[This Transcript is Unedited]

DEPARTMENT OF HEALTH AND HUMAN SERVICES

NATIONAL COMMITTEE ON VITAL AND HEALTH STATISTICS

SUBCOMMITTEE ON POPULATIONS

Meeting on:

DATA NEEDS FOR ASSESSING SURGE CAPACITY AND HOSPITAL PREPAREDNESS

February 14, 2007

Hubert Humphrey Building
200 Independence Avenue
Washington, D.C.

Proceedings by:
CASET Associates
10201 Lee Highway, Suite 180
Fairfax, Virginia 22030
(703) 352-0091

TABLE OF CONTENTS


PARTICIPANTS:

  • DONALD M. STEINWACHS, PhD, Chairman.
  • LARRY GREEN, MD
  • GARLAND LAND, MPH
  • WILLIAM J. SCANLON, PhD
  • C. EUGENE STEUERLE, PhD
  • KEVIN C. VIGILANTE, MD, MPH

Staff

  • DALE HITCHCOCK, ASPE, Lead Staff
  • AUDREY BURWELL, OMH
  • NANCY BREEN, PhD, NCI, NIH
  • LESLIE COOPER, PhD, NCI, NIH
  • MIRYAM GRANTHON, OPHE
  • SUZANNE HAYNES, PhD, OPHS
  • CILLE KENNEDY, PhD, ASPE
  • JACQUELINE LUCAS, NCHS
  • EDNA PAISANO, IHS
  • SUSAN QUEEN, PhD, HRSA
  • HARVEY SCHWARTZ, PhD, AHRQ

P R O C E E D I N G S (1:43 p.m.)

Agenda Item: Call to Order. Welcome and Charge to the Participants.

DR. VIGILANTE: My name is Kevin Vigilante. We are going to go around and
introduce ourselves before we get going.

I do want to thank everyone for braving the wintry mix. It sounds like a
medley of tunes, but we appreciate your coming.

I think we will have a very interesting program today. We have some great
speakers here and discussants. I couldn’t restrain myself.

I felt rather guilty that we can’t provide you lunch, but we did get from
the cafeteria some odds and ends, here on the corner of the table, mixed nuts,
apples and sort of snack bars, if you are about to get hyperglycemic or
anything.

Why don’t we start, by way of introductions, with Jim and go around the
table and then we will talk about what our agenda is and a little bit of
background.

[Introductions made around table, around perimeter of room, and on
speakerphone.]

So, I would like to sort of give some backdrop to this, and I think Bill
and Jim will also add some color commentary.

I think for context it is important to understand what the NCVHS is. The
NCHVS, the National Committee on Vital and health Statistics, is a statutory
public advisory body that advises the Secretary on health data, statistics, and
national health information policy.

Since our mission is really focused on data and information, we are coming
at this problem from that perspective.

We have, in the wake of Katrina and other recent events, really thought it
would be very important to focus on data needs relative to preparedness and
response.

Now, of course, that is a very, very big topic. It is a very, very big
area. I think today’s session is really a first step in an exploratory process
for us to understand better what we ought to be doing in this environment as a
committee, to make recommendations to the Secretary.

I think at one level this is a bit of a brain storming process. So, we
don’t want to be constrained too much on sort of driving hypotheses about what
we ought to be doing and focusing on.

On the other hand, we are constrained by our mission, which is around
information and data. A hypothesis going into this, at one level, is that there
is probably data that we are not collecting that we should be, data that we are
collecting that we perhaps shouldn’t be, different parties collecting data, the
same data, but asking for it in different ways.

Thirdly, there is probably data that a number of agencies and entities are
collecting that should be sharing and it would be more efficient to do that
way.

I think that is sort of the going in sort of perspective. We may come out
with a different perspective. I think, after hearing from folks today — there
are a number of folks with presentations — we really don’t want this to be
sort of a didactic session.

What we really want is a robust exchange of ideas and discussion. So, don’t
feel inhibited in any way. So, that would be my perspective on the data. I
would ask Bill and Jim to add their perspective as well.

DR. W. SCANLON: I would just underscore that we are a committee that is
focused on data and sort of information.

While we may sort of have collectively in this room many opinions about
what the state of surge capacity sort of is in this country, it is not our role
to try and recommend policies that are going to address that. It is really to
focus sort of on the information side of it.

Now, for me personally, when I wear another hat, which is as a member of
the medicare payment advisory commission, there is an intersection here of
interest.

One of the things that is sort of a reality is that surge capacity may be a
very important service that hospitals provide to this country, but there is a
question of, if we are going to sort of recognize it and pay for it, how can we
sort of do that in a reasonable and reliable way.

The need to do that was brought home to me when I actually used to be at
GAO. We did some work on emergency room capacity.

One of the sort of sets of comments that we got from some hospital
executives is that they had eliminated a lot of their essentially surge
capacity.

They had sort of taken to heart the message they were getting from payers
and sort of from the American people, which is that they needed to be more
efficient, and they had right sized.

So, if, in the process of right sizing, we are throwing sort of out the
essential services that we may need at a point in time, we are doing the wrong
thing.

So, we need, in terms of setting sort of the direction through the payment
policies that we have sort of in our public programs, we need to recognize sort
of what are the multiple types of services that we want, including that stand
by capacity that represents surge, and make sure that it is there when we need
it, make sure that we are paying sort of sufficiently for it, and that we are
getting it when we pay for it.

So,I bring that kind of perspective to this as well. Again, we are here
today more focused on the issue of what is it that we should be measuring in
surge, how well are we doing it today, how can we do it better for the future.

MR. J. SCANLON: I will just add to that, and then I will be setting the
stage for this particular focus as well, but I just might, in terms of
perspective, the National Committee on Vital and Health Statistics actually has
a long and distinguished history of advisory HHS and the health care and public
health community on data issues, data needs, and so on, really going back a
long time.

Many of the initiatives and many of the data resources that we now take for
granted actually arose out of recommendations from this committee.

I am here at HHS and I am also on the receiving end of many of the
recommendations as well. So, at any rate, we asked the committee, the
department did, to kind of look at this issue at least in an exploratory way,
and I will give you a little bit more history as we get started.

It arose out of experiences more recently, over the past five years, six
years, seven years particularly, and a desire to get a better grasp on what the
actual standards and best practices and data and resources and approaches are
in this area.

There are broader public health preparedness issue and measures that we
will be interested in later, but we are focusing today on, in essence, the
health care preparedness, I guess, and surge capacity is one dimension.

DR. VIGILANTE: I think we are going to kick off with Rick Niska and
Katherine Burt sort of doing a presentation about —

MR. J. SCANLON: If I could give just a little overview?

DR. VIGILANTE: I am sorry, yes.

Agenda Item: Overview of the Data Challenge.

MR. J. SCANLON: Let me again, just to focus, I am Jim Scanlon and I am here
with HHS. My office is sort of the corporate departmental office of planning
and evaluation.

So, we don’t have operational programs to run. We are supposed to be the
think tank, the analytic office, the evaluation office, and the data mining
office. So, we have initiated a couple of activities and tried to develop data
in this area.

In terms of history, I probably don’t need to remind this group but,
beginning with 9-11, beginning with anthrax scares, following Katrina and Rita
response efforts almost a year and a half ago now, and then with an emphasis on
preparedness for all threats and all hazards, and now more recently with the
focus on pandemic preparedness, a lot of different strands have come together
now.

They have clearly forced us to take a look at the whole national response
and preparedness apparatus in the United States, what the play book is, what
the responsibilities are, and what the dimensions are of preparedness, and what
the responsibilities are.

I think that Katrina particularly, the federal response, the local and
state response to Hurricane Katrina and, later, Rita, I think really pointed
out some of the — again, you could argue that those were unique situations and
maybe don’t make a good model to plan upon, but they clearly showed us some
weaknesses and limitations in a lot of things, from the federal level to the
local level, for the protocol and the play book and responsibilities, and for
some of the basic assumptions about how the whole national response plan is set
up.

The national response plan — you know better than I — is kind of the
national play book and model for how emergency incidents are to be handled.
There are various scenarios for various threats.

As a result of Katrina and Rita, that really had to be looked at again. I
think the basic framework was considered sound, but there needed to be a lot
more fleshing out and clarity.

One of the fundamental assumptions, though, of the national response plan,
the various scenarios and the incident command situations, was that there was,
in essence, the community level would be handling first response, would be
handling most emergencies.

Then, if the event proceeded to the scale and the complexity and severity
that additional resources were required, then the community leaders would call
upon the states.

If those resources were overwhelmed as well, then the call would be upon
the federal level, but it had to be initiated, the request had to be somehow
initiated at one level, to the top.

Then the federal government, where we are, has a number of resources that
can often be augmentary, sometimes primary.

In terms of augmenting, maybe the health care resources, the public health
and the medical support, and the human services dimension as well.

In Katrina and others, HHS has the major responsibility for the whole
federal interagency response. It is called the medical and public health
response, and there is a whole set of array of resources and so on that can be
provided.

Again, much of this framework rests upon this local government first
response, community first response, not just the local government, then the
state, then the federal level — well, regional levels, and interstate contacts
and so on, and then the federal level.

I think what we all found was that a lot of assumptions were made about
readiness and preparedness and resources and capabilities at all of those
levels.

I think the experience with Rita showed that probably some of those, at
least, were not well founded assumptions.

One of the critical areas, and certainly not the only area, was in the area
of health care, health care organizational and hospital preparedness at the
community level.

Again, I think there were some marvelous, very exceptional examples. At the
same time, I don’t think anyone thought we had a systematic assessment before
the fact of how really — well, this is what we are assuming is readiness.

We didn’t really have a systematic sense of where everyone really was. The
reason for assessing systematically is really to keep improving, find out where
the gaps are, where attention and resources need to be put.

So, we in our office, when we did lessons learned at the federal level at
any rate on Katrina and the federal response, I think everyone thought we
needed to get a better sense of, can we assess, can we measure, can we
periodically take stock on where we are with the whole array of preparedness
issues.

Some of them are states, some of them are directed at the state level and
the preparedness level, but some were clearly at the community level, and some
of those interests were clearly at the level of the hospital or the community
health care surge capacity or preparedness or readiness in general.

So, we began in our office to begin looking at what did we know, what did
people say were the best practices, what did folks say were the requirements
for hospital preparedness, again, for all sorts of hazards. I guess pandemic
introduces a whole new level of consideration of what constitutes preparedness.

So, we in our office, we started with a literature review in terms of what
did the literature say, what did the specialty societies say, and so on, about
what constitutes hospital preparedness and what do we know about it. Then we
proceeded to some further work.

GAO, I think as Bill said and as Cindy said, has undertaken a study to look
at performance measurement and so on in the preparedness areas as well. We
shared our papers with them.

We are continuing to look at, in our office we sponsored some questions
added to the survey you will be hearing from Kathy and Rick in a minute. It is
their survey of emergency departments, and we supported some questions a couple
of years ago now, I think it was, on emergency preparedness in hospitals. We
are probably looking to do an update or at least expand that as well.

So, there are a lot of activities in these areas. There are a lot of ways
of sort of testing preparedness. I mean, some folks like tabletops, some folks
like exercises, some folks like the certification route, and then there is the
survey route.

We have had other activities in HHS. Within our hospital preparedness
program — I think we have those folks on the line as well — these are
cooperative grants to states for hospital preparedness, and they hava collected
some data as well.

I think we are kind of looking at how can we pull this all together in a
systematic meaningful well without over-burdening folks, to kind of get a good
sense of, are there measures we can use, are there standards we should look at.

Are there approaches, perhaps certification process or others, where there
is data available, as Kevin said, that the department and all of you can rely
on and pull together to kind of give us, when we take stock on this where we
are periodically, what kind of data could we use.

That was what prompted, I think, our request to the committee. The
committee, as usual, put together a very capable group of experts, and we are
looking forward to the discussion today.

DR. VIGILANTE: Thanks, Jim. Rick and Kathy, do you want to dive in?

Agenda Item: Panel One – Federal Perspective. Ongoing
Federal Surveys and Available Data.

DR. BURT: Thanks. Both Rick and I work on the National Hospital Ambulatory
Medical Care Survey, which is a survey, an annual representative survey of
non-federal general and short-stay hospitals.

It excludes military, federal, VA types of hospitals. We have a sample of
about 700 hospitals that we are going to be talking about as the data base
today.

Primarily we are going to be talking about two facility supplements that
were added to these hospital surveys, which is primarily a survey of visits to
emergency departments and outpatient departments.

We added extra facility questions on staffing capacity and ambulance
diversion, and on bioterrorism and mass casualty preparedness, the latter being
funded by ASPE.

These were done in 2003-2004. The data were collected over two years. From
that sample we also have the visit data for over 76,000 encounters.

The response rate for that survey was 85 percent. There are adjustments on
the weighting process to account for non-response and to weight up to national
annual estimates.

So, our frame, which is the Verispan hospital market base, has about 5,100
total non-federal hospitals. That is short stay hospitals and general
hospitals, including childrens general hospitals.

When we surveyed them, we found that about 92 percent of them had emergency
departments, and about four percent had no emergency department but did have
outpatient services, and another four percent had no outpatient or emergency
room. So, we will be talking primarily about the 92 percent of hospitals that
do have an emergency department.

This is the distribution of these hospitals by the four major census
regions. About 68 percent are in the south and the midwest.

The civilian population has very similar percentages of people in those
locales. It is a different story when it comes to MSA areas. Thirty-eight
percent of hospitals are in non-metro areas, but only 16 percent of the
population is in those areas.

So, what happens is that the metro hospitals tend to be bigger hospitals.
There are a lot more beds, a lot more emergency department stuff.

Unfortunately, the demand on emergency departments has been increasing over
the last 10 years, whereas the number of emergency departments that were
operating actually declined, which means that any one emergency department,
especially in metro areas, was seeing more and more cases.

The IOM, in their reports, the three volume reports that came out last
June, clearly indicate all the problems with crowding in emergency departments.

In metro areas, about 64 percent of the EDs experienced some form of
crowding during the year. Now, we defined crowding as either having an
ambulance diversion period, where you send ambulances away to other hospitals,
they don’t let them come, or greater than three percent of the ED patients left
before being seen by a physician or other health care provider, or that the
average waiting time for urgent cases was greater than 60 minutes.

Larger hospitals are more likely to experience crowding. Therefore, even
though 64 percent of metro hospitals experienced crowding, that results in 84
percent of the visits in those areas going to hospitals that experience
crowding.

So, one of the problems with crowding is this problem of ambulance
diversions. This shows that, of the metro hospitals, only about a third don’t
do any ambulance diversions.

That doesn’t mean that they don’t want to do ambulance diversions. There is
a fair percent that are not allowed to divert any ambulances away.

You can see there is a small set of metro EDs that have quite a bit of
their time, up to 20 percent of their time, on diversion.

The metro areas in 2003-2004 had about 400 hours each year in diversion
status, which resulted in about 500,000 ambulances being diverted away from the
closest emergency room, and that results in about one per minute.

The leading reasons were the lack of inpatient beds and the number of ED
patients. So, these are equally responsible for most of the diversions that
occur.

So, it is not just an ED problem. It is a hospital problem, and this is
well documented. This particular chart — by the way, we have these two reports
over here, if you would like to see them, some of these charts are coming out
of these reports.

This particular chart shows that, among the EDs in metro areas that had as
much as 20 percent of their time on diversion, that they also tended, on
average, to be the large volume hospitals as it relates to bed size.

Their mean was about 311 beds, and their occupancy rate was, on average, 81
percent. Now someone mentioned about cutting — about the expense being more
efficient, so some hospitals got rid of their excess capacity. They do this by
closing down beds and having higher occupancy rates than they maybe used to
have, which leads to some waste.

So, hospitals that had no ambulance diversions, down in the bottom
left-hand corner, on average their bed side was small, their occupancy rate is
only about 60 percent.

So, what kind of measures do we have? These reports go into all kinds of
details about what kinds of services the emergency rooms provide, their
staffing, their space.

One of the questions we asked was about the number of treatment spaces. So,
when we weight that up to a national total, there are 65,700 ED treatment
spaces nationally. This, of course, was 2003-2004, which was about 14.6 per ED,
but we know there is great variation across EDs.

Population-wise, it turns out to be 2.3 per 10,000 persons. Most ED visits
lasts about — the treatment lasts about two hours in general. If it is serious
and people are being admitted, it is going to last longer than that, about
three hours, and that doesn’t include the waiting time outside.

We asked about whether they recently expanded the space, and 16 percent of
EDs said that they had, within the last two years, expanded their number of
treatment spaces, and 32 percent said they intended to expand them within the
next two years.

So, this is 2007. In theory, those 32 percent have already expanded their
treatment spaces. So, hopefully these numbers are higher now than they were
back then.

The midwest has a higher spaces per capita. I went by it pretty quickly,
but we have in general more hospitals relative to the population in the midwest
than we do in other regions, and they also, therefore, have more ED capability.

In terms of inpatient beds per population, it weights up to 614,000 staffed
inpatient beds nationally, which is about 137 per hospital with an ED, and that
corresponds to about 21.4 per 10,000 persons in the population.

Again, the midwest has a higher per capita rate compared to the south and
the west. So, getting these numbers and sort of weighting them up to national
totals, I am not sure if that is a useful metric or not, but we thought it
could be used for modeling how many you may need or the impact of expected
different scenarios.

Like in a pandemic, how many is it likely to affect and then you can see
what the rate per population is, and you will know how bad off you are going to
be in terms of providing hospital care for them. Rick is going to go ahead and
talk a little bit more about our bioterrorism supplement.

DR. NISKA: The data that I am going to show you are from the supplements
that had to do with the national hospital ambulatory medical care survey, the
bioterrorism and mass casualty supplement.

I pulled out some of the stuff that had to do with surge capacity
specifically, and there were several questions we asked.

These questions were basically asking hospitals to describe their mass
casualty emergency response plan, were these issues dealt with in that plan.

We found that 73 percent of hospitals, actually a pretty good number, had
some sort of policy on cancellation of elective procedures and admissions, so
that you could deal with whatever was coming in requiring emergency treatment
or possibly surgery in the case of an explosive or type or other trauma
casualties.

Sixty-five percent had the establishment of an alternate care site, in
other words, something outside of the usual emergency department. It might be a
tent set up on the hospital grounds, it might be other areas of the hospital.

About 16 percent had back up plans in writing for medical utilization of
non-clinical space. To translate that into English, that might be lining the
hallways or using other areas of the hospital for dealing with patients who
were not seriously injured but, nevertheless, needed to be evaluated and dealt
with in some way. So, utilizing that non-clinical space.

Only half of the hospitals had actual memoranda of understanding, written
agreements, with outlying hospitals to accept inpatients during a declared
disaster.

The interesting thing, and what I didn’t show you on this graph, was that
many of the hospitals, something in the 80 percent range, had actual plans that
dealt with people coming in that needed to be dealt with, but there was a big
disparity between the plans and the actual memorandum of understanding, meaning
that we could actually send patients to these hospitals. So, only half could do
that.

About 45 percent had plans for stockpiling antibiotics and supplies. Now,
how you view this, some people may say this is great that so many hospitals
have these in place, especially the stockpiles, rather than depending on the
CDC stockpile.

You could look at it a little bit more pessimistically by saying, we really
need to ramp this up, that more hospitals could be doing more. I will leave
that to the higher thinkers than I.

These last two issues, I actually stratified these on several variables,
but urban rural issues corresponding to the MSA/non-MSA group seems to come up
in discussions quite a bit, how many resources should we divert to urban areas,
and should we leave the rural areas uncovered or should we deal with them
aggressively as well.

For these, such techniques as converting the post-anesthesia care unit to
be an intensive care unit essentially, you can see that about 40 percent of
hospitals overall, but there is a big urban rural difference there. It was more
likely to be done in urban hospitals, and about half or a quarter of rural
hospitals, and that was fairly significant.

The other thing was activating unused space. This isn’t just lining people
in the halls. This is actually opening up decommissioned board space that is
available to be used if you had the staffing and you had the equipment. In
other words, actually increasing the capacity of the hospital space that was
originally designed, but is not now, being used for clinical care. Only about a
little less than a third had plans for doing that and, again, the urban areas
were much more likely to do that then the rural areas.

I looked at the question that we had about the national disaster medical
system. Basically, the responders were hospital administrators or they were the
bioterrorism coordinators or somebody who was in charge in that hospital of
coming up with emergency plans.

We asked the hospitals if they were hooked into the national disaster
medical system, in other words, were they designed to receive patients during
the NDMS.

If you remember the previous slide about the MOUs, where you refer people
out and hopefully somebody at a higher level of care will take them.

What was interesting to me was not so much that only about a third were
designated to receive patients through NDMS, again, with a bit of an urban
rural split there, but the third column, which is how many hospitals actually
didn’t know whether they were designed or not.

Now, I asked people about that, and some of the ideas that we didn’t
actually ask about in the survey but might serve as trying to figure out why
this occurs is that some of these hospitals may have been designed at NDMS
hospitals and then, with mergers and selling of hospitals and being bought up,
it just kind of went by the wayside.

Maybe the hospital had never been utilized as an NDMS hospital and the
people just kind of forgot that they were designated. That is a possibility.
That is speculation as far as this survey goes, but I thought it was really
interesting, that a third of the hospitals and a little less than half of the
rural hospitals did not know their NDMS status, whether or not they were
designated or not.

This is actually — this has to do with availability of resources, critical
care beds, ventilators, negative pressure, isolation rooms, personal protective
suits, and decontamination showers.

Some of the questions that I have gotten that I haven’t published yet but
people are always interested in national estimates about how many we have in
total, so I will show you a slide having to do with that.

Again, a big urban rural split between just the sheer number of critical
care beds per hospital, about 21 on the average, about 29 critical care beds in
urban hospitals and only six in rural hospitals.

The same trend goes for — actually, it is not a trend, these are
significant differences at the .01 and less level — mechanical ventilators,
again, about 12 mechanical ventilators per hospital.

Negative pressure isolation rooms — this actually surprised me because
maybe I have dealt with a lot of rural hospitals in my time and have seen fewer
negative pressure isolation rooms, but there are actually seven per hospital on
the average, in rural areas about three.

I couldn’t fail to note, the last time I was doing temporary duty at the
Indian Health Service hospital I go to, which is an urban hospital in the
southwest, that there was actually a negative pressure isolation room in the
actual emergency department. So, I was pleased to see that.

Personal protection, just having the PPS available, now we didn’t ask what
type of suits, whether it was a full PAPRS with the self contained respirators
and all of that, or just the suit that you can throw on to protect yourself
about chemical splashes. We just asked about personal protective suits, and
this is what we found. About 14 suits per hospital were available for such
things as chemical casualties coming in.

Decontamination showers, I don’t think this is statistically significant,
but there are one or two showers per hospital.

Now, what we didn’t ask, and it might be interesting if we do this again,
is to see how many were multiplace decontamination showers versus a single room
with a single shower head, and also if people or hospitals have plans for
actually setting these up.

I remember talking to some folks from an area hospital here in D.C., that
they actually had a plan to set these up in the alley-ways before people get
into the emergency rooms so they can do the chemical decon before they come in.
So, maybe we need to ask about creative ideas like that if we do this again.

Okay, I split the data a little bit differently to look at nationwide
totals. If you want to know how many critical care beds we have in the nation,
there are about 97,500, and so forth, for mechanical ventilators, negative
pressure isolation rooms, et cetera.

I would like to emphasize that we do a nationally representative survey
with a complex sample design that allows us to make national estimates. So,
these would be fairly good for planning, as Dr. Burt mentioned, in terms of
modeling disaster scenarios, how many nationwide beds we have, say, for a
pandemic flu situation that would affect the entire nation.

Then we split it. Kathy, actually she did the work on this, to split this
out per 10,000 population, the critical care beds, and we only have about three
critical care beds per 10,000 people, which is fine usually but maybe not in a
pandemic situation, or mass casualty, about two mechanical ventilators per
10,000 population, only one negative pressure isolation room per 10,000
population, not too many personal protective suit. Well, I guess that has to do
more with the hospital staff, but again, only about actually one
decontamination shower per 30,000 on the average.

So, some conclusions from the work. We observed that crowding in the EDs at
present limits our ability to handle an influx of cases.

The IOM report was excellent in pointing this out, and it is really
something that has been bandied about in the emergency medicine literature and
practical practice for years, the fact that EDs are crowded and we wonder what
would happen, if they are already crowded, what are we going to do during mass
casualty situations or mass epidemics.

There is an observed disparity across the regions in their ability to
handle a surge due to natural or manmade disasters.

We saw that the midwest and the northeast tend to do a little bit better. I
think Kathy may have emphasized the midwest because that may have been the
statistically significant part.

Surge capacity planning for specific elements is found in only a quarter to
three quarters of the hospitals depending on what you are looking at. So, that
can certainly be done better.

We saw an observed disparity favoring urban hospitals in both emergency
planning, having things actually written down in the plan, and available
resources, things to actually manage these emergencies.

I think that was our last slide. So, I will just leave that up. I don’t
know if we are doing questions now or later.

MR. J. SCANLON: One question. You show the average number, the mean number.
Do you have a distribution? It would be good to know the percentage of
hospitals that have none or one or two on the scale. Do you remember that data?

DR. NISKA: We haven’t analyzed it that way, but there is no reason why we
couldn’t. It is a continuous variable. We could put them into bands and see how
that goes.

DR. BOENNING: Could one conclude that hospitals currently do not have surge
capacity?

DR. NISKA: That sounds like some of the questions you get from reporters
sometimes, are we prepared or are we not, and we wonder what the implications
of that are going to be.

Again, it depends on how you look at it and what you are preparing for. You
know, I guess it is kind of good that almost three quarters of hospitals can
cancel their elective cases and so forth, but if you are more of a pessimist
you should say, well, hospitals should be able to do that routinely because
they always bump cases for traumas come in and why are only 73 percent.

I think we could do better. This is 2003-2004 data. We have a few cases in
2005 which we haven’t looked at yet, which we can certainly do as we crank out
these papers and everything.

I think we could do a better job and hopefully this data will help us plan
on a national level. I used to be with the hospital bioterrorism preparedness
program which provides grants or cooperative agreements to states and
territories to try to manage this stuff. Hopefully data like this can help plan
this in a more focused fashion.

I don’t know if we are prepared or not. We are somewhat better prepared,
but we could do better, would be the short answer.

DR. KELLERMAN: Two related questions. One is that a lot of this information
is self reported survey data. Do you have any independent validation that some
of these numbers that are reported are, in fact, real numbers, given that there
is a pretty strong social desirability or bias to report issues.

Second, you mentioned that the survey data was sort of 2003-2004, and we
are well into 2007 now. Do you have any trend data to suggest that we are doing
better or worse, for example, with things like ambulance diversion, and that
statistic, which the media really picked up and reported pretty regularly, of
half million ambulances a year. Is that getting worse or is that getting
better?

DR. BURT: Well, I will answer your second question first, which I think you
already know the answer to, because I told you.

We did ask, in 2005, we asked hospitals to tell us about their ambulance
diversions. We started asking in 2003 and, since 2003, our percent missing the
information has been increasing to a point such that we cannot publish anything
from 2005 on ambulance diversions because 50 percent of the hospitals didn’t
tell us the answer to that question. In 2002-2003 they did.

DR. KELLERMAN: I set her up for that. She didn’t know I was going to do
that, but it is concerning to me that, both from the perspective of social
desirability bias and self reporting, that if hospitals aren’t sharing that
data on something that fundamentally important to readiness and preparedness, I
think that has implications for the rest of our discussion over the course of
the afternoon about what we are going to do and what we are going to measure.

I can’t interpret it because I wasn’t there and none of us were, when
individual institutions answered. Either they are not tracking the stat, which
would worry me, or they don’t want us to know the stat, which worries me even
more.

DR. BENTLEY: Or, Arthur, this building and the people in it have proposed
cutting medicare payments by billions of dollars, proposed cutting medicaid
payments by billions of dollars, proposed cutting hospital bioterrorism
preparedness by millions of dollars, and you ask yourself why it is that
hospitals don’t want to answer surveys, whether it is from HHS or whether it is
from us or whether it is from the Emergency Medical Physicians Association.

Two thirds of hospitals don’t make money on patient care. One third of
hospitals are losing money in total, all money.

You take that data. You say, every dollar that gets spent answering a
question, and now there is a new plan to go out and require the plan to daily
report its supply of blood on hand as opposed to relying on the AABB and the
Red Cross.

We are simply getting back from our members that they don’t have the
resources to respond to the increasing and ever present demands of people who
want data on this, that and the other thing and be able to care for patients in
their community. You have a choice. Every dollar that goes in to data
collection doesn’t go into caring for patients.

DR. VIGILANTE: I think that is part of, frankly, the mission of this
hearing, is really to understand as well the burden of reporting, because it
does affect the quality of the data that we get. I think it is something that
we are going to be purposefully exploring.

DR. BENTLEY: One last comment? There was a survey done, and maybe Melissa
can tell us, because I think she is on the phone, that I think it was your firm
designed. Booz-Allen Hamilton designed it, went out to hospitals collecting
data, came back in, and our best understanding — somebody can correct me if I
am wrong — is OMB is still sitting on that data and has not released it.

If you want to collect data, you want to motivate physicians, hospitals,
anybody to submit those data, you can’t collect it and never release it and
never do anything with it.

DR. AUF DER HEIDE: I think a corollary to that is that you shouldn’t be
collecting data unless you can specify what is going to be done with it.

You know, we have had a history in recent times at the federal level of
making all kinds of reports and collecting all kinds of data and nothing ever
results from it. I think that is an important aspect of this whole scenario.

DR. SANDERS: Jim, you are right, and Sally Phillips actually has more
background on the reason why that data — it has to be reported back in a
certain way.

DR. PHILLIPS: You are right, that survey was fielded now I believe almost
two years ago and we are still trying to figure out how to get that data back
to the state level so that they can filter that down to their hospitals. I
think what has been said here has been adequate, unless you want me to go into
more detail. I can.

DR. SHULER: I had hoped to be here with Melissa Sanders, but it turns out I
am here instead of her and she is on the phone.

I don’t have the answers to all the issues that have been raised, but there
are a couple of comments that I would like to make.

For starters, we have a lot of work to do on our data, but I think we have
got some data that are more recent than what you have seen here, and which
would present a slightly different picture.

There has been a lot of learning and growing in this program over the last
few years. Some of the data that I have seen from 2003-2004 reflects, I think,
at least the data that we have an inability on the part of the states and
hospitals to answer the questions, and I have seen that get a lot better over
the years. I think we are asking the questions better now and getting better
answers.

Secondly, some of these data were collected several years ago before there
was a large increase in the amount of funding for the national hospital
bioterrorism program.

I do know for sure that we could present more recent data on, for example,
decontamination expenditures, personal protective equipment, and things like
that, which there has been a lot of investment in those things over the last
few years.

As I said, we have got a lot of work to do on our data. As we have been
growing and changing over the last few years, we have changed the questions
that we have asked a lot.

Consequently, we don’t have the ability to show you year to year change on
quite a few things, because some of these questions haven’t even been asked for
two consecutive years.

We are moving in that direction. We are creating a system. We are planning
to do a lot more analysis of data and make the data available as appropriate
for decision making, and we collect it twice a year.

DR. GREEN: A question for Katherine, Richard and possibly Art Kellerman.
What is your opinion about the value of a distance to care measure? Secondly,
do you have any data about distance to emergency care?

DR. BURT: Well, we have the capability of doing distance to emergency care
but, based on the zip code of where the patient resides and the zip code of the
emergency department. Of course, not everybody comes from home. So, we haven’t
done a lot with it.

DR. GREEN: Is it your opinion that it really doesn’t pay off? It is not of
great value?

DR. BURT: The stuff in our data set, the only distance measure we have
would be from median zip code to sort of median zip code, and it would only be
from the residence.

We do have a statistic somewhere about where they came from. No, no, I
don’t think we do. So, we don’t know where they were when they decided they had
to come to the emergency room.

In that regard, I am not sure that what we have is useful. We would have to
ask a different question a different way which probably is not in the medical
record and, for our survey, if it is not in the medical record, we don’t get
it.

DR. HANFLING: I want to thank you for an excellent presentation but I want
to go back to Art Kellerman’s point. I think, relevant to the discussion for
the rest of the afternoon, we should be able to distinguish between prospective
data collection ala ambulance diversion and what I call situational awareness
or the ability to create real time or near real time situational awareness,
which is going to be very critical for figuring out some of the issues around
surge capacity limitations that your retrospective data has demonstrated.

I think it is important to distinguish the two. Then the question is, in
the data tool that you use, was there any assessment at all in terms of either
on a regional or a state level what sort of real time situational awareness
capabilities were in place at that time?

DR. BURT: We did not ask those questions. The ambulance diversion data in
2002-2003, part of that supplement, it wasn’t just a single question.

We had them track and write down prospectively every single diversion
episode, which is a costly thing for a hospital to do. We don’t pay them
anything to answer our questions.

It gave us better data than just asking a single question, but we didn’t
ask about real time, do you have the ability to tell me immediately how many
beds are vacant.

Now, in our 2007 survey we are asking some questions on that order, about
how often do they do their census, is it only once a day, is it twice a day, is
it real time, that kind of thing.

DR. HANFLING: I would suggest — I would be interested in Jim’s sense of
this — but the ability to conduct that sort of real time surveillance really
may be protective to health care facilities ultimately in terms of being able
to forecast what may be coming.

In the context of trying to — not trying to sort of create mandates and
throw a whole lot of money down a rabbit hole, this may still be an area — and
I think hopefully at least in the context of when I give my presentation — try
to hit on why this is a worthwhile investment. The question is, whose
investment.

DR. BOENNING: I hope we get into it more in the second panel and we also
have representatives here from the Maryland state system, who have a state
system named FRED, I believe, that keeps track of daily bed availability.

DR. NISKA: What would be interesting in terms of future data collection is
to tie it to the requirements of both the CDC and the HRSA guidances on
bioterrorism.

One might be that real time surveillance, how many emergency departments
are actually hooked up directly to the health department where they can feed
data directly, so that you get more of a real time surveillance versus the
usual surveillance which is retrospective monthly reporting sort of thing.

DR. GAMACHE: I just had a couple of quick comments from the data. Before I
went back into informatics, I ran the public health department’s bioterrorism
preparedness program.

On the stockpile issue, at the time, communities were asked not to do that,
because really it was a capacity issue with the antibiotics and getting enough
there for the S and S, never mind for individual stockpiles.

That was around the time when it just started to change. So, having 45
percent, I think, at that time frame is really quite good and I think it is a
lot better now.

I think in the future also some of these hospitals realize they can’t
handle that inventory management part of it. So, they are actually getting
collaboration with other facilities.

So, the hospital itself may not have it. The community may have it or it
may be part of a partnership. I don’t know how you design a question that is
going to look at the community perspective on that.

Also, on the PPE, the same thing has been done where a lot of that is now
done by the state homeland security or the state emergency management, or they
manage the PPE or the decon part of it beforehand, and the hospitals are
collaborating with that.

So, the hospitals themselves may not have it, but it may be part of the
community plan. I think it makes doing the surveys in the future much more
complicated, because not everyone has done this the same way, but they have
tried to do it more so that they are combining and coordinating their resources
much more effectively than when they first started this program.

I think it is quite a hard thing to do these surveys. I am not trying to
get at that, but I think in the future it is going to be hard to figure out how
to do these surveys just focusing on one area.

DR. NISKA: When I was first writing up these data, just because of my
institutional memory of how fast the funding got out for hospital preparedness,
et cetera, I made the point in at least some of these papers that this really
represents baseline data, because even though we got the funding rapidly out to
states in 2002-2003, that didn’t necessarily translate to states getting it out
to individual hospitals.

So, I made the point that the 2003-2004 — actually, in essence, it is not
a precise cut or anything, but it in essence represents what our baseline
status is before all of this federal intervention came in.

So, it would be fascinating to see what the follow up data that HRSA has, I
would love to see that study see the light of day, partly to address Dr.
Kellerman’s thing about response bias or social good bias — I forget the term
you used — but that people are responding one way to the grant making people
and another way to NCHS, which really doesn’t provide them their grants.

That would be interesting in terms of a comparison, and also just seeing
what has happened since the money did eventually make it to hospitals and
people have been working on their programs.

MR. J. SCANLON: Number one, I am not sure which side would be the more
correct. I think that is why we are asking you to think about what are the best
measures and approaches.

I did want to ask, are there standards at hospitals to maintain
accreditation or become designated in certain ways, that have already been
agreed to, best practices or standards that constitute preparedness, and is
that the basis that you based the questions on, or was it expert opinion?

DR. BURT: It was all of that, but yes, JAHCO has specific requirements for
what the hospital has to have in their emergency response plan.

So, our questions along those lines were based totally on the JAHCO
requirements. The things that weren’t part of the requirements were based on
expert opinion from the Secretary’s office and HRSA and other places all over.

MR. J. SCANLON: But isn’t that, in the spirit of reducing the burden, if
there is administrative or accreditational data available, not so much on
individual facilities, but at least in a statistical sense, isn’t that another
way.

If JAHCO or others have statistics on the number of hospitals that have
passed or are in conformance with these standards, even on a statistical
anonymized basis, that would be useful to know as well. I just don’t know.

DR. BURT: I don’t know the extent of their accreditation, that they have to
physically go and check everything. Are they actually reading the response
plans.

To answer our first question, we didn’t ask to see their response plans. We
just asked them the questions. Does your response plan contain, and they said
either yes or no.

Now, they could say yes and it could be no. I don’t know what JAHCO does,
but not all the hospitals are accredited.

We did find that hospitals that weren’t accredited were less likely to have
done the drilling and the training that is sort of implied that should be done
every year to be ready.

MR. J. SCANLON: Is the JAHCO process kind of a self assessment or are there
visitors and an independent —

DR. BENTLEY: Let me answer a couple of questions. There are basically two
ways that hospitals get accredited. An accreditation has two purposes. One is
to participate in medicare and medicaid. A second, prior to medicare and
medicaid, is just simply as a good housekeeping seal of approval.

MR. J. SCANLON: Standard of quality, standard of care.

DR. BENTLEY: Because of medicare and medicaid, you are either accredited by
the JAHCO or, if you are an osteopathic hospital, by the American Osteopathic
Association, or you are accredited by a state agency. Particularly when you get
to the smaller hospitals, it tends to be by the state agency.

If you look at the JAHCO accreditation standards in this area, what they
require — and it makes it real difficult for people in this building trying to
do preparedness planning — they require, as a very first step, that the
hospital go out and assess the likely process — called the hazard
vulnerability analysis — in its community.

So, if you are in Richmond, Virginia, at the top of that list is going to
be chemicals. If you are in Norfolk, Virginia, at the top of that list is going
to be naval munitions.

That is the most likely cause of a major incident. If you are next to a
nuclear power plant, you are going to have that as your major concern.

So, part of the data collection difficulty JAHCO faces is that it is driven
off, what is your likely problem or likely problems, in a probability sense and
in an impact sense.

Then you design a range of plans that cover hopefully a whole set of
incidents, but focused primarily on the most likely ones.

So, it is very hard for them to go abstract and say, okay, we are going to
take the response plans for Richmond and match them up, say, against Chapel
Hill, North Carolina. They should be different and they will be different.

What you can say is, they have a plan, JAHCO has reviewed it, they have
done two exercises a year, one based on an internal incident, one based on an
external incident, but that isn’t the kind of detail that this survey has
historically looked at.

DR. MARCOZZI: With all deference to my colleague from the AHA, I am from
Duke University. You mentioned Chapel Hill, so I have to come to the table now.

I find the discussion pretty interesting. I am Dave Marcozzi from the ASPRS
office, previously OFEC, renamed in the pandemic and all hazard preparedness
act.

The discussion really ranges from the tactical level about how is the
survey conducted and whether or not it is practical to do a policy of the kind
we are discussing, the breadth of that, that is really challenging and I find
that pretty interesting.

I just want to try to align some of the arrows. I would suggest and
encourage all of us take a look at the bill that was recently passed back in
the December, that actually had the AHA’s support, that kind of repositioned
and tried to take a unifying strategy from the states and hospitals to their
preparedness grants and cooperative agreements and submitting them to HHS.

Then, from HHS, submitting the same parameters to congress every four years
in what is being coined, the national health security strategy.

In addition, this is a very interesting discussion about, really,
situational awareness, which is another mandate within the bill.

It mandates in the bill, number one, a situational awareness capacity
similar to biosense, that looks at mitigating an event and, in addition, the
management of that event.

That management of that event certainly impacts what we are discussing here
with regard to surge capacity. So, the pandemic and all hazard preparedness act
at least took a step to try and align the arrows in our discussion today. So, I
would encourage you guys to take a look at it.

DR. VIGILANTE: Rick and Katherine, thank you very much. It was a great
presentation. We appreciate the effort. Terry, you are up next.

[Brief recess.]

We are going to move Dan up here.

Agenda Item: Regional Trauma Center Preparedness and
HAvBED.

DR. HANFLING: I appreciate the opportunity to be here. I will try to focus
my remarks around situational awareness in real time, or near real time,
medical situational awareness.

It is funny to be here today in the immediate aftermath of this ice storm
that I know affected some folks more than others.

When I think about surge capacity, the ice storm of 1999, which the media
has been talking about a lot the last 24 hours or so, we had an unbelievable
onslaught of patients with mostly fractured extremities, from that event.

When I turned to the hospital administrator and I said, we are going to
open up in our new atrium, he said, oh, no, you are not. I said, oh, yes, we
are.

We basically instituted a real time surge capacity plan right there and
then that allowed us essentially an additional 20 beds in a space that was,
until that point in time, sort of public space in the hospital. It is just, I
guess, ironic that we are talking about surge capacity here today.

So, let me focus on situational awareness. I think it really provides the
ability for hospitals to make the choices that need to be made in the context
of an immediate uptick in demand for patient care.

I think that it also helps us with understanding some of the more complex
issues, as Jim Bentley and I talked about earlier this week, regarding the
implementation of altered standards of care, which also figures into the
discussion around surge capacity.

When I think about the kind of things that we talk about with regard to
responding to a surge in demand for care, we are either going to have to be
able to estimate the demand and then calibrate the supply, or we are going to
have to increase the supply, and that may just be a matter of improving
distribution, or we are going to have to scrap the system and create something
completely different.

So, from a surge capacity perspective looking forward now, situational
awareness is really going to be critical for us to be able to manage these
sorts of events.

If we don’t, we are going to be closed for business, and I think that is a
point that I would emphasize over and over again and, again, as was mentioned
in the beginning remarks, Katrina shows us where, in the most critical of
circumstances, this affected health care and the delivery of health care, and
continues to, to this time.

So, some of the questions from the committee, what data is needed, how can
that data that already exists be shared or how can information about hospital
capabilities become available in a real time format, I think those are all
legitimate questions.

I am not sure I am going to give you the answers, but I will try to point
you at least down that road.

When I think about the data that are needed and how they are best obtained,
really what I am trying to figure out is, where are the patients and where are
the resources to manage those patients. For me, it is as simple as that.

Again, based on very real experience, when we look back five years ago to
the 9-11 response here across the river in Arlington, Virginia, Anova Fairfax
Hospital, which is the level one trauma center in Northern Virginia, received
zero patients from that event.

This comes out of the Arlington County After Action report. Why was that?
Well, there was an absolute fundamental lack of situational awareness in real
time around where the available resources were, i.e., trauma resources in this
case, versus where the patients ended up.

This speaks to much deeper broader issues that probably go beyond the focus
of this group, but clearly, situational awareness will go a long way toward
matching available resources, available medical resources, to patient needs.

I think that is really critical because, there lives are hanging in the
balance, basically. Although we don’t think that there was a significant
adverse outcome with regard to those patients who were then secondarily
triaged, the next time around we may not have those opportunities. These may be
bigger events, they may be more wide scale or they may be more sustained. So,
it is important to match those resources.

I don’t have a slide to talk about anthrax, but when we talk about
sustained events, like the biological attack or emerging infectious disease,
this becomes even more critical, because these events start slowly and then
continue and plateau over a period of time.

There you really want to have an idea of where you are able to move
patients based on available resources. Again, going back five years, you recall
the events started here, just up the hill.

Three of the five patients in the national capital region came to northern
Virginia hospitals, two of them my hospital, and yet this was, at the outset, a
D.C. event. So, situation awareness is critically important.

Some of the efforts that we have put in place in northern Virginia to
address this has been what is now five years on a fairly sophisticated
coordination of all the regional hospitals — all the hospitals, I should say,
within our northern Virginia region, under the HRSA division.

So, the state of Virginia is split into six different regions for HRSA
grant delivery. The northern Virginia region has created this framework around
the integration of all the emergency response disciplines and then the
hospitals specifically, in what we call the Northern Virginia Hospital
Alliance, is integrated under our regional hospital coordinating center.

So, we have created now a format for regional health care coordination and
cooperation that addresses a number of the points that were asked in Rick’s
survey about MOUs and so forth.

In addition, we have put in place an information management platform and a
communications capability that really links us together.

This regional hospital coordinating center is a concept actually that I
took to the state just immediately after 9-11 and said, you know, we really
need to create this regional clearinghouse.

Truth be told, I borrowed very, very selfishly, I guess, from a concept
that was already in place in the state of Maryland, in the context of the
Maryland Institute for Emergency Medical Services, a real systemwide and
statewide system, as well as from the D.C hospital association, which had also
put in place at least a communications tool.

We took this to the next level and actually were able to put this in place
across the state of Virginia. in Northern Virginia, I would posit to you, we
probably have taken this the farthest down the road toward really integrating
this health care coordinating function.

What does that look like? Well, we talked about ambulance diversion as sort
of the lead in to some of this discussion.

When I was the medical director for Fairfax County fire and rescue starting
in 1998 we were in the throes of an ambulance diversion crisis here.

I was quoted as saying that ambulances were circling the beltway, even
though I didn’t say that. That is how it came across in the press.

The fact is that we needed to get some real time situational awareness in
terms of where we could safely send patients and not have emergency departments
put up the barrier and say, we can’t take them.

In fact, we put in place what was initially a fax-based system that then
became an internet based system, that we purchased from a vendor.

After 9-11, all of a sudden, there was a lot of interest focused on this,
and communities across the country were putting in ambulance diversion systems,
and there are a couple of big vendors who have done that.

Well, what was very interesting was that we had, in northern Virginia, one
vendor. The rest of the state of Virginia had another vendor for all the other
60 hospitals across the state of Virginia.

You know, I felt pretty strongly that the vendor who I had selected back in
the late 1990s was doing a good job for us and I said, it should be more than
just the commercial interests of a vendor and having to kowtow toward well,
these guys have a bigger slice of the pie and we are going to move to them.

We began to talk about the need to create standards for information
exchange, basically. So, in combination with the fact that we had a system in
northern Virginia, the rest of the state of Virginia had a different system,
and then my colleagues, John Donahue amongst them, representing the state of
Maryland, had another completely sophisticated, very well honed system, we
really began to have the discussions around how to coordinate the creation of
those standards. In doing that, essentially there came an understanding that we
needed to move this forward.

Now, I will digress for a second because some of you may be familiar with
this slide. This comes out of the medical surge capacity and capability project
that was done here in the office of emergency preparedness.

Essentially, you know, what this highlights — and put the red arrows there
in, those are my arrows, the rest of the slide comes out of the book — but the
fact that, from my perspective, we really need to focus on coordinating the
hospital, which is at the very bottom or tier one, to the coalition of
hospitals or other health care providers, which is at tier two, and then to the
state level, which is really tier three, which is sort of jurisdiction.

It is actually pre-state, but jurisdictional incident management, and then
you get the state sort of looking down.

So, from a conceptual framework, I think that this is useful and may be
useful as discussion continues forward around these issues, at least in terms
of theoretically laying this out.

Now, how can hospital capability information become available in real time?
Out of the discussions that we had in northern Virginia with the MIMS folks in
Maryland, with the D.C. hospital association and then with Richmond and the
Virginia department of health, we essentially worked toward trying to figure
out how to create those standards.

Actually, credit to Sally Phillips and AHRQ, who took this up at a national
level and basically funded this HAvBED project, the origins of which, actually
in part, came out of discussions that we started here in the national capital
region.

So, I will show you that in a second. What I won’t show you is the other
key area of real time situational information acquisition that I think is
critically important, and that is in the real of patient tracking and I think,
again, alluded to in the introductory comments in regard to what happened in
Louisiana, Mississippi and os on, but particularly in New Orleans.

Where were the patients, where did they go and how do we get a sense of
that. That becomes critical information different from bed availability.

These slides from my colleagues, Steve Cantrell at Denver Health, which
highlights essentially where the project is right now, this HAvBED project,
essentially here you see a map of the United States.

You can actually scroll down by FEMA, NDMS regions or by other regions,
geographic regions, if you will, which I think you can see listed here.

Here we are looking at the northwest and, if we hone down further, we are
looking at the state of Washington and, within the state of Washington,
Seattle.

Within Seattle, it is hard to pick up, but there are green dots that
represent where the hospitals are and then you can cone in and, actually, for
any given state or city or region or so on, based on voluntary reporting of
this information, at this point in time, you can get a sense of available beds
as defined by the currently accepted now NDMS categories for beds, which are
listed here, ICU, med/surg, burn, pediatric ICU, psychiatric, negative
pressure, operating rooms and so on.

Then you could actually scroll down and you could look at each individual
health care facility to identify the information resident in each individual
facility.

DR. BOENNING: Dan, how often is that updated?

DR. HANFLING: Here are the issues with that. It is a voluntary system right
now. I showed Seattle so that I could talk about it without having to share
with you the difficulties that we have in northern Virginia, but I am really
going to talk to you about our northern Virginia experience.

We actually have just re-upped our regional hospital MOU, our regional
emergency operations plan for hospital response, if you will, in which we have
now gotten the buy in of our board of directors, who are our 13 hospital CEOs,
to actually have our health care facilities update on a once per shift basis.

Now, we know that the compliance with that will be poor at first, but I
think it represents a huge step forward for us, to have commitment of our
administrative leadership. I mean, we are talking about CEOs of all the
hospitals in northern Virginia to say that this is important for them because
they know ultimately this allows us as a region to protect ourselves and to be
able to gauge what it is that we have to deal with going forward into some as
yet undefined crisis.

How that is going to play out, I hope I will be able to give you results
down the road. I can tell you, going back to Art Kellerman’s point, with regard
to ambulance diversion, at least in our community, it was such a crisis that we
realized we were better off working this out together.

Although we used to say, look, ambulances, they have wheels, hospitals
don’t have wheels. We are not going to be able to run away when an ambulance
comes driving up our ramp.

It is to our mutual benefit to be able to take care of patient needs. So,
we have been pretty successful in being able to report that data. It is an open
system. So, everyone can see everyone else’s availability per se, and then we
run reports and, at the regional level, we review that on a monthly basis and
so on.

We are hopeful that around some basic information with regard to bed
availability and basic resources, that this is going to meet with the same
successes.

I put the caveat which is, although you are looking at this from a top down
approach, my sense is that it is also important from the bottom up, in terms of
being able to make informed decisions with regard to information, to delivery
of patient care.

I will finish by suggesting to you again — some of you have seen these
slides before — if we can’t estimate the demand and calibrate the supply, or
if we can’t calibrate or improve the supply, we may have to just come up with
an alternate model for care delivery.

You know, maybe this is kind of just being too cute, but my concern is, on
the Y axis we sort of have what I Call echelons of care, which really start at
the home and go all the way up to the hospital.

On the X axis it is sort of echelons of contact, how do you get information
out. We have been talking up to this point about information for us. We are in
the health care sector, we have got to figure out what is going on, but really
ultimately we have got to figure out how to get this information to the
citizens. We have got to get this out into the community.

I would suggest to you that, if we don’t do a whole lot of effort in that
realm, we are going to have a lot of peer to peer networking, Aunt Phyllis is
going to call Uncle Bob and somebody is going to call Marjorie up the street
and Kevin is going to call and say, what do I do. Nobody is going to have good
information and they are going to flood hospitals.

I think that that is to our detriment and ultimately puts us at tremendous
risk, as we saw in New Orleans.

If we can work out what I call echelons of contact — that is probably not
the best phrase, maybe echelons of communication — that are incorporative of
broader strategies and including, for example, using local AM radio and, at the
national level, what I would posit to you is taking advantage of our existing
cable networks, our satellite radio and television networks and so on, and
getting information out in that manner.

You know, we may do a better job at being able to direct flow and reduce
the crush on demand for care where it isn’t always necessarily the hospitals
that have to deliver that care.

That is kind of an aside, but I think that has to be considered in the
context of discussions around assessment of data needs, because ultimately I
think the real stakeholders here are our citizenry, who we hope we can give
them good information to make informed decisions about where they go.

DR. VIGILANTE: Thanks, Dan. Questions for Dan?

DR. SNYDER: Some of us from NIH had a very good opportunity earlier this
week to see the military counterpart of what you just presented.

I just wanted to bring to the committee’s attention that it may be useful
to involve the military folks, the Department of Defense, at some point as this
evolves, to involve them in the discussions.

What they showed in terms of real time data management and capacity to
identify in a federal express type manner what is happening with each and every
one of their patients, whether it be in the Iraqi theater or elsewhere around
the world, is truly remarkable.

What they have developed and what they have actualized, what is operational
at this point, is worthy of this committee’s attention or considerations.

The other thing I just wanted to bring up is that one of the things that
some of us who have actually spent time in New Orleans, either during our
careers or visiting in the aftermath of the hurricane, one of the things that
we have learned, I think — and I will give an example — the universities and
academic institutions in that area have expanded the concept of the classroom,
physically, geographically, what have you, including using hotel space as
university space.

I guess, reasoning by analog, I have yet to hear much in the way of a
discussion of concepts of what a hospital bed is, or what an expanded concept
of what an emergency department bed is.

There is obviously the concept of hospitals without walls and emergency
departments without walls. I am wondering if there isn’t a risk of insular
thinking here, particularly by those who have vested interests.

I worked in an emergency room for 15 years, but I no longer represent that
element of health care. I know some of you do and indeed perhaps have vested
interests in emergency departments with walls and hospitals with walls.

I wonder if part of the thinking and part of the discussion should be
expanding the concept of what it means to be an emergency department, what it
means to control the concept of an emergency bed or a hospital bed, if you
will.

DR. HANFLING: First of all, both points are well taken with regard to DOD
and their real time capabilities. I think that is worth looking at.

We have had some discussions with DARPA and I have heard that there
probably are some opportunities to sort of share from their experience in the
real world experience, particularly over the last couple of years.

You know, the model that I put up at the end really, I think, begins to
address that point, which is that, on that Y axis there are echelons of entry
into the health care system.

I didn’t go into it in great detail but, in fact, there has been a lot of
discussion around the development of alternate care facilities, and alternate
care facilities, in my mind, being different than alternate care sites.

The ice storm of 1999, when I set up 20 beds in our newly minted hospital
lobby atrium, that became our alternate care site, and we delivered limb saving
orthopedic care right there and then, and we put up — well, maybe I shouldn’t
be describing this within this building, but we put up, as best we could,
patient privacy barriers and we put in a light box and so on and so forth.

Take that to sort of the grandest level. There has been a lot of discussion
and, in fact, AHRQ just recently participated in a project that came out of the
office of emergency preparedness here at HHS, looking at the allocation of
scarce resources in a mass casualty event, and recognized — in fact, I was a
co-author on a chapter that focused on the whole issue of alternate care
facility development.

So, the same way that you talk about hotels being used as school rooms, you
know, we have talked about other buildings of opportunity, even possibly
hotels, being used to deliver care.

I think that what is interesting is that the common denominator is, if you
can set that up in a community — and there have been some communities that
have done some work on this, for example, in upstate New York, the state fair
grounds, which Rick Hunt has done a lot of work around creating a surge
capacity facility if that was required — but the bottom line is, you need to
integrate them into emergency management.

You need to integrate them into incident management, is really what I am
trying to say. One way you do that is by linking the data and linking the real
time understanding of what kind of patients are there versus what kind of
patients are in the hospitals versus what kinds of patients are still out in
the communities. So, this actually I think is very much tied into that
expansion of thought around where you deliver care.

DR. AUF DER HEIDE: Could I make a comment kind of expanding on these
thoughts? You know, Henry Corentelli, who was the co-founder of the disaster
research center and the author of a landmark study of EMS and disasters, made
the observation that, in order to do effective disaster preparedness, you
really need to take a systems perspective.

I am kind of happy to hear some of the thoughts that are coming out of the
discussion just passed because it kind of alludes to that requirement.

There are a couple of observations from a number of field studies and case
reports and so on that I think are relevant to this.

One is, just to kind of introduce the concept, really traditionally a lot
of our focus on surge capacity has been on hospitals and, to some extent on
alternative care facilities.

The aspect of this surge preparedness that seems to be lacking is on the
other non-hospital medical facilities in communities, for example, physicians
offices, pharmacies, assisted living facilities, dialysis centers, urgent care
centers, and so on.

You know, when you look at the epidemiology of disasters, there are a
couple of interesting findings. One is that the overwhelming mass of casualties
do not have very serious injuries, and many of these injuries could be taken
care of in a non-hospital setting.

The other observation which is of interest is the fact that, in a number of
recent disasters, the majority of casualties going to hospital emergency
departments were not trauma. They were medical cases.

In a substantial number of those, it wasn’t people who were injured or made
ill in the disaster, but people who had lost access to their routine sources of
medical care.

One example, the derailment that occurred up in Missisauga, Ontario, with
the chlorine release, that involved the evacuation of 250,000 people and three
hospitals and three nursing homes, people evacuated not knowing that they were
going to be gone for several days, thinking they were only going to be gone for
several hours.

They didn’t take their medications with them. Then they couldn’t get a hold
of their doctors to get them refilled because their doctors had evacuated and
the pharmacies had evacuated.

So, I think when we are doing surge capacity we need to think about ways of
reuniting people with their routine sources of medical care.

I think we need to talk about making sure that routine sources of medical
care can survive and function in a disaster and maybe surge a little bit
themselves. If they don’t, then their patients are going to end up at the
hospital.

One component, along this line, is not only being able to surge, but the
aspect of survival, and this is something that can be done before the disaster.

If you have medical facilities built in the flood plain, or if they don’t
have any back up power, then they are going to become disaster victims and
their patients are going to end up at the hospital.

So, I think when we are talking about data collection, we need to take this
into consideration. I had provided a handout with some data on this. I don’t
know if it got circulated, and also a paper on diabetic needs during Hurricane
Katrina, which pointed out that there was a prevalence of 11 percent of the
population with diabetes, and some of the problems they ran into because they
didn’t anticipate the needs for people who had evacuated.

So, one of those pieces of information might be what is the prevalence of
chronic diseases and ongoing diseases in the community. What kind of resources
might you have to provide if people lose access to the routine care for those
procedures and how that figures in.

That would include things like what medicines are people taking, how many
diabetics so we can talk about insulin and glucometers, how many dialysis
patients there are and what alternatives there are to treat them and so on.

I think that kind of broadens the perspective and gives us more of a
systems view of what the actual community health care surge capacity is.

DR. VIGILANTE: Thank you. We have distributed that paper.

DR. GREEN: Let’s assume for just a minute that we are most interested this
afternoon in systematic assessment of the capacity to surge.

Given Dr. Snyder’s comments and yours, whether it is your atrium or his
hotel, what is your thinking about the metric?

DR. AUF DER HEIDE: Who are you asking?

DR. VIGILANT: Well, he is looking at Dan for the moment, but we will take
the answer from anybody.

DR. HANFLING: Erik, let me take a shot at it, and then I would like to hear
your thoughts, too. You know, actually the key metric, to he honest with you,
is not so much flat space or ED treatment space, I think, was the phrase that
was used in the HRSA survey.

It actually is the care giver that can stand by the bedside to provide that
care. That, I think, actually is the key metric and is the rate limiting step,
if you will, around which the ability to surge in demand for care actually gets
meted out.

DR. SNYDER: And that means different things around the world in terms of
who is that care giver, what their level of competence is, training, et cetera,
et cetera.

DR. AUF DER HEIDE: You have been talking about real time data collection,
but I think it also would be important to collect some data after disasters to
find out how many people are showing up at the hospital who otherwise might
have been able to get care, had their routine sources of care been able to
survive and function.

That had some real important implications for where you might put
resources. Just to give you an example, in Hurricane Aniki in Hawaii it was
estimated that the cost of DMAT, disaster medical assistance care team, was
$1,500 per patient contact.

Most of these were not for critical injuries or illnesses but basically for
minor emergencies and family practice type of care.

You kind of wonder, if you had taken that same amount of money and had a
cadre of trailers with generators on them, that you could have taken around and
gotten physicians offices and urgent care centers and other sources of care
back up on line, if it might have been a more efficient use of resources. I
think we need to kind of expand our horizons and think a little bit outside the
box.

DR. VIGILANTE: Thanks. We are running a little bit behind and a little bit
of sequence. I would like to propose that if Terry and then Mike can present
back to back, and then have a little discussion, and then we will move on from
there.

Agenda Item: DHS Office of Health Affairs and Our Role in
Interagency Medical Planning.

DR. ADIRIM: I am Terry Adirim. I am from what is going to now be called the
office of health affairs at the Department of Homeland Security.

Basically, I am just going to give a couple of minutes about what our
office does. We don’t generate the kind of data that I think we are talking
about today. We are more of a coordinating and integrative function.

HHS, under HSPD8, is the lead in medical planning and response, under
HSPD5, if it is an incidence of national significance we then take the lead.

What happened was — and this is just a story that I heard kind of third
hand — but during TOPOF-3, Secretary Chertoff needed medical advice, and there
wasn’t anybody in DHS who could provide him medical advice, and he decided that
he needed to have a medical advisor.

So, in 2005, a chief medical officer, Dr. Jeff Runge(?), who is an
emergency physician, undertook that position.

This past year it was realized that there were a lot of other functions
that were needed to integrate a medical response and coordinate amongst various
agencies, and we have since undertaken some other tasks as well.

So, the chief medical officer serves as the Secretary’s principal medical
advisor, coordinates Homeland Security’s biodefense activities, as well as
DHS’s medical preparedness activities.

We are also the point of contact for federal, state, local and private
sector for medical and public health issues, and I will show you a little bit
about what our new org chart is looking like.

We have Dr. Runge, after March, will be the assistant secretary for health
affairs. We have a principal chief medical officer. Both are emergency medical
physicians with extensive EMS experience.

Then we have three divisions within our organization. We have weapons of
mass destruction, biodefense line. We have an associate chief medical office,
Till Jolly, who is an emergency medicine physician, under medical readiness.

We also have an associate chief medical officer who will be doing sort of
an occupational health type function.

A little bit about what kind of data we do generate, which isn’t really
within this line, but basically what WMD and biodefense division does is
collect information with regard to early warning for infectious disease
outbreaks.

Currently we have just undertaken the program biowatch, which was under
science and technology. it is a environmental monitoring and detection program.
Over 30 metropolitan areas have these filters in various places, like metro,
and collect information daily or more than daily, on infectious disease type
things.

Biosense is at CDC. Bioshield, we have a small component of that, because
Homeland Security provides the threat analysis that goes into the medical
countermeasures that are procured under bioshield, which is housed at ASPR.

We are now undertaking the national biosurveillance integration system, or
NBIS, which integrates all of this into one system, in order to collect data in
one central location, and they will be housed under the office of health
affairs.

Under medical readiness, which is where I work, I, too, am an emergency
physician. I am a pediatric emergency physician. I worked at children’s
hospital for many years here.

Our function is primarily as an integrator amongst agencies and coordinator
of medical disaster planning. For example, we are very active right now with
pandemic planning along with HHS.

We also take DHS’ lead with regard to state and local government and
private sector partners. For example, we work with our grants in training that
provides a lot of the funding to the various places in this country where EMS
and first responders get their funding. We aid grants in training with that
particular function.

As opposed to being a response unit, we are essentially a policy and
planning type unit and participate with ASPR and other parts of HHS like the
CDC, to plan for disasters.

Some of the questions that came up in our minds with regard to information
are very similar to what has already been talked about.

How do we collect this data and disseminate it to such disparate entities.
There seems to be a big silo effect in government and getting this data and
even knowing it exists is an issue, and especially for us, because we are not
housed at HHS, yet we have an integrative function.

So, how do we get this data? Where is it? How do we share it? I think these
are important issues for us. That is basically what I had to present today.

DR. VIGILANTE: What we are going to do is have Mike present his, and then
we are going to have questions for both, just so we can start to have some
economies timewise. Thanks very much.

Agenda Item: Case Study: Value of the Electronic Health
Record in the VA.

MR. VOJTASKO: Let me go ahead and get started. In the interests of time I
am going to try to go through this rather rapidly.

First of all, obviously I could not be there in person, but getting out of
the West Virginia panhandle was not possible this morning.

However, I do appreciate the opportunity to talk a little bit about what
is, in terms of bed surge, in terms of two programs that the VA is involved in.

There has been a little bit of mention of this in the previous discussion,
but I just wanted to give a little bit more detail.

First, we are basically 25 years now. VA under the public law, has the
requirement to support the Department of Defense in war time.

We do this through a plan that was put in place in the early 1980s and
still exists today. They are updating it based upon some changes in both DOD
and how the VA does business, but the bottom line is that, under the
legislation, VA can surge its beds based upon a reordering of the care that we
provide for our veterans.

In this case, when requested by DOD, active duty patients get bed priority.
In addition to that, we do the normal surge-type activities in terms of early
discharge, transfers to some of our secondary support hospitals, and so on and
so forth.

So, that is one activity that we have been reporting beds to DOD for, as I
said, about 25 years now, first on a computerized system that they had in the
early 1980s. That was updated twice, and I will get into that in a minute.

The second area is the national disaster medical system, which basically
was developed again in the early 1980s and basically put into place about 1984.

Of course, most of us are familiar with the response component with the
DMATs and the other response teams, but also there are two other components of
it that we saw during the response to the 2005 hurricanes, and that was the
evacuation component and the use of NDNS hospitals that have signed up for the
system.

During Hurricane Katrina and Rita, there were approximately 2,500 patients
that were evacuated to about 10 of our federal coordinating centers plus two
DOD facilities.

Those places were, as you recall, evacuated under rather the worst
circumstances, basically the kind of things that you might see in a combat
situation. It was rather chaotic.

Yet the patients got on board the aircraft and got to these reception
aircraft and got to the types of care that they required.

So, how we do that, a large part of that is done based upon our
coordination with DOD and their evacuation system, which serves both the NDMS
and the DOD itself.

Earlier I think somebody talked about a system that DOD now has that they
have been using, especially with the current Iraq war.

That system is an adjunct to their patient regulating system. It is called
the joint patient tracking application and it does indeed allow them to track
patients basically all the way from the battlefield in Iraq to Landstuhl,
Germany, and to the United States and into whatever hospital active duty
military eventually arrive.

The lead responsibility for the NDMS definitive care, which basically is
the hospital portion, as you see, is concentrated in the major metropolitan
areas.

Each area has a coordinating center, either a VA or DOD facility. Each of
those areas are selected based upon their ability to receive and distribute
casualties.

Being that most of that transport is by air, as we saw in response to the
hurricane, they have to be available to a major airport or something of that
nature, that can handle military aircraft.

These are the locations of the federal coordinating centers. They basically
cover most, if not all, the major metropolitan areas of the country.

We heard earlier, I believe, about a survey that I believe was done in
2003-2004 — correct me if I am wrong — with a third of the hospitals did not
know that they were NDMS facilities.

That may not have been surprising for that period of time. I will tell you
that I would hope that, since Hurricane Katrina and Rita, that has changed
dramatically.

Especially in October 2005, there was a new memorandum of understanding
developed between DHS, HHS, VA and DOD.

Part of that memorandum of understanding required new memoranda to be
executed with all the NDMS hospitals. So, between the response to the
hurricanes and the visibility of NDMS there, and the fact that just recently
over the last six months both VA and DOD have updated those MOUs, I would like
to think that today NDMS at those facilities is a lot more visible.

These are the primary STC responsibilities. I won’t go into detail except
the last two, and that is really the subject of why I am even speaking about
this.

One is to report local NDMS patient bed availability and then to receive
patients to the area, as we saw in the aftermath of the hurricanes and
transport the patients to the local hospital.

The bed count reporting is basically part of the same system. >From the
VA and DOD medical treatment facilities and coordinating centers, they report
the bed availability to a function that is located out at Scott Air Force Base.
It is called the global patient movements requirements center.

On the other end of it, the casualty requirements, movement requirements
are also reported to that entity. That entity matches up the patient
requirements with the available beds and the transport assets to move the
patients to these beds.

These are the basic bed and casualty reporting categories. I say basic,
because there is a whole system that stands behind this.

There are also 138 specialties and subspecialties that are mapped to these
five, and actually can be broken down simply to med surg.

It is based upon coding, but we can get rather definitive about the patient
requirements and then map it to these five patient categories.

Under NDMS, remember, the patient reception portion is important because it
is at that point that, when the casualties are received, in one of these
federal coordinating center patient reception areas, there is actually a
reception team that basically does the triage and actual assignment to a
particular NDMS hospital in that area.

The system that has been put in place since the early 1990s after the first
Iraq war to accommodate this bed and casualty reporting is the transcom
regulating and command and control evacuation system, commonly referred to as
TRACES.

The JPTA or the joint patient tracking application, is an adjunct to that
system. So, we can not only track patients and report beds and then track
patients through the evacuation system, but also then track patients to the
actual hospital where they could be cared for.

So, in essence today have about 25 years experience with ability to rapidly
report patients, rapidly report bed capability, but it is always — as we talk
about metrics and as we talk about this whole area, is to remind ourselves of
the obvious, and that is what we are really talking about is patient capability
and capacity.

Some have said that what we really need to be reporting is simply that,
ability to care for patients and not think about beds at all.

I would like to make one other proviso, that while we are focusing either
regionally, statewide, nationally or what-have-you, and we want to look at
numbers of beds nationally or whatever basis we can to surge to take care of
patients, that we must not forget that there needs to be a system and a
mechanism to move patients to those beds, and a mechanism at the local level to
coordinate the arrival of those patients and then move them to the local
hospitals.

So, as I said, I tried to do this very quickly. This is the last slide, but
to hit the high points, as I said, to indicate that there are some things going
on, have been for a number of years.

I would hope that all or some of the lessons learned and some of the
current capabilities that we are already involved in can be incorporated into
ongoing and future discussions in terms of surge capability in the United
States. So, having said that, any questions of comments, I will be most happy
to address them.

DR. VIGILANTE: Thanks. I think our plant is to have Art make a few comments
as discussant, and then just open it up to discussion, to discuss your
presentation as well as Terry’s.

Agenda Item: Discussant. Art Kellerman.

DR. KELLERMAN: Thank you. I am just going to offer a couple of short
comments. Corey Slovis, a Vanderbilt Emergency Physician, told me that most
people can’t remember more than five things. He always takes every lesson down
to five points.

I am going to offer two short comments, two very brief stories, because we
are in Washington and stories count as much as data, and then I will offer one
idea that I think plays off of comments that I have heard several people make
up to this point.

The two comments are, first, the best way to assure that a system can
function under extreme circumstances is to have systems that function
efficiently and effectively on a daily basis.

We don’t have that today. That came from the very first presentation that
we heard this afternoon. We are diverting hundreds of thousands of ambulances,
our ERs are swamped with patients. Our hospitals are full all the time and
understaffed and financially stressed. So, we have a huge challenge ahead of
us.

The second issue is, if we want to generate the data and metrics to more
effectively manage that system, we should figure out how to do that in a manner
that facilitates that daily function.

A tool that is operationally useful on a daily basis that helps hospitals
meet their mission, helps doctors and nurses do their job, gets the right
patient to the right place at the right time, and that can ramp up or function
more effectively in a mass casualty event, not something you pull out of a box
or you open up a room and turn on the power and suddenly everything is working
great, which is the model that we currently have.

Two quick stories that illustrate that opportunity and that challenge. The
first is, last August, about a week before I left to come up here for this
fellowship, I made the rash decision of embedding two reporters in my
department for a week.

They were with me on Saturday night about 3:00 in the morning. A CAT
scanner had gone down. We have seven critically injured trauma patients in our
four-bed critical care bay, four of them who were on ventilators with severe
brain injuries.

Our recovery room was full. All our ICUs were full, all our operating rooms
were full. Our hallways were completely jammed with patients. Blood was rolling
out on the floor and I got a phone call from a doctor in Delanaga(?) Georgia
with a critically injured 21 year old, single vehicle versus tree, begging me
to take the patient in transfer.

I ran over and looked at our rudimentary — and that is a generous term —
diversion spread sheet and the only trauma center that handles adults in the
metro Atlanta area of over 4.5 million people that was not on diversion, was
between this doctor and me.

I ran back and I said, well, there is one trauma center open and it is
North Fulton Hospital. He said, they have already turned me down twice.

They can’t do that. Well, they have done it. How about — doctor, please
don’t. I have called seven hospitals and everybody has told me they can’t take
this patient. He is going to die if you can’t help me.

This is with a reporter snapping pictures and scribbling on his pad. I
said, send a patient. What in the hell is a doctor doing calling seven
hospitals in a single cover emergency department in north Georgia who needs
care for a patient.

That drama plays itself out across this country every night with people
trying to get folks to the right hospital.

The second hospital is a post-Katrina story. We were part of the NDMS
system in metro Atlanta. We got 19 aircraft in six days and evacuated 1,600
patients. That was a small percentage of the estimated 300,000 who self
evacuated to Georgia, the majority of whom ended up in the metro Atlanta area,
and we were not pushed as hard as Houston or Baton Rouge or Dallas or a couple
of other places closer to New Orleans.

We got 600 acutely ill patients who went straight from Dobbins into
hospital beds in a city that, on any given day, has six to eight hospitals on
ambulance diversion.

Our counties decided that they didn’t need to open their EOCs because this
wasn’t a county level disaster, and the state EOC was swamped, and this was a
regional issue sort of below the state level, although they were open and
scrambling in a lot of other ways.

We muddled through but it was incredibly difficult. The real difficulty was
not the first or the second or the third day. It was about days five through
about 25, when we kept having people get shot and have heart attacks and
seizures and all the other things that happen in a metro area with 4.5 million
people, and now we had absolutely no inpatient capacity.

So, we ended up scrabbling together and tried to create a coordinating
capacity to figure out where we could get the next patient into a bed.

It was an incredibly difficult circumstance, one that was played out in
receiving cities around the country, but nobody paid attention because we were
all fixated properly on the Gulf.

So, the one idea, playing off the ideas that I have heard today is, as a
pilot program — because congress and administrations love pilot programs —
let’s look at developing five or six regional, not air traffic control centers,
but ambulance traffic control centers, that would function on a day to day,
night to night, basis, to monitor occupancy rates, emergency department
crowding, on call specialty coverage, which is an increasingly difficult
problem.

It doesn’t matter if you are a trauma set if you have got no neurosurgical
coverage tonight and you have a patient with a bad brain injury.

It monitors diversion status with the goal that it is going to be always
open all the time, but we have got to get there first.

That would not be for every single inbound ambulance, but it would
certainly be for the inbound ambulances and interfacility transfers with
critically ill and injured children and adults.

It would generate the metrics on a day to day, shift to shift basis that
could go straight into that fancy EOC that I have seen somewhere in this
building, that is a phenomenal data collection resource point, but doesn’t have
this kind of data showing in any city today.

Learn from Maryland. We haven’t heard about it yet, but we are going to.
Learn from Northern Virginia. There are some systems out there that are most of
the way, and maybe in a couple of communities all the way there, but let’s do
this in the cities that DHS would say are our most problematic cities for mass
casualty events.

See how it happens for two or three years. I think it would be a dramatic
improvement. It would improve day to day care of critically ill and injured
children and adults.

It would also create an immediate situational awareness that would allow us
to manage a mass casualty event, whether it is due to bioterrorism, a transit
system bombing, the emergence of an infectious disease or a natural disaster.
Thank you.

Agenda Item: Discussion.

DR. VIGILANTE: Thank you, Art, and thanks to the presenters. I just am
going to open it up now for comments on these most recent presentations.

DR. SNYDER: Art, are you prepared to give those half a dozen or so regional
traffic control centers a set of standards so that they could communicate with
one another?

DR. KELLERMAN: Absolutely. I think in the mind’s eye of the people who have
conceived this, there should be a uniform platform, a uniform set of functions.
I would envision this as something that would also absolutely require a state,
if not a community, partnership in terms of matching funds, because you have
got to have buy in at the local level.

Yes, the data should be collected and reported and managed in a consistent
manner, so that you can compare apples to apples and oranges to oranges. I
wouldn’t have each city inventing their own system and their own data metrics.

MR. J. SCANLON: If I could follow up, are you envisioning this as more of a
communication system, or would it be decision making?

DR. KELLERMAN: I would say a little of both communications and decisions
personally. This is clearly an idea that would have to be worked out.

I think back to that doctor in north Georgia. I suspect that doctors in
western Maryland today can call MEMS and say they have got a patient with
problem X. It is one call and the next thing they hear is a helicopter or a
ground ambulance is en route, and we will find the hospital to take your
patient.

I mean, you go back to taking care of the patient which is, after all, what
doctors really kind of want to do. Yes, I can imagine all kinds of folks,
managed care is going to get all upset, but if it is not our hospital, what
about this and that.

That is fine. If your hospital is on diversion and this hospital is the
closest with the appropriate staffing, you can cut a check and work with it
later. I think it would be in everybody’s interest to have this kind of
capability, but it would be operational and it would produce data of immediate
benefit to hospitals.

If I am having an acute MI and Mr. Bentley’s hospital has an interventional
cardiologist who can take me in the next 30 minutes, I want to go there. I
don’t want us to have to figure it out over the next 90 minutes to two hours.

DR. BENTLEY: I don’t disagree with what Art has suggested, although I think
it is more for the routine than the mass casualty setting.

We are behaving this afternoon as if we are going to have control and be in
charge of the patients when we have a mass casualty surge, and I really think
we need to disabuse ourselves.

If you look at what is happening, in every mass casualty incident around
the world, the medical system did not have control of the patients.

People went by scores. Every hospital is overloaded. It is not a question
of where we allocate patients from. Every hospital is overloaded.

Moreover, they are overloaded in many ways with patients who have the very
same medical need. So, it is not an allocation matter.

Whether you look at Taipai or Turkey or Indonesia or Pakistan, we have had
the luxury as a company to not — you could say we had one in Katrina, but I
really don’t think we did on the scale that a mass casualty surge is designed
to look at.

I think we ought to say to ourselves, separate from the other discussion
today, and not disagreeing with Art that that would be an improvement for
health care generically, but we have got to get beyond the point where we think
we are going to be able to direct and then charge patients.

There are going to be thousands and they are going to go — I will talk a
little bit when It is my turn about some things I think we should do as well to
complement what Art is suggesting, but I think we have got to be very careful
that we don’t get this model that says it is kind of an air traffic control
system, where we are going to know who the patients are, we are going to send
them where we have got resources. We are going to be swamped at a much higher
level than that.

DR. KELLERMAN: What I would say is, for the patients who self evacuate, the
walking wounded, the ones who get thrown in the back of somebody’s car, you are
right.

Erik Auf Der Heide, who is on this call, I think has written the definitive
analysis of the real experience of disasters versus our conceptual models.

For people who require transport, who are severely ill or injured, you
actually can do this fairly rationally. Nobody, I think, does it better on an
operational basis than the folks in Israel, where they have had a lot of
experience with this and have, in fact, built this sort of capacity.

One of the interesting issues in Israel is, if you go into their hospitals,
their inpatient corridors are set up to become inpatient wards in multicasualty
events because they double as their military hospital system as well.

I think there are some lessons we can learn here. I am not suggesting this
is a definitive answer. I am suggesting we can get the kind of surge data we
need on an ongoing basis by developing a tool that will benefit hospital
operations on a daily basis, not asking people to fill out a form or go to a
web site every day or once a week.

DR. AUF DER HEIDE: Art, just let me amplify on that a little bit. I think
what the data in the United States shows that, in most disasters, they are
characterized by a maldistribution of casualties.

A lot of this is due to the fact that a lot of the patients are transported
by non-ambulance vehicle. The fact remains that some of those patients, and
most often the more critical ones, which are transported by ambulance, you want
to be able to define which hospitals are being overloaded, and to redistribute
the load and to take those ambulances over which you do have control and make
sure they don’t go to places that are already overloaded.

I think the other aspect of that, for example, during the 1994 earthquake
in Los Angeles, ambulances were bringing casualties to hospitals that were
being evacuated because they were damaged by the earthquake.

So, having a real time intelligence and sort of traffic control
establishment, I think, really makes some sense based on the data and the
experience in the United States.

DR. HANFLING: Just to follow up on that point, though, the concept of our
regional hospital coordinating center that we have now in place in northern
Virginia, it started as what we call Medcom, which was really a medical
communications clearinghouse.

It is, again, I think, for discussion beyond this panel, but there is a lot
of education that has to go on in the hospital EMS community around decision
making with regard to moving patients.

I understand it. Again, I speak about this from their perspective having
been in the fire service, but the point about who is going to take control, I
think, is an interesting one.

That is why I go back to saying that it is ours to take back control over
and we have to have those real discussions with EMS.

Then we also really have to share information with the patients who
hopefully, in an informed capacity, can make some of those decisions ahead of
time.

DR. VIGILANTE: Jim, why don’t you do your presentation, because I think
this is going to dovetail quite nicely.

Agenda Item: Panel Two – State and Local Perspective.
Hospital Preparedness: Many Things to Many People.

DR. BENTLEY: In light of the time, let me try to make four points, although
admittedly, given some people know, some of those points will have subpoints.

One, there is less money available for data in the health care system now
than there has ever been. Now, that seems contrary to what a lot of us
experienced, but what we observe is, we are getting more and more push back.

As places have been pushed to be more efficient and as payments don’t grow
at the rate that at least the providers and practitioners would like them to
grow at, the people who are getting cut out of the system, at one time, when
the hospitals had a lot of assistant and associate administrators, assistant
this, that and the other things, there were lots of people to fill out
questionnaires or submit data.

That is disappearing. I fact, we had to shrink our annual survey because we
basically got enough push back from the members who said, shrink it or we won’t
complete it. It is that simple.

Secondly, as this committee looks at its task, I really hope it will ask,
what decisions are the data designed to address.

I am dismayed, I guess is the right word, that we collect an awful lot of
data without a real good idea of what we are going to do with it.

If we really want people out in the field to provide that kind of data, we
need to be able to communicate what decision that is designed to address.

Third, in terms of surge, which has at least been one of the themes today,
a couple of at least my observations, and what I hear from our members.

One, at least for the past five years, there has been too much of a
tendency to think of surges and inventory.

Surge is really a process. How do we go about taking whatever we have and
expanding our capability to care for people.

It has a couple of limiting steps. One is to think in the way that Dan
said, can we match demand to resources. In that sense, we need to have a model
of demand management we think about and, at least as we think about it, there
are a couple of things, four things, that we need to think about.

One, can we rapidly identify the non-recoverable patients. We have got to
help the public understand a military definition of triage in a terrible
situation, rather than a civilian definition, if you will, or a normal
definition of triage. Having identified those, how do we use the least
resources for them.

Secondly, can we maximize self care. How do we, as Dan has suggested,
maximize the ability of people not to present to any part of the system if they
can care for themselves.

I am old enough to remember when we taught people to do that back in the
civil defense era, or at least we thought we had them taught to do that, and
there are still a few buildings around town, if you watch, where you can still
see the sign up on the building. I hope they don’t have the foodstuffs and so
forth down in the basement.

Third, how do we maximize the distributed ambulatory care. How do we get
people to go to, if not their routine site of care, a distributed site of care
because, if they don’t, they are going to collapse on the hospital.

That is where they know, or at least they think they know, they can go, and
there are usually a set of highway signs that tell them how to get there.

They can’t go to the health department. They have no clue in most
communities where the health department facilities are. So, they will go to the
hospital.

Lastly, in the area of demand management, how do we develop shelter
capabilities, whether it is for people who are on dialysis, the Red Cross will
not take a woman who is pregnant more than 32 weeks. So, if you can’t house
that person in a Red Cross shelter, they are going to wind up in a hospital.

If they are only 32 weeks pregnant, you may have four or five weeks where
you have that patient really boarding, and it is a bad use of hospital or other
resources.

In addition to the demand management, going to the other side,w hat are the
resources the hospital really needs.

I would agree with the VA gentleman who said we ought to think less about
beds. If you look at what it is, one, as Dan said, clearly staff is the rate
limiting step here.

It is what makes, Jack, some of the use of alternative site facilities less
successful as people have explored them.

If you don’t have staff, and staff who know what to do, and equipment and
so forth to put in those sites, you may be using them for housing people but
you are really not using them for medical care because you don’t have the
capacity to do that.

What did we really learn in the last couple of years? It is utilities. If
you don’t have water, if you don’t have sewage, if you don’t have electricity,
stand by generators.

If I could get somebody to invest in anything, we don’t have stand by
generators that can last under load. Every major experiment that has involved,
or minor disaster that has involved, using them, the generators go down and we
lose capacity.

The hospital wants to know, what is the status of the water system, the
sewage system, the electric system, the gas and the oil.

Secondly, transportation. When we were working with the Department of
homeland Security in the very first days of Katrina, the key issues were two.
Could we provide security to the facilities that were there, public safety, and
could we get gasoline in the tanks of employees who only had enough gas to go
one way. If they went home, they were out for the rest of the duration.

There was no electricity to pump gasoline. There was no priority at the
pumps. We need to have the transportation system.

Interestingly, in Rita, on a call where we were working with the Texas
Rangers, they said, flat out, go to our web site, download our logo.

If anyone wants to get through who is an employee, make up their own
certificate and sign it. Nobody out there in the sheriff’s department knows
what my signature looks like. It works.

Third, if we are going to have resources, we really need the supply chain
to work. Do the big distributors have the capability to really make the things
move.

We live in a just in time world where Art and his colleagues are using it
up in the ED almost as fast as it is coming through the system.

We have got to keep — for our members, it is in many ways more important
to have data on that supply chain and where things are needed than it is on
patients.

Lastly, the public safety services. The hospital — there are four refuges
in our society, as best we can tell, the police, fire departments, houses of
worship of whatever denomination, and the hospital.

That is where people feel they can go safely and be taken care of in a
disaster. We have learned in every disaster the police and fire station doesn’t
work because there is nobody there. They are out in the community.

Most churches, synagogues, what have you, aren’t very helpful because they
are very low staff enterprises, and they are really not equipped to handle it.
So, people come to the hospital who don’t need the hospital.

That is the place where they expect there to be pharmaceuticals if they
need them. There is light. There is power.

One of our hospitals in Meridian, Mississippi used up a week’s supply of
food in the first 12 hours because they fed 3,500 non-patients three meals a
day. That was the only place they could eat.

Now, in terms of data, let me make five very quick observations. One, go
back to my point, what decisions are the data designed to impact. If we can’t
answer that, we probably ought to ask what we are collecting the data for.

Secondly, is the data useful to the submitter. I would underline Art’s
point. If it isn’t, it probably isn’t very good data and it will just wither
eventually.

Third, is the data collection process consistent with the one that has been
established as part of the national plan, that is, federal to state to local,
back up, up and down.

We get an extraordinary number of complaints from physicians we talk to in
their offices or from hospitals, that all kinds of other steps, whether it is
federal direct down to local, or whether it is study centers or whatever, keep
adding additional links in that chain and, the more independent links there are
in that chain, the less people know what to do.

Fourth, is data a byproduct of routine operations? If it is not, it won’t
be very good data. Lastly, simpler is better, and let me close with a story.

When patients were being evacuated form New Orleans, under the NDMS system,
the people in New Orleans couldn’t say where the patients had gone.

We put in place a reverse tracking system. That is work with every hospital
that received patients to say who did you get, where did they come from, so
that families could learn. Art may remember lots of hospitals in Georgia
filling out that data.

The reason for that is elegantly simple. I had a hand in the 1970s when
ASMR was the air force predecessor to this under the national defense medical
system.

You could not leave the ground as a pilot if you did not have a manifest
that you carried with you, and that you left with the dispersing officer. If
that plain went down, we wanted to know who was on that plane. It was that
simple.

In Katrina, when we called and were working trying to get that data, the
people on the airfield down in New Orleans did not have it. The people landing
the plane did not have it.

The reason was simple. The vendor that won the contract for the data system
won in a competitive contract where, to win, you wanted to find the most
sophisticated, complex, all needs system you can develop, lap top based.

We called the people in New Orleans and asked, where is the data and they
said, sorry, our lap tops lost power and we have no way to recharge them.

We had a very sophisticated data system. Had it worked, had they had
battery capacity, had they had lap tops, it would have been ideal.

It was designed to feed back out through the net. We need to remember, in
most disasters, sophistication is our enemy, not our friend, and we need data
systems that can work in very simple ways. The more elegant they are, the more
complex they are, the less likely they are to survive.

MR. J. SCANLON: If I could, your annual survey, if I could go back to this
a minute, this goes to all hospitals who are members and non-members as well.

DR. BENTLEY: Yes, it goes to every hospital in the United States.

MR. J. SCANLON: Do you get any data on this area, preparedness?

DR. BENTLEY: We get som every limited data and we do a couple of other
sometimes special surveys during the year on a kind of fax back kind of basis.
For instance, we have added questions on negative pressure rooms.

This is the constant balance of, with 1,000 masters, both us and a variety
of agencies and public interests and so forth, how do you keep the survey no
larger. So, if you add something, you can add anything you want, as long as you
tell me what you are going to take off.

DR. VIGILANTE: I think in some ways this goes to the heart of the issue.
The Hospital Quality Alliance, you have hospitals beset by all sorts of
stakeholder groups who want to measure quality in some way.

Hospitals are bombarded with requests for data, often, in different
formats, to answer the same question, and it is a huge burden of reporting.

So, the Hospital Quality Alliance is trying to harmonize these measures, so
that there can be consistent data requests that you answer once, that satisfies
everybody.

Now, we have NCHS asking hospitals for data, we have AHA asking questions
— on preparedness, to some degree — HRSA, JAHCO. Do we need sort of a process
to harmonize the collection of preparedness data that will satisfy a broader
number of stakeholders, and identify that data we need in real time, or almost
real time, and that data we need periodically or episodically over the course
of the year or annually. Do you think there is a role for that kind of activity
kind of going forward?

DR. BENTLEY: I think as the HRSA hospital bioterrorism preparedness was
originally conceived, it would have had that role.

What we have found, and what our hospitals have observed, is once you
decide that each state and locality was in the middle tier, then each state and
locality decided what it is they wanted to add, subtract, delete, modify or
change, and it lost that vision that was once there.

Whether or not we can capture that back, there is just an enormous tension
between what different levels of government — say anything about day to day
operations — different levels of government think they want to have.

Each level of government thinks their need is the most important. So, you
sit down with, whether it is state hospital associations or state health
departments, and it often, for reasons that are very appropriate for that
health department or that state hospital association or something else, it
makes sense individually, but the sum of them don’t make sense at all.

DR. KELLERMAN: Let me mention quickly, you made a comment about quality and
I am going to say this and wake up Sally Phillips on the phone.

A couple of months ago I attended an AHRQ briefing, or at least there was
an individual there talking about a new quality survey for hospital care and
hospital experience of the patients.

Five items on this survey dealt with bathroom use and bathroom conditions,
because that is an important part of the hospital experience. Not one item
dealt with emergency department care experience.

I asked the architect and they said, well, that is a minority — not
everybody that gets in the hospital comes through the emergency department.

I said, okay, do you have any items on your ambulatory care survey. They
said, no, because the emergency department is a hospital based provision of
care.

So, we have got two major surveys dealing with quality of patient
experience. Neither of them think that emergency care is relevant to the core
mission of their survey.

Thank God for NHAMSIS and Katherine Burt and people who have at least mined
that data set. Part of our problem is that we do have lots of surveys and lots
of data, but we are not measuring some of the critical items that we need to
measure for preparedness and for life and death care.

DR. BOENNING: I think that is ultimately where we want to get to in the
discussion.

DR. PRITCHETT: I am Beverly Pritchett from the D.C. department of health.
The simplest answer is that most departments of health, at least at the local
level, and most hospital associations, have been gathering this data for
several years.

You should have one of the handouts that is over here, that has a listing
of both the New York state survey, one of them has the listing of the D.C.
survey — the two-column one is the D.C. survey.

That is all the data elements. As you can see, that is two columns on front
and back, about seven pages worth of data, very, very detailed.

The New York one is the one that is called critical asset survey. It is a
little easier to read. The D.C. one is an electronic spreadsheet that I have
that we use.

Then the other format is actually screen shots of the system that they use
in the District right now through the D.C. Hospital Association.

This is only kicked in when an actual emergency event happens, such as a
train derailment over at Metro Center a couple of weeks back.

The reason that this is — and you can see that it contains many of the
pieces of data. This is very simple. It talks about equipment, talks about
personnel, talks about Q resources.

The reason that it is only kicked in is because this information is all
manual. If anybody tries to tell you otherwise, they haven’t been in a hospital
in a very long time.

What I would tell you is that I come at this from a very unique
perspective. in the fact that I have 27 years military experience. I have
commanded the hospital over in Iraq, the referral facility, the first year of
the war.

I had MASCALs in that hospital weekly. We learned how to do the things that
you are all talking about today on a day to day basis over there.

It was frustrating, but yet the military does it very well because they
have at least taught it for years. Unfortunately, for the last four years now,
they have actually been doing it again.

I would tell you some things. First of all, in the role that I now have as
the director for emergency health and medical services administrator, I am not
only responsible for emergency preparedness in the District, but I also provide
oversight, regulatory oversight, to EMS.

So,I kind of have a split role there, but everything that you have said
today are the kinds of things we grapple with every single day.

What you need to know is that, at least in the national capital region but,
from what I understand, across most of the rest of the country in large cities,
we are operating at disaster level every day, every single day.

The most recent statistics for our diversion is one year of closure in our
facilities last year, one year of man hours of closure for the District
hospitals.

The other statistic that you need to know is that only 60 percent of
inpatients in the District’s hospitals are District residents. Forty percent of
our inpatients come from Virginia and Maryland.

As John would tell you, the reverse is true also, that they oftentimes get
D.C. patients int heir facilities. When we talk about, as Art had said earlier,
you need to have this over-arching system that controls that is going on up
there.

It is a wonderful thing. We have one in D.C. They have one in Maryland. the
problem is, there is not one for the national capital region.

You know, this is the basis, elements, 101, of health care administration,
is that you have to think about your population catchment area.

Your population catchment area is not defined by state boundaries, and that
is where a lot of these issues come into play.

What is data to me is not necessarily valid data to John or to our friends
over in Virginia. What is affecting me today, though, is very likely going to
also be affecting Dan Hanfling within a matter of minutes or hours,
particularly in a MASCAL state. We don’t have the systems that cross talk.

Now, in the national capital region, I am very happy to say that we are
starting to have systems that are going to be the very exact same systems that
will be able to hopefully connect all this data.

It is called HC Standard. Virginia state uses that, and I think that is
what you were referring to before. John has told me that Maryland is willing to
do that.

It is a system that basically sucks up any other system’s data, translates
it, and provides you with a dashboard.

It is very simple to use. You can manipulate any data that is input into
it. I go back to the original thing. All that data is manual.

The only way you know what bed availability status is, is if you go down to
the admissions office or you walk into the emergency room, and you look at
where the cards are not in the slots.

That is true today in 400 bed facilities, not just 20-bed facilities in
rural United States. So, I want to emphasize the fact that, whatever you do,
whatever data requirements you prescribe, you realize that, if it is on an
emergency basis, it is somebody with a pen and a paper sitting there, counting
that, and inputting it into a data terminal.

It is not automatic. Until we come to a point in time where there is a
sensor that goes across the doorway and the stretcher rolls out and you know
that that is an empty bed, it will always be a manual system.

DR. VIGILANTE: Beverly, thank you very much.

DR. MILLER: I come at this mostly from the surveillance angle of things.
So, if my statements aren’t fully informed by 30 years of emergency department
experience, I apologize in advance.

In listening to Jim talk and also in listening to others talk, there were
two things that came out that I thought should at least be brought to the
forefront.

When you talk about getting all of this data and emergency use, we
generally tend to talk about two sort of use cases.

The first use case that, for example, Art mentioned is sort of everyday
hospital, ambulances, running around trying to figure out where to best
allocate.

The second case that we have spoken about is say the Katrina case, where
things in a certain area seem to be shut down.

I would pose that there is probably an intermediate case that involves a
large part of the communications infrastructure being down, but not a large
part of, say, the hospital infrastructure.

I am from Atlanta originally. Back in, I think, 1996, we had a big
hurricane come through. Power lines went down, phone lines were down, trees
were across roads.

You didn’t have flooding. You didn’t have massive flooding that would cause
hospitals to evacuate.

So you would have, say, an almost fully functioning emergency room, but
what you don’t have is the communications infrastructure to effectively convey
data on capacity or openings.

So, as you think about what data to collect and how you process it, I would
encourage you to think about that middle case, which I think is probably going
to be more like most of the disasters you will experience than, say, the
massive shut down in an area of Katrina, or the everyday that a train
derailment might be like. That is my first comment.

My second is on Jim’s point, that it is important that we maximize the
number of people who are doing self care.

I think a key part of that is not only telling people — I think we have a
tendency to try to come to an answer.

This is what you need to treat yourself for, influence, in the case, say,
of a bird flu outbreak. What I think we also need to focus on is what things
are not.

Richard Danzig wrote a very convincing paper, I think, on the need of
focusing on what we are not dealing with as well, in communicating to the
public.

The sort of self care basic prevention tasks that they can be performing is
going to be as much influenced. I think, by what it is as what it isn’t.

In the absence of knowing what it is, knowing what it isn’t could help
influence those thoughts. That would be my two centers.

DR. VIGILANCE: Thank you. Roland, why don’t we five in.

Agenda Item: Public Health Informatics Needs at the
State Level.

DR. GAMACHE: When I looked at the questions, I first was going to talk
about trying to issue vital or death certificates very quickly in a surge
capacity.

Reading the questions, I knew I was wrong. So, I asked five minutes of Dr.
Boenning and I called him and, after an hour of incredibly interesting talk, I
realized that I was pretty well off base.

I think what we have in place here — and I am going to talk about what we
are doing in Indiana. I am just going to talk from the perspective of the
state, but I have had discussions with Ohio, Rhode Island, Michigan, Illinois,
Kentucky, Wisconsin, Denver and Minnesota. They are all on a similar track as
this.

So, even though this is our perspective, I don’t think we will be that far
off from what other states are trying to do at this time.

Also, I want to agree with what everyone has said here. There are a lot of
things I was going to talk about that I am going to skip by, and I think I have
about five minutes of presentation here.

I am going to use some public health words here, because we are going to
divide it into surge. There are two parts of here, the surveillance of it, and
then there is the outbreak management part, and I think we are talking more
about the outbreak management part of the surge capacity here, than actually
the surveillance, although they are both hand in hand. We have questions in
both of them.

I am going to bring up two systems that we already have in place, newborn
screening and FAS. I am going to try to show how this links into what we are
talking about here as well.

It is an outbreak management thing, and then some other systems we are
trying to put in place in there. So, these are just some of the areas we are
looking at.

In order to do this, I think we have to give a little bit of perspective
from Indiana. These are health information exchanges in the state.

The green ones are functional health information exchanges. They have a
positive business model. They are making money at this point in time, or are
real close to making money. They are very close to breaking even.

The blue one is one that is, in fact, in place already —

DR. VIGILANTE: Do you mean clinical data of the patient?

DR. GAMACHE: Yes, and it is what is available electronically. So, it is not
all clinical data. Like we don’t have weights on people, but if it is a lab
result or diagnostic of some kind, that is already available electronically, or
if the group has an EHR, that information is out there and available.

One third of the medications are available on most of these systems as
well. That is because of the people for the groups that are doing this. So,
anyway, and we have a bunch more that are coming into place.

What that translates into, though, is over 40 percent of the population in
Indiana is already part of an HIE, which when I looked at that information, I
was really surprised.

That doesn’t count the tertiary referrals. All of these places are tertiary
referral centers. So, we actually have more data than that in these systems,
but for day to day, what is in place already, 40 percent are already in an HIE.

So, with that in mind, the other parts that I want to put in there, too, we
have newborn screening that is already sent to us by HL7, R1 codes, and we are
working with Reigenstrief and, of course, with Mark Overhedge as part of that
group. So, we are getting that information.

This is part of the infrastructure in place to get the reporting, that we
will talk about surge capacity, all reportable lab data.

The next part I think is the big one. We get chief complaint data on
presentation to all emergency departments, or not all of them. It is actually
over 70 now in near real time.

So, it is pretty much as soon as they come in, they are entered into the
admissions system, and we are getting the chief complaint information on that,
in order to do surveillance activities.

We have about 120 EDs in the state. So, we are over half. It is much more
than half the population. It is probably about 80 percent of the population.

We divided our state into 10 districts. So, there are at least two
hospitals in each district that are covered. So, we have pretty good statewide
coverage at the same time.

We are also working on the first stage of NEDS to go live in March 2007.
So, we have a lot of electronic systems getting ready to go

We are trying to find ways to share information that we are collecting with
the health department, with the hospitals and with the clinicians as well.

Then I am going to kind of skip this because I think we can get some of
that information .I think a lot of that has been said already, and I am not
disagreeing with what anyone has said here. I agree with all that has been said
already.

This is, I think, one of the big areas. The chief complaint data comes in
and it is put in one of these categories, and we are able to look at this
information. We put it into three-hour buckets and then do an analysis.

I have been in public health for 15 years and there are people there — I
am still kind of a neophyte, even with 15 years of experience there.

In all that time, we always get called by the provider when there is an
outbreak before we can find it ourselves.

In the last six months, we have found three outbreaks before the providers
have called us. This is one of those. This was the carbon monoxide poisoning,
six cases from this emergency department, and it was triggered.

What we have here is a way, then, of saying, here are people coming to the
emergency department, and we have an infrastructure in place to report this.

What we don’t have, and if I could do this system as well in order to tack
surge capacity with the people who are already there, the discharges don’t get
reported well. They are dirty, they take a lot of time, and it depends how
important they are, but the infrastructure is in place to do that.

We are working on a couple of places trying to study this and see what we
can do to help improve that. That was essentially what I really wanted to talk
about, and everything that has been discussed so far.

A lot of the different surveillance systems that we are putting in place
can help look at the surge capacity issues with this right now.

We are working very closely with Ohio. I think in the next month or so we
are going to be sharing information with Ohio on the two county border.

Really, the two counties on each side of the state, they share this ED
information as well. We are proud of the fact that we were the first state this
information on 70 hospitals with the CDC, in the biosense program, so we are
sharing this with the national program as well.

They quickly followed. Ohio was not far behind us and several other states.
So, they have over 300 hospitals that they are getting some kind of information
like this from at this point in time.

Some of the other things we are looking at, immunization through claims
based data. Again, this is part of that information back and forth.

We are looking at public health messaging systems. I just talked to Mark
and they are ready to go and provide the service for us, so we can have ways of
sharing information among public health communities.

The other big issue here, when we look at hospitals — and we have heard a
lot of talk to this — we really only have one city in our state where there
are enough facilities that we really have to worry about how we divert
patients.

Most places have two hospitals. When I have talked to providers in those
communities — I know a lot of you work in EDs — when there are two hospitals,
you have to make one phone call, because you are at one of the facilities
already.

So, it is relatively easy. It is when we get into some of the places we are
talking about here, where we have multiple jurisdictions and we have multiple
ambulances serving those communities.

So, we have really one community we are looking at. They are trying to put
a system — Marion County health department has been very active in this and
trying to develop a system, and they are getting ready to go live with this.

We just signed the MOU to share he FES data, the ED data with them. They
are going to have the data from all the ambulance runs and the chief complaint
information from the ambulance real time.

We are looking at surveillance systems to monitor this. Their big interest
is finding out where patients need to go, to provide diversion real time to
ambulances. So, that is something that is going to happen there this year. I am
very excited about this. It cost over $2.5 million to set this system up. It is
very, very expensive to do this real time.

Part of what we are doing now, when a person picks up the patient, they are
making the call, a 10 or 20 minute drive from where they are, the location, to
the emergency department, diversion occurs in that time frame.

So, that is part of the issue as well, and why they are looking at it in
the system and trying to figure this out in real time and manage it quite a bit
better than they are right now.

This also helps with outbreak management and trying to manage that surge
and get it forward. All the other issues, the big one that came across in all
the claims that we did, we can find lots of space to put patients. Trying to
staff those beds is really the big issue for the health department and for the
state as well.

How do we staff those places. We are trying to address that issue. We are
looking at different professional development ways to do this, and how we can
use health professionals in ways that are a little bit kind of on the edge to
make sure that we can staff these areas where there is not as urgent care
needed. A lot of that has been addressed already by people who did a very good
job of it.

The other part of it is to look at, almost every hospital has an electronic
admission and discharge system. We are not looking to gather data any different
than the way the hospital already manages its data already.

That is true in the FES data as well. We are not asking them to enter this
data again. It is collected off the electronic admission information.

So, they are already entering it into their system. It is a very
uncomplicated system that just extracts the information from their electronic
system already, their electronic registration system, and collects that.

We are looking to try to do the same thing with the admission and discharge
at the hospital. Again, the admissions come in, I think pretty well.

This is, again the discharges and some of the other dirty data and some of
the coding that we are really trying to work on, but there is a project going
on right now to try to pilot this in some other hospitals to see if this can
actually be a way to look at patient census.

Right now, when you look at the census, you start off at the correct figure
and it starts to rise. You have patients who are there 31 days before they are
discharged, but they really were discharged a long time ago. So, we are looking
at ways that we can try to identify that and move forward with that
information.

My point here was to get some information out quick and stimulate
discussion. Dr. Boenning said about five minutes and I think I was close.

DR. SNYDER: Jack, a quick question for you. Are you saying that somewhere
on a server or a network of servers in the state of Indiana or Marion County,
that you are actually capturing pre-hospital electronic clinical information?

DR. GAMACHE: We are going to. We are not right now. The pre-hospital
information they don’t have now. That is going to go, in fact, this year.

Everything is in place except capturing the data. The data is being
captured– it is being captured in one place but it is not being integrated now
with the hospital information. That is the next step.

So, there is really no good way to analyze it at this point in time. The
Marion County health department is going to do that analysis and they have some
very good capacity and great people behind it to do that work very well.

DR. HANFLING: A question about the implementation of the ESEN system. Did I
understand you to say that reports are being delivered back to hospitals based
on the input of their data, chief complaint data? That is the first question.

The second is, with regard to the successful pick up of the CO poisoning in
patients, what is the mechanism for alert and notification to the hospital?

DR. GAMACHE: Good questions. Right now hospitals can get that information
but not everyone is signed up for it at this point in time. A lot of places
don’t have the capabilities to do that or don’t want that burden on them.

When we do see an outbreak, the first thing to do is call the emergency
department. So, it is a phone call. We also have state epidemiologists in every
one of the districts.

So, we follow up with the health departments. So, basically we notify the
emergency department, we notify the local health department involved and we
notify our epidemiologist in that district that we have a potential outbreak.

So, that is kind of – -usually the follow up dismisses most of the cases.
We find out what is going on very quickly.

A lot of the alerts that come up are not necessarily outbreaks, but the
ones that we have, the few that we have — you are seeing here the three — a
quick phone call identifies them very quickly. So, it is not a lot of extra
work but occasionally it does come up with a quick follow up.

DR. HANFLING: I am not sure how many hospitals. I think you said there were
— how many hospitals in the state?

DR. GAMACHE: Seventy out of about 120.

DR. HANFLING: So, those calls go out serially to each of the 120 hospitals?

DR. GAMACHE: It would just be to that one facility affected. It is usually
just one place that is affected in that case.

We did have one in Marion County where it was more than one facility. We
coordinated with the local health department then to make sure that we talked
to everybody. That is a coordination effort that is — the state health
department doesn’t know all the details well enough locally in order to work
that way.

So, we partner with the local health departments. They are much better at
addressing those issues in their community than the state is, but we do offer a
lot of technical assistance for other health departments that don’t necessarily
have all the expertise that they need or the staffing that they need. So, it is
kind of helping or an aid in that case.

DR. HANFLING: This represents, I think, a very good story, if you will take
that phrase, in terms of the role of syndromic surveillance.

You know, there has been a lot of discussion, a lot of controversy, even,
about what the cost benefit is. It also speaks to really the need — again,
back to the charge of this committee — to look at proliferating that
information in real time.

What is CO poisoning on one day could be the outbreak of some emerging
contagious infectious disease the next, and then you really want to get that
information everywhere as fast as you can.

DR. GAMACHE: We are working along those lines with the IHAN system as well.
So, we have st that up and then we are working with our health information
exchanges that have several thousand physicians already signed up.

There are over 4,000 in central Indiana. There are another 1,000 I think in
the Fort Wayne area. So, we have over half the physicians that we can actually
get messaging out to very quickly.

Then we have alerting systems already for most of the major health
departments and working to get the other ones connected at this point in time.

When you are in the emergency department, the best thing still seems to be
the phone calls, because they have to go some place else, they are looking some
place else for this information.

It is not always easy when you are getting overwhelmed with patients going
in. You are not going out to the computer to see if there is an alert at that
point in time.

DR. HANFLING: Pushing the information and going back to Beverly’s point
about coordinating, now bring it back at least to the NCR.

While we don’t have a data management system that is coordinating the three
jurisdictions, by putting in place a radio system, in northern Virginia, that
was tied to the District’s HA hospital mutual aid radio system, and that in
turn was tied up to SISCOM in Baltimore.

We actually have the real time capability right now to talk by radio, 800
megahertz radio, amongst all the hospitals in northern Virginia, the District
of Columbia and the state of Maryland.

So, it goes back to your point about collecting the data, having sort of a
data management capability, but also having the means for alert notification
and pushing information out or at least getting people alerted about something.

DR. HANFLING: It is the aware part of situational awareness.

DR. AUF DER HEIDE: It was mentioned that we had a two-way radio system
throughout the area there. I think it is important, in any of our deliberations
about real time data collection, that it is based on two-way radios. One of the
first things that goes out with almost predictable repetitiveness is telephones
and cell phones.

DR. GAMACHE: That is why we include our epidemiologists in this. That is
our most redundant communication spot, is a two-way radio. All the hospitals do
as well, but that is our point of contact to make sure.

So, we have looked at that redundant communication capability and a lot of
it is through radio and back ups of that sort.

That is kind of important, because in our case, what we are always afraid
of, being along the fault line there, is that we will have an earthquake. We
are also worried about that infrastructure with all the antennae falling down
at the same time, too. So, there are issues there as well. So, we are looking
at a whole bunch of different systems for that.

DR. BOENNING: There is a similar system like that called SERIS, like ESENS,
that is in use in the southwest. In the interests of time, I don’t think we are
going to be able to see a demonstration during the formal session here, but
Andrew Miller might be able to demonstrate this after adjournment. Another
resource you want to keep in mind is amateur radio services.

DR. AUF DER HEIDE: One issue related to that, that could be a stumbling
block, is that you are not allowed to encrypt data sent by amateur radio,
although I would imagine legislatively that could be fixed.

DR. VIGILANTE: Any other questions? I think our colleagues from Maryland
have a contribution and then I would like to sort of complete that in a few
minutes, and then we can have some discussion and wrap up about next steps and
where we are going from here. I am sorry, first we will hear from Sally and her
point of view.

DR. HANDLEY: While John gets set up to show you things on the screen, I
would like to talk about just a couple of things that happened or are happening
in Maryland that might be germane to the discussion.

One of those is, within Maryland, we have a communications backbone,
microwave, that can override a fiber optic background, but it is purely stand
alone.

Really, a paramedic in Ocean City down on the coast can get a referral or,
not a referral, can get orders from a physician in the ED in, say, western
Maryland, if needed.

That communications hub, that we call SISCOM or EMRC, sits at MIMS, and it
facilitates the communications between the paramedic and the hospital.

The fact that that place exists, that capability, led to a program that we
worked on with Baltimore City, where we talked about this air traffic control
notion.

Baltimore city was wrestling with the fact of a lot of their ambulances
being locked up at hospitals due to over-crowding issues.

So, what we did was, for certain categories, severities of patients, a
Baltimore city dispatcher sat in our communication center as a pilot and the
paramedic, prior to transport, would call and say, I am headed to X with an Y
priority patient. Is that okay. Sure, go ahead, or the answer might be,
negative, divert to.

Now, what makes that capable is not only our ability to talk from field to
hospital and all those communications pass through MIMS, but also we have a
running status at any time of a program called CHAT, with what we call yellow
alert, red alert, where a hospital, typically the ED, will call us and say, you
know what, we have no inpatient capacity or we have no acute bed capacity.

We will put that hospital on red alert. That will show up on a publicly
available web site. You can go look at the status of any hospital in Maryland
at any time, see what type of alerts they are on.

So, we have that ability to know where to divert those ambulances to. With
that, the amount of over-crowding — I say over-crowding, over-crowding still
existed — but its impact on EMS was lessened, and the return to service times
certainly was improved for Baltimore city and their ability to intelligently
route a patient from maybe the intended facility of that medic to maybe an
equally appropriate and slightly further facility.

So, that said, we have that. We also have a program called EMASE. When we
talked about syndromic surveillance, biosurveillance, half the state right now
is on an electronic real time patient care reporting system. The medic comes
in, hands off the patient to the ED, they fill out an online web-based report
at the hospital.

Those data sit at MIMS. They are available to the jurisdiction and also to
the hospital, but that PCR Is coming in real time for literally half the state.

So, at any time we can see ramp ups in certain types of calls. Those types
of real time centralized data do offer certain possibilities in this arena.

Lastly, what John can speak to more, is this program we have called FRED
which deals with the other side of the house, which is what resources are
available within the health care environment.

JOHN: Let me just start that one of the first things we did when we started
to talk about FRED, which is the facilities resource emergency data base, is
that we didn’t want to impose any more entry of data or burden on the folks who
had to take care of the patients.

So, we do have this place called EMRC. Basically, during a disaster — this
is an old presentation, so I probably am going to be jumping around on it —
during a disaster, because EMRC was the everyday patient, the patient o has to
talk to the hospital, then they call EMRC and then they get connected to the
hospital.

So, they talk to the ambulances every day. So, when the ambulance gets
stressed, who do they talk to? They talk to EMRC. We have got 60 patients,
pleased call all the hospitals and let us know what they can take.

So, basically you see a list of things that they might, in any disaster,
might be collecting information about.

So, basically, one of the things that we had to do was link up all of our
partners into a system that could get information when a disaster was going on,
when an incident was ongoing.

So, we came up with this FRED. FRED basically is a centralized web-based
system that we can communicate with folks.

Normally, if you are monitoring FRED on a regular basis, it does page and
text message folks, but if you are monitoring, this is the monitoring page, and
I am just going to walk through quickly and make a couple of extra points.

When an incident occurs, our operators, or one of our field operation
support teams that actually go out to the disasters to help facilitate
information, will go ahead and begin to enter data about what is going on.

This is our train wreck, which is one of like the last straws that we had
underneath Baltimore city. It was a fire and we had all these hazardous
materials.

So, basically, this is what I kind of scribbled on my yellow piece of
paper, faxed to EMRC and faxed out to all the hospitals. That was pre-FRED.

So, we had all these hazardous materials. Once they enter the data, then
they push a button, it goes ding, ding, ding, ding, and all the hospital pagers
go off, text messaging devices, whatever you have. You get that information and
you get what was plugged in, in the beginning.

DR. BOENNING: How did you know that the train was carrying all those
hazardous materials.

JOHN: It took about four hours trying to get information from CSX and
everybody else. It sounds very simple. Actually, it took about four hours from
the time I found out about the incident to the time that we got the information
to all the hospitals.

That was like one of the last straws. That was unacceptable, especially
since I had one hospital that was in the smoke. I had one hospital that was
right on Howard Street and that was the tunnel.

Once we send that alert out, there is space on FRED for them to enter —
again, hand entry — enter in what they have available.

This is what I call dynamic information. You also have static resources,
like everything that is parked in the firehouse is on our web EOC, which is the
emergency management application.

So, that is everything that is parked in the firehouse but, just because it
is parked in the fire house doesn’t mean that it is available to go an
incident. So, that is the static, is what is on web EOC.

The dynamic is what do you have I can use right now. So, here you see
medications. On FRED we also have — you can go through. Once they enter that
information, it gets plugged into a central web site or a central spread sheet.

Here are all the different types of resources that you can get, and we have
actually expanded on this. It is beds, medications. You see NDMS.

We do our monthly or quarterly NDMS down through FRED, as does Delaware.
Delaware and Pennsylvania are also using FRED. So, beds, equipment, et cetera,
can all be catalogued on FRED.

We get in touch with lots of folks. Mostly it is hospital and public health
folks who are using FRED. However, everybody else still has access to it. So,
if we told them, you need to go to FRED now and enter this data, it could be
done fairly easily.

We can send out stuff basically geographically or by discipline, and that
is just to different groups, and we also control what folks see. So, there are
lots of security issues that we can throw in there.

Again, that is the static and the dynamic sorts of things on FRED. We have
talked about the CHATS. The CHATS is kind of the daily, the hospitals
available. FRED is, there is a big disaster. I am just going through the types
of systems.

Then you have web EOC, which helps you communicate the situational
awareness and can actually help to assign some resources and that sort of
thing.

The next step that I think has been talked about has been patient tracking,
and that is another issue that needs to be dealt with globally.

the last thing that I could say is the practitioners’ availability, is
another issue that really hasn’t been dealt with very appropriately.

DR. AUF DER HEIDE: Your patient tracking, can you explain — is this a
prospective system for determining where the patients go?

JOHN: Right now, I was just kind of, as I was finishing up on FRED, trying
to hit on some of the different types of systems that we have a need for.

Patient tracking, at this point — and I will probably get shot by some
vendors, but I have seen no off-the-shelf patient tracking program at this
point that does what I need it to do on the street.

I need to know what beds I have, what I have used, where the patients have
gone. So, there is really no off the shelf system right now.

We are working on a patient tracking project in the national capital region
which has become kind of a little bit stalled at this point, but it is moving
forward.

We are learning lots of stuff from it and, starting on Friday, I am
embarking on the same thing for the state of Maryland, to come up with a
statewide patient tracking system.

It can’t be done by one jurisdiction, which we are finding out in the
national capital region, that it has got to be kind of a statewide sort of
thing.

I am looking at Dan and I am looking at Beverly, because whatever I do in
the state of Maryland has to merge with them, not just them, but with West
Virginia and Pennsylvania and Delaware and NDMS, because we have done a lot of
work with them on patient reception and preparing for that.

All of those systems, although we might have six or seven different
vendors, it is going to come down to the standards of what data fields we are
saving and what we are being able to transmit. So, there needs to be a lot of
work on cataloguing what those data points are.

DR. AUF DER HEIDE: I think another thing that you might want to clarify,
there are really two kinds of patient tracking.

If you recall, back in the day when we were using START triage tags and you
have a unique number on each triage tag and you would tear those off and then
you would centralize all that information, that kind of system doesn’t tend to
work because most patients are transported by private vehicle or other
non-ambulance conveyances.

So, any kind of patient tracking system that you develop really has got to
be retrospective, and that is going to the institutions that have received the
patients and collecting the data on that end.

JOHN: I have to disagree with you on that respectfully.

DR. VIGILANTE: We have to wrap this up.

JOHN: Just real quick, any patient tracking system that you have has to be
able to be used by all of the disciplines.

In other words, if there is a declared disaster and we have communicated,
for example, across through FRED that there is a disaster, then those hospitals
need to be prepared to identify the patients coming in from that disaster,
potentially even use the pre-hospital triage tag and begin to identify and
track those patients immediately.

If you don’t track those patients immediately from the beginning, the first
encounter with a health care resource, you will not be able to find all those
patients. I am still looking for patients from our AMTRAK incidence in the
1980s.

DR. VIGILANTE: I am sorry, I am just very conscious of time. It is getting
near dinnertime. It is cold. Great presentation. Thank you very much. I think
it is a great resource that we can come back to.

JOHN: If you would like, we could host something a little bit further north
in Baltimore and show you what we have gotten. We are not perfect by any means.
We still have a long ways to go, but we are more than willing to share.

DR. VIGILANTE: Thanks so much. Actually, I am going to move right to Sally.
I would just like to hear your thoughts, Sally. I know you were going to
provide some comments at this point in the presentation.

Agenda Item: Discussant: Sally Phillips.

DR. PHILLIPS: We all know I talk fast and I am usually short. So, no
problem. I think first of all the sort of key point that just got made that has
been made throughout the time is that a lot of work, a lot of activity, a lot
of money have already been invested in tons and tons of data sophisticated, and
some really sophisticated ones.

The HERD(?) system, FRED, ESENS, ISIN, are just a couple that have been
named today. I think one of the things that we are hooked on, and I think we
are all kind of thinking in the same direction, as Dan mentioned, is no matter
what these systems are, everybody wants their system. They have their system
now, they are comfortable with it, they are developing it out.

What needs to happen is to make sure that we are starting to work from some
standard, so that there is the ability to talk within these data systems, and
to retrieve data out of these data systems, that isn’t labor intensive, that
doesn’t create something new, that the interface is there that allows these
systems to share data and extract data.

If it is a manual data entry system at the hospital point — and we all
know that it is — then let’s have it entered once into the system and then let
these other systems be able to talk to it and pull data from that.

Certainly the way we approach the bed availability, coming out with those
standards, we do have a patient tracking project we have been doing for the
department that next month will come out, and there has been a group working on
that with some of this data, standard language, and that will hopefully help
all the systems start to be able to talk to each other again.

I think it behooves us all, as we look at these data systems, to approach
it from that direction, to say, we already know there are vendors out there, we
already know they are home grown.

We already know people are well down the pipe for that. Are there some ways
that we can tweak all those systems to be able to exchange data, draw real time
or static data into a common pool.

I know there are pieces that we need to think about. As we look at an
approach of some data, I have heard a lot of discussion around using this data
for real time decision making, how to provide care, how to get people to care,
how to transport, whether it is the day to day emergency system, Art, or
whether it is the emergency system of these catastrophic events we are talking
about.

We are talking about using these systems to help allocate resources to the
places to get the best care to the most people.

The piece I haven’t heard too much about is making sure that this data is
real time available and build a pool of data that is available for research.

There would be an investment in building some of the research, the
evidence, that we need to compile that data into better responses, better
allocations, better day to day decisions around emergency care.

Our pool and our data base for that is very shallow, and I would hope any
data decisions that this group would come out for recommendations, you would
also be talking about it not being just for use during the disaster, but as far
as helping us do these disaster responses better through an evidence base.

The last point that I wanted to make was, during the disaster, Jim Scanlon
and I spent hours and hours and hours collating and pulling data. So, this
project is near and dear to my heart.

We needed to know really good information at the time of the disaster. What
did that system look like before it started, minimum data about — someone
mentioned earlier — just the demographics.

Was this an intensive area for dialysis or diabetic patients. You know,
what is the health care system going to be impacted, and to what level and
scope.

That minimum data base collection, there are some pieces out there of it
that we have. We need to make sure we know where it is, but we also need to
build that base so that, in every community, we can touch that and tap that
when the disaster to happens, to know when we move to the next level.

The next point was, you know, as these states manage their acute
situations, we have got to be able to look at how large different areas, in
managing those events, get the data that they need to make the decisions, to
move the patients.

As Mike was talking about, using JCTA or a tracking system to know where
people have gone, how they have managed the event.

Again, look at that data after the fact, build on lessons learned so that
our evidence base of disaster response is better.

The last part is looking at that recovery effort and how that data needs to
help us build the recovery. I remember sitting in a room like 30 days into the
event and someone saying, did we know what New Orleans looked like health
care-wise before this happened? How do we rebuild it back to where it was.

Someone was saying in the room, we don’t know. Well, we did know and there
was data, but knowing where that data is and what is there, and then how do we
anticipate, as you mentioned earlier, the after shock, the deferred care that
is going to occur, not only the exhausted system, the resources that have run
out, but the time, the temperature, the people, the resources that are going to
have to ramp right back up again for those deferred access to care.

We need a better data system that retrieves what that system looked like
before the event, how it was impacted and anticipating those after shocks.

It doesn’t take rocket science, we all know it, but as far as we are going
to build data systems and know what kind of data we want to collect, that is
what we need to make some of those decisions, so that, when we can decide when
it is time to move back to normal, back to a better standard of care, back to
regular staffing, back to continuing services, do we know what back means and
how do we get back to, and then how do we define and measure what that steady
state is.

Somebody said earlier that they challenged Art with what is the metric for
that. That is the question that I ask every day and people ask me every day and
I think we are going to have to really grapple with some of those metrics.
Those are the aspects, I think, of the kind of data system we are talking
about.

DR. VIGILANTE: Sally, that actually is, I think, a great summary of the
problem that is before us. I am just going to turn to my colleagues here on the
committee from the NCVHS.

We have to sort of take this in, what we have heard today, and process this
with the perspective of the mission of the NCVHS, what is within its scope of
mission responsibility.

I think that certainly finding ways to collect data prospectively so that
we understand where our vulnerabilities are, B, to collect data as you said
from a research perspective, to understand in an area in which the evidence
base, frankly, for best practices is fairly thin, if it is based on data, to
sort of enrich that environment.

Then also, real time data to sort of guide decisions intra event. I think
those are all three different kinds of categories of data, data collection and
modalities of it, being mindful all of the time of the burden that is placed on
those who must report the data and, to the extent possible, harmonize the
requests in a way that reduces the burden going forward.

I would invite other comments from Bill or Jim or Doug or anyone else in
the room as we think about and contemplate next steps. For folks in the room,
we would love to have your advice.

Agenda Item: General Discussion and Plans for the
Future.

DR. W. SCANLON: I agree with you on those dimensions. I went from early in
the session of taking notes and thinking about what we could drill down on in
our discussions, to thinking, thank goodness that we are going to have a
transcript that we are going to be able to try and digest over sort of months,
because there are so many new dimensions that were being sort of introduced.

We thought that surge capacity was a multifaceted sort of beast before we
came in here. We thought sort of we needed to be concerned about how elastic it
was, not sort of how it is kind of a fixed concept.

You all introduced dimensions we never might have thought of. Who would
have thought we had to think about meals as part of surge capacity.

I think we need to work, sort of in terms of what our next steps are, work
through how we best sort of approach the critical dimensions, how we sort of
can put together more information gathering on our part to identify some
strategies with respect to the different types of data needs that we have, that
are going to be reasonably feasible, sort of economical.

That, I think, are kind of the immediate steps in our work plan in terms of
this. You didn’t disabuse us in any way of the notion of how important this is.

I think of it, when we talk about — there are two types of needs, I think,
that are potentially satisfied from the providers perspective.

Sort of one is, can we design this in a way that there are operational
values through the exercise. I think there are hints here that that may be
feasible.

The other part — Jim, you are very sensitive to this about sort of the
pressure on the revenue side. I think of this as part of demonstrating sort of
hospital’s community benefit, which is something that I think is very much tied
to how much hospitals are paid and how we regard hospitals.

I therefore think that there is an indirect sort of positive benefit over
the longer term. IF we can develop sort of that area sort of more, we will have
both a better system and hospitals will be in a better situation because of it.

Also today you have told us that we can’t be thinking only of the hospital.
The demand management side is an interesting sort of aspect. Where do we stop
when we start to sort of explore that area in terms of thinking about what
capacity should a community have in terms of dealing with a disaster.

MR. J. SCANLON: I just — we had some very good discussions. I am always
impressed when we get smart people together in a room who know the subject, and
there is a lot to think about.

I think we have to sort this out sort of as an exploratory, but I do hope
we weren’t coming to the conclusion that there is no way to know in advance
what systematic readiness or capacity is, or to assess it on a systematic
basis.

I mean, if we just — again, I think we were sort of discouraged from the
survey approach. The table top and the exercise approach are fine, but it only
tells you about those particular institutions, as sort of a natural measure of
capacity.

I hope there is more optimism for sort of getting some regular readiness
information that is valid. Otherwise it is kind of, well, the plan is the first
casualty of any disaster, and we will know after the fact how good we were.

There must be some way of continuing improvement. We learn from past
experience. Isn’t there some way to measure along the line.

DR. VIGILANTE: I just wonder, in the AHIC community, you know, that the use
cases really focus on biosurveillance.

I just wonder if understanding is really on the demand side of the
equation. I just wonder if the notion that the supply side, in terms of — it
is just something to think about.

MR. J. SCANLON: They do have other activities in HHS dealing with health
information technology, to promoting and accelerating health information
technology to help all of health care.

In one of these cases, where they looked at, well, what sort of standards
are needed and what kinds of technologies is needed, electronic health records
and so on, there was one for emergency response.

It was more or less a first responder sort of a field system, which is not
quite what we are talking about here.

The other thing I think you made clear today is that these advances in
health information technology, to the extent that they are interoperable and
affordable, will help this dimension as we move along.

Hopefully it won’t have to take a manual tally — we are a long way from
that yet, but there does seem to be an underlying strategy for moving forward
on health information technology that would help some of these areas.

As you say, it is extremely complicated. In a way, every incident is
unique. So, we know afterwards whether you were prepared for it, but isn’t
there some way to think about this on a more prospective forward looking basis.

DR. AUF DER HEIDE: I would also emphasize the need for retrospective data
collection. Since disasters are unique in so many ways from daily emergencies,
sometimes the only way you can learn the lessons is to look how things worked
afterwards, and benefit from those lessons for the next disaster.

I think there are some good examples in the literature of how that has been
used successfully. So, let’s not rule that out.

DR. VIGILANTE: Thank you. Art, I think, is poised to make a dramatic
statement.

DR. KELLERMAN: Looking around the room at the caliber of the people and the
federal agencies, I would be remiss — first of all, I do want to emphasize how
pleased I am that this committee and that DHHS and various agencies are looking
at this issue. It is incredibly important.

As a member of the IOM committee that put three reports out last June on
the future of emergency care in the United States, of which disaster
preparedness is a very important piece, and we termed the system, at that
point, at the breaking point, I do want to emphasize that this is incredibly
important.

The information is not theoretical. This is not something we are modeling
for some future event. You could argue, as I think you made very eloquently in
talking about the District, we are in a mass casualty situation every day in
communities across the United States.

I talked about the Newsweek reporters that hung out with me. That story has
not run. They have other more pressing issues like girls gone wild and some of
the others that have made the cover, and Anna Nicole.

This story obviously did make a cover story some time back, which is why I
put this up here, and just to remind everybody, this issue appeared September
10, 2001.

DR. HANFLING: Just one other, I think, point of note, that is our trauma
center. That is one of our surgical residents.

DR. KELLERMAN: I plead with DHHS and Homeland Security and AHRQ and
everybody around the table, and non-governmental state and local and private
partners, we have got to fix this issue, and it has not gotten the attention it
needs.

I think you manage what your measure. So, your commitment to measuring this
issue is vital, if we are going to be able to really make progress, but I
wanted to compliment you all for giving your time and putting this on the
agenda on such a challenging day.

This is really a daily and nightly crisis that we have absolutely got to
come to terms with. Thanks for indulging me with this slide.

DR. VIGILANTE: I am going to let Doug say a word or two and let Jim have
the last word, but I do want to thank everybody for coming. This was a great
exchange of ideas.

NCVHS, I think, provides a nice forum, a non-judgmental forum, for this
kind of discussion, for people from different agencies to share ideas and
thoughts.

Of course, we are only three or four members of an 18-member committee, but
we will certainly sort of process this information and take it back to the
committee and figure out what our next steps are, and probably will be in touch
with you folks again.

I want to thank Doug and the staff of the NCVHS for really doing the heavy
lifting in setting this up. You all really did a great job and I really
appreciate all of your efforts.

DR. BOENNING: Thank you, Kevin, and Bill Scanlon to my right and you,
Kevin, to my left, we spent many hours on teleconferences and meetings getting
ready for this.

I appreciate everybody’s turning out today on a relatively bad weather day,
and I appreciate the patience of everybody on the telephone line.

I want to thank everybody for their excellent presentations and thoughtful
comments, provocative comments and interesting stories.

I want to leave people with one big question and that is what is next. What
is the next step for this group. I don’t think, in the remaining four minutes,
that we have adequate time to flesh that out, but feel free to send any
comments, thoughts, denouement, to me via e mail or give me a telephone call.

I do think that this is a pilot meeting, of sorts, and that we will have
further meetings and discussions, and I certainly hope that you can all remain
interested and participate. Thank you very much.

MR. J. SCANLON: I think I have said enough previously.

DR. VIGILANTE: Thank you very much. Drive home safety or metro home safely
in the wintry mix.

[Whereupon, at 5:30 p.m., the meeting was adjourned.]