what
they need to do. They need to improve uniformity of coding,
the ease of data access and ease of data pulling and the
quality of the data. They also need to participate at higher
levels of management in particularly multi-level
efforts than they have been doing. I'd say that there's
some going on there. It's mixed. They are not acting
proactively at present to agree to test more rational forms of
care than it is to reduce that necessary care. I think because
there's a fear of being perceived as denying care
[inaud.] But they need to support research at the
implementation level. They've supported it at the sort of
intellectual level. But they have been reluctant,
not impossible, but it's very difficult to change care in the
system to do things differently. It takes an extraordinary
amount of leg work and arm twisting and so on
to get to that point. And finally, a point that you
mentioned, Russ, regarding social media, many members
don't all do the same things. I'll talk for Kaiser Permanente
that I know the best, has begun actively to participate in the
interventions over the Internet. And, in fact, had
begun to design systems that are supposed to shift care from
the office to the Internet. And what they need to do with those
is to evaluate them more rigorously and evaluate the
cost of all this as well. And to do that, they need to get
all the researchers. The researchers for these systems
need to collaborate more aggressively with the
university researchers. Because we simply don't have the arms
and legs necessary to take adequate advantage of the
extraordinary data resources. And I want to tell you those
data resources are beyond your wildest dreams if you're not
familiar with them. You can dream great dreams and
ask questions you haven't ever been able to ask before.
You are also frustrated and have problems
when you try to use those data, but [inaud.]
>>>DR. RUSSELL GLASGOW: Thank you, very much.
I think that was worth two minutes. So well stated.
I'm going to offer one friendly provocative amendment
to that. I would say that not only the HMOs, but most other
health care systems that I know of, despite the billions that
we've spent, despite the huge advantages that we have, we
have one glaring hole, i.e., patient reported measures and
outcomes, and how do they relate, and not being collected
systematically on a national basis. Now, the VA might be
doing that. And with that, I will turn to Martin. And then
we're getting about time for the 45 second, one point summaries.
>>>DR. MARTIN CHARNS: There are other people in this
room I know from VA who probably can say things that
are of more importance than what I will say. But I just
want to point out two things about the electronic health
records in the VA. We've been talking about them in terms of
their research value. And, in fact, one of the attractions
eighteen years ago when I moved to the VA was the ability to
use these data. I mean, it was like a gold mine. But the other
thing I want to point out is how the electronic health
records and associated IT can be used for changing practice.
And there certainly are...there is literature around things
like the use of reminders and other assists that come from
the technology. So don't forget that side of it which has also
been used quite advantageously in the VA system.
>>>DR. RUSSELL GLASGOW: Okay. We're going to start now
and we're going to end with giving Martin the last word.
We're going to do our 45 seconds, starting here.
But when we get halfway through, we're going to pause for
a final call out by somebody who thought he was going
to sneak out and is now non-attending like a third
grader by the name of Kurt Stange who is called out.
And he is going to tell us what PBRNs and primary care is
going to do by next Tuesday to advance the multi-level agenda,
research agenda here. But I'll give him a minute and a half to
think about it before he comes up. So, please start.
>>>DR. ERNEST HAWK: So I've been fascinated over the last
day or so to learn more about multi-level interventions.
I think they're extremely complex. I think they're very
promising. But I remain unconvinced that necessarily
moving in three different directions all at the same time
is necessarily going to afford better outcomes than doing
so through individual parallel efforts or serial efforts.
So I would love to see a demonstration project that
would prove the benefit of this to really put enough impetus
behind it that we could actually change the minds of funding
institutions, et cetera. So I'd love to see this whole concept
tested in a rigorous manner to demonstrate that it's
more effective, more efficient, et cetera, like we hope it is.
>>>DR. RUSSELL GLASGOW: A CER challenge called out
to the multi-level thing. Thank you. Good model. Muin.
>>>DR. MUIN KHOURY: So can I get two 45 seconds?
>>>DR. RUSSELL GLASGOW: Two 45 seconds?
I'll give you one minute and fifteen seconds. Last offer.
>>>DR. MUIN KHOURY: Okay. So one of the reasons I'm
excited about this conference and one of the reasons why
I spent so much time traveling between CDC and NIH is that
I believe in the value of joining both a research
institution with an action oriented institution that's
trying to solve the problem. It's not just trying to
study them. So by definition and here what I want to
say is that whatever we call this, health care or health
or public health, in order to attack the problem,
whether it's smoking, obesity or whatever -- you need
multiple levels. You need that from a solution perspective.
It's just that the literature hasn't caught up with that
idea. People are studying one thing, one thing at a time in
their own institution. I mean, public health -- and I put that
as a population health perspective, that worries about
disparities, worries about access, worries about
everything -- is trying to attack the problems from
multiple viewpoints. And we don't have the scientific tools
to do that. Health care is an arm of public health. So for
those of you who think that medicine is a branch of public
health, I'm with you all the way. And we're going to have to
find a way to all get that science to back up the actions.
So, a few years ago, I wrote a paper called Will Genomics
Widen the Schism between Medicine and Public Health?
And my thesis is that genomics provides us with an opportunity
for the first time to take an inherently biomedical model and
put it in the context of population health. Because with
that, we might be able to finally bring that biomedical
community and the basic science researchers to think like
health services or public health people to see the
outcomes of all their technology, to see whether it
will benefit, but not harm people.
>>>DR. RUSSELL GLASGOW: Thank you. That was worth 1:15.
But now all my other friends are going to be mad at
me because they don't get 1:15 too. So, Dick. Get up here.
>>>DR. RICHARD WARNECKE: My take home point is that from
what I've heard about the Affordable Care Act and
all the other things is that whatever interventions or
innovations come out of these multi-level studies,
one of the questions should be will this advance or will it
narrow the disparities that currently exist?
And it's something that you do at the start of thinking
about new things, not after you've done them.
Because by then, the differences are too big to fix.
>>>DR. RUSSELL GLASGOW: Well said. Okay, Kurt and then
we have our last two comments up here.
What is primary care and the PBRNs and the Annals
of Family Medicine going to do by next Tuesday?
>>>DR. KURT STANGE: First, [inaud.] might want to do when
something comes up, send it to the different committees
with a cover letter that tries to frame it appropriately.
[inaud.] might look at that and say, you know, the fundamental
approach that [inaud.] networks have is a committee
based participatory approach. But we've been focused lately
on just [inaud.] community and most [inaud.] now are really
trying to take a broader community perspective.
At the same time though, we get the infrastructure for a lot
of fact-based medical research is really the information,
the technology, the EMR kind of platforms. And the CTSAs
are trying to work with PBRNs to provide that.
So I think what PBRNs might do with this multi-level
research is say that if they could just [inaud.] try to
act on that idea. If we can't do it ourselves,
then we need to partner with others.
>>>DR. RUSSELL GLASGOW: Excellent. Okay. Martin and
then I see Brian just came back in the room.
Brian, we called you out earlier up here. So the question
is what's Implementation Science going to do by next Tuesday?
If you want to respond to that, you have 45 seconds after
you find your way to the mic. But we're going to go to Martin
and then Steve is going to close this session out for us.
>>>DR. MARTIN CHARNS: So three points as quickly as I can.
First off, I take away that the definitions of levels still are
messy. I'm an optimist. So I come away with the glass more
than half full. I think if we get 90 percent of the way
towards defining what levels are, that would be
very helpful. And maybe the next ten percent we put
off a bit. Secondly, and perhaps Brian can comment on this,
this is in response to a question from the audience.
I think that implementation science recognizes
multi-levels. But I'm not sure when implementation
science was in its infancy if it really did.
Because there were a lot of studies that we saw that I
think were single level. My third point is really as I
reflect on all this as an organizational theorist,
I think the concept that we're dealing with is that of an open
system, versus a closed system. And, you know, an open system
is one where the boundaries are permeable. And, in fact, in
practice in the real world, the boundaries are permeable.
But the thing is do we conceptually and in practice
treat it that way? Or do we treat it as if it's a closed
system and we can operate in our little part and ignore what's
going on outside. And it's easy to operate in a system and
treat it as a closed system, but you do so at great peril.
And as us doing interventions, we do so at
the risk of our interventions failing.
>>>DR. RUSSELL GLASGOW: Thank you. And that tied into
the second point here about the complexity of the future and
our situation, ignoring that at our peril. Brian.
>>>DR. BRIAN MITTMAN: My twenty second response to the first is
the journal will do what it always does, which is welcome
and encourage submissions on the topic. The second thing I think
is something we don't do often enough. And that is an
editorial or commentary or other more proactive effort to
identify issues and proactively seek out submissions.
And that's something that we'll talk about.
>>>DR. RUSSELL GLASGOW: I want to thank all of the
panelists here. I want to close with this one slide here
that Einstein anticipated this conference fifty some
years ago. As you can see there that we can't keep thinking
at the same level that created the problem for us in the
first place. But with that, we are going to close with the
words and the notion that Steve is going to tell us and
explain, in the words of Brian Weiner, he is going to
explain at the bottom of the last slide the infamous
"and the one and two." That was the key challenge for
the conference. So we're going to close this session
with Steve telling us about the "and."
>>>DR. STEVE CLAUSER: Well, I don't know about the "and."
But that clearly talks about connectivity. And I think that
one thing that for me that's important is just the moment I
think we have right now in history to really have to...
to almost be compelled to address these issues in
some way. Because, I mean, Kelly talked about health reform and
Muin talked about the tsunami of genomic medicine. But there's a
lot of other things going on in the cancer care delivery system
and the community health system that I think are major changes.
That no matter whether you want to define reminder systems that
affect both patients and providers, you know, as one or
two levels, clearly, they're not going to be sufficient to
address the kinds of pressure that are occurring given the
expectations now with us reaching out to the community
to try to get community engagement, not only in basic
research with bio specimen collection, with clinical trial
participation. Oh, and by the way, providing patient-centered
medical care. Those things are all happening now in kind of a
perfect storm. And I think that the only way I can imagine
having an opportunity to deal with this is if we really take
a very systems point of view in terms of how these
different factors interact with one another.
And if it's not multi-level intervention research,
tell me another game in town.
>>>DR. RUSSELL GLASGOW: Thank you so much.
And with that, that is the end of this session.