It is a work that might
create a South Carolina

level number of companies and here it

was faster or teaches

his partners.

She's given them every

policy they

ever check back

in how they work.

What they've done very

well we keep this very informal today

that intro was a great lead in to
my first lives six and a half.

That's how long I've been here six and
a half years and

I high point now because almost nothing.

I'm going to talk about today
happened before there were in

a period of very rapid transformation
with respect to health I.T.

as an industry as a technology.

And the adoption of health.

So it's a very.

Interesting topic at this point in time.

In the interests of trying to
squeeze this talk into an hour.

I will spare the history Gail and

I were just talking about that
happened before two thousand and

four in two thousand and four something
quite remarkable happened the president

the United States at that time was
George W. Bush talked about computerizing

health records as a national priority
in the state of the Union address.

You think about it.

That's a pretty remarkable thing that
the president would highlight something

is geeky and
technical is computerized medical records.

In the state of the Union address and
then in April

of that year he created the Office of
the national coordinator for health I.T.

a new agency whose job it was to oversee
programs to actually make that happen.

For we get into the details
of all that in this era

of fragmented government in which we live.

We can hearken back to this house in
period decade ago when there was actually

cooperation at least around
this issue you recognize these

two folks that this is is.

Hillary's statement saying that
it's important that all health

care providers use.

Health Information Technology and

here's Newt Gingrich's statement
saying essentially the same thing.

Yeah from back then from yeah.

And then this is these are statements
from the Obama and McCain campaigns

in one way that I'm in two thousand and
eight or two thousand and seven I guess.

That's right two thousand and
seven both calling for health I.T. so.

This was a very nine character
verse feel nonpartisan issue and

to a large degree still is.

I mean I see I couldn't really find
very much on people in the political

spectrum saying we ought to defund
the office of the national coordinator we

ought to roll back and
not have health I.T. that the I'd

say right now the biggest political
issue which will elude to in

a little bit later is just the timing
with in general the Republican side.

Pushing for.

Taking a little longer to do.

So why is there such remarkable consensus
around this when there is doesn't

seem to be consensus around much of
anything else in this country these days.

Well this being the United States you
can imagine it there at the root.

There are are dollars.

There are many other reasons but
the root issue is money and

the root problem is expressed
here our health care system.

Really delivers a remarkably poor return
on the huge investment in making it.

We're currently spending about two point
eight trillion dollars a year in health

care which is roughly twice what the other
advanced industrialized nations.

Span and yet they outlive us.

I could spend a whole hour showing you
slides documenting how poor the U.S.

health care system is as compared to
health care systems around the world but

that really isn't the point of the talk
but this slide just intended to be

the first of a few showing how financial
issues really are at the root of these

federal policies got a huge audience I bet
I think we can squeeze you in Jamaica.

Another aspect of that is waste.

This is from the Iowa.

Publication I think this is from eyes.

To get which I am publication this is for
basically showing the details of a number

that's been floating around for as long as
I've been in this field which is the long

time that roughly a third of all U.S.
health care spending is a waste of money.

So not an unnecessary services being
a huge actually the biggest part of it.

This is where health I.T. really comes
into play excessive it ministration

cause people tend to forget
just how much more the U.S.

spends on administrative costs than
it needs to and so on and so forth.

And.

The rest of the bad news from a health
care economic point of view is our costs

have historically been rising faster
than those in these other countries.

So not only do we spend
twice as much as they do but

it's going to get worse over time
although in the last year or two.

Since the Accountable Care Act
went into effect and

nobody knows whether there's
a relationship there or

not there's been a noticeable slowdown
in the growth of U.S. health care costs.

I personally without any
evidence just from my feet

on the ground sense of things think
that the Affordable Care Act is actually

gotten the attention of the healthcare
industry and people are starting to pay

attention to efficiency but
that hasn't been proven yet.

So why did all this lead to so
the demand for health I.T..

Well let's let's drill down and
look at the health care system.

Bit deeper.

But first the cartoon.

You know we all know that a crazy
government driven health care

system would be a bureaucratic nightmare.

And fact this is a cartoon
that is drawing that is

prominent on right wing web sites showing
just how kind fan of really complicated.

The health care system we would have
if we actually followed Obama's advice.

Of course the brutal reality is we
already have a stupid bureaucratic health

care system.

So the thinking is just ludicrous.

It doesn't mean it isn't widespread but
it's ludicrous and

this diagram from I think this is
from one of Bill rouses publications

describing health care is
a complex adaptive system it and

attempts to define the nature of
the system in more engineering terms

with one of the key characteristics
being misaligned incentives.

So.

Is all the news bad now many of you
heard me speak before seen the sly.

We're very good at delivering
high technology care.

So if if you.

Do you decide to have a heart attack or
go.

Our Cancer or
get into a major accident your

odds of surviving these are our mortality
rates so lower numbers are good.

Are better here in the United States
than they are elsewhere in the world.

The problem of course is that
that's a part of health care and

it's really chronic disease
not the dramatic high

technology stuff that everybody
base intention to but

Monday in problems like diabetes and
hypertension congestive heart failure

chronic obstructive pulmonary disease
increasingly cancer which is sort of

become a chronic disease as we're able to
keep these patients alive for many years.

This is really what's driving health
care costs along with technology.

Those are the two major drivers.

And we're making the link to health I.T.

here just bear with me and you'll see and
in managing crime is these primary

care not not high technology
care the absolute converse

of high technology care primary care
the family doctor the general Inter.

As the pediatrician the O.B.-G.Y.N..

Are played a key role and I'm not I'm
not going to read slides to you but

there is good data from
the Commonwealth Fund showing.

That primary care where it's available.

Means people are more likely to
receive preventive treatments.

They're going to have fewer
preventable emergency room visits

fewer hospital admissions.

Better preventive near better management
chronic disease reduce mortality and.

There's actual data to show that where
there are is a greater availability of

primary care.

There are lower mortality rates and
I actually look for

but couldn't find a slide I have showing
there's actually lower health care costs.

There was a very good study published I
think in the New England Journal that

showed that they veil that a primary
care reduces health care costs.

So here's another problem we
have United States that you know

you don't hear about nearly as often.

Here we are over here.

Hungary Poland and Greece and the only
three countries reporting to the O.E.C.D.

that have a lower percentage of primary
care doctors in the United States.

Those advanced in dash realize nations
like France which I meant to point to and

forgot which spends half
as much as we do in lives

longer half the doctors are primary
care doctors in France here.

It's twelve percent.

So the whole system is a line around
the living this high technology

highly specialized care.

Which is exactly
the reverse of the kind of

system you need to manage chronic disease.

The reasons for this are or
shouldn't be surprising.

These are primary care doctors for

the most part they make a lot less money
a lot less money than specialists.

I was just at my medical
school reunion a few.

Weeks ago talking to medical students.

Many of them said they love to go into
primary care they find it very attractive

they know it's important.

But you know they're going to leave
medical school with exactly the same debt

as the students are headed
into these specialties.

How are they going to pay
it off at these salaries so

that's a problem we as a society need
to deal with in some policy related way

I personally believe there should be some
program that forgives medical school debt.

Maybe so much money a year for

every year that a student practices
primary care would be a smart investment.

So what is the result of all this.

Well you have this very highly
fragmented system where

care is mostly delivered by specialists
these patients with with chronic

diseases actually tend to have
multiple chronic diseases in

part because current diseases
cause other chronic diseases.

So this study this is
a stunning study shows

that Medicare beneficiaries with five or
more chronic conditions.

These are twenty percent of the Medicare
beneficiaries about eight million people

they drive half of all the Medicare costs
eight million people and here's why

because in the average year they're
seen by fourteen different doctors.

And.

The net result is this is the the care
network of the typical primary

care physician here in the United States
interacting with eighty six other doctors

for their multi chronic disease patients
and over two hundred doctors in overall.

And historically this is how
they've tried to do that now

from there's probably nobody in Georgia
Tech who doesn't understand that that

doesn't work that you're not going to
coordinate something this complicated.

Is in paper records that are can only
physically be in one place at one time.

That are probably illegible anyway.

And here's the results.

So I.

Very well study problems.

The more doctors a patient sees

the more likely they are to receive care
from doctors who don't have complete data.

In fact twenty percent of
the time in the eyes say

specialists see patients
with no record at all none.

They have nothing.

With respect to what's been going on
with this patient you think maybe that

relates to that big block where the.

Inside of medicine pointing to unnecessary
tests and procedures are being done yet.

It's a huge huge part of it.

That's actually been studied.

And mistakes occur.

So the more doctors and see patients the
more mistakes by that is these are cross

country if you notice the U.S.

is of a leader in incomplete data and
medical mistakes.

In large part because we don't have enough
primary care physicians so patients

are being treated by doctors who really
don't have an overview of that patient.

There are of course other problems

that I won't get into today we have
enormous health disparities here in

the United States your everything
from your access to health care

to your life expectancy depends on income
where you live how educated you are.

And there's there are many areas
of the country we in Georgia

have huge areas of the say where there
is no health care access even if.

No pediatricians know.

And patient safety is yet
another issue we.

In large part because we mishandle data so
effectively with this

fragmented system healthcare system
does a great job of killing people or.

Putting people on medications programs
that lead to complications and

hospitalizations.

I'm about a third of all hospitalizations
are due to misuse of drugs.

And the most common Misuse of Drugs
are patients being on too many drugs

drugs that overlap drugs that interfere
with each other and so on and so forth.

So all of this leads to
the essential roles of health I.T.

with and this is my slide and
this is sort of my ranking

of of importance at the very
top is coordinating care.

I'm and we're going to continue to have
a fragmented specially driven health care

system in this country for
a long long time.

There is no way the medical schools
couldn't produce enough primary care

physicians if the system changed the
students decided they want to go into it.

It's estimated about one hundred thousand
primary care physicians short in

this country and

it's only going to grow worse because most
of the primary care physicians are old.

Back to back train back when I was train
when we didn't have this high technology

system and
there wasn't this huge income disparity.

The second.

And I think from
a Georgia Tech point of view.

Very interesting.

Thank you.

The second and third from a very
interesting for me to actually second

third and fourth are very interesting from
a George actually they're all interesting

from a Georgia Tech perspective as
they all lead to research these grants

that are coming out of Man S S F for Smart
Health the opportunities are coming out

of and I they're all driven by this list
the second one is continuity of care.

This is making sure that all
the doctors who take care of a patient

have a complete picture of the patient.

But even if they do the care is still
what we call episodic the patients only

seen every so often.

What's going on in between
those visits to the office and

if things are not going the way they
should is that patient going to actually

end up in the emergency department or
the hospital before they have that next

visit to wherever they're scheduled to see
so solving that is called continuity of

care these terms are old I was working
in the early seventies and I don't.

Early family medicine apartment and our
chairman talked all the time back airport

nation of continuity of care as
the overriding goals of health care and

I was working on health I.T.
then and this is why we were doing it.

The third one is newer because we
didn't really have technology to

think about this back then we were
trying to do patient gave him and

his mind a technology called
the telephone we would call patients but

patients need to be more
involved in their own care.

Doesn't matter what
the health care system does

because chronic disease is
an everyday proposition if

patients don't do the things they
need to do they don't comply.

They don't manage their medications
right in their diet right in their

exercise right.

You're not going to get a good result.

So the furred and now with mobile and.

Please wonderful.

Technologies.

Of course a lot of people in
Georgia Tech recognizes the ability put

sensors in their patients home.

We It feels like we're entering an era
of of technology empowered patient.

Although we're certainly not there yet
and we're it's very early in this case.

And despite all the talk.

It's very hard to find research that yet
shows any particular benefit but

certainly over the long term.

A fourth.

Essential role of health I.T.

and fledgling is here is this is really
is areas clinical decision support.

I will show you a slide in a minute.

From that says that doctors make
the right decisions about half the time.

I tell my students in my seminar.

If you see a question on any quiz or exam
I give you that begins with what percent

of the time do fifty percent say always
going to be a good guess in healthcare.

So only about half the time to doctors
make the right clinical decision and

you know it's hopefully not
because they're stupid.

It may be because they have very
incomplete data about the patient.

But there's a significant.

Possibly they don't know what the right
the clinical decision is and that a cold

knowledge is is coming is increasing at
a rate that no one can keep up with it.

So it's been obvious for a long time
that computers ought to be tools for

clinical solutions support
an excellent quality reporting

this is a relatively new development
in the last ten or fifteen years.

And at the bottom of
the food chain here but

I think still very important and
interesting is the belief and

there's increasing evidence that it's more
than a belief it's a reality that we can

mine mountains of electronic health
records and actually discover new.

New knowledge.

Now it's actually been
shown that this stuff works

this is a study from the New England
Journal of Medicine which is the top

medical journal on than if you're
a researcher in a medical school and

you get published in
The New England Journal of Medicine.

Then you know you're sort of at the top of
the mountain can't doesn't get any better.

And this shows highly
statistically significant.

Pete values of point zero
a one point zero five.

That the achievement of standards for
diabetes care.

In other words doing the right things.

Is much higher at sites with the Amar's
than sites that are on paper.

And higher and actually not quite
as statistically significant but

still over five is is very respectable
statistical significance is outcome so

this is measuring something called process
all return to that in a minute and

most quality measures in health
care process space just the belief

that if you do the right things
you'll get a better result.

For diabetes and I'll explain this in
a minute that you can actually measure

outcomes and so both process and
outcome results are better.

It's a much better.

Yes.

Wow This study was done
in federally sponsored

safety net clinics which are clinics that

the federally qualified health centers
that take care of the poorest people so

that makes it even more impressive and
these letters tend have a char's.

Here in Georgia the safety net
practices have been more aggressively

putting any age hours than normal clinics.

I won't go through the details here but

it was also shown in something
called the physician group practice.

I know it's a politician demo but

what it would turn is this was
the predecessor to the new outcome based

instead of payment programs that
are in the Affordable Care Act.

Has done it ten high performing
health care sites around the country.

And these people these sites were put
under reimbursement scheme where if

they got better.

Quality metrics and

were able to reduce costs they would
get a share of that cost savings.

One of the sites the Marshfield Clinic
got like half of the the money.

So ten sites one side got half the money
only three or four got any money at all.

And these are the Marshfield clinics
explanations of how they did it and

Riddle throughout it is their
advanced use of health I.T..

This is none of this is new.

To my knowledge the first person to
complain that health data was not.

Being maintained in a way that
made it useful and available

more useful was Florence Nightingale back
in one thousand nine hundred sixty three.

She was far more than a nurse.

She's very interesting woman.

It's interesting to read about her shoes
as you say she was a social reformer

statistician and
the font founder of modern

nursing a renaissance woman
in one thousand sixty four.

This guy Dr Larry weed.

Who's one nine hundred sixty
nine book actually launched me

on my career when one of my prefer.

Cers gave it to me to read see given your
students' books might be a good idea.

Occasionally while I family
never give them back.

So I'm reluctant to confirm that I have
a copy of this of his sixty nine book and

it's hard to find and
I lent it to a student never got it.

I wish I could remember
who I wanted to but

Larry we wrote he's still around but

is getting up there in age we still around
wrote very directly about the need for

medical data to be recorded properly so
that could be used.

To provide adequate care and
also contribute to medical knowledge.

In one nine hundred seventy seven.

The Office of Technology Assessment
which doesn't exist anymore.

Studied the landscape of existing
electronic health records.

There were four.

I'm happy to say I was one of them.

And determined all the things I've been
saying that they get him to increase

efficiency monitoring quality here
a little little eliminate duplications.

And be supplied data for research.

But nothing really happened until two
thousand and one gal I don't know.

Did I mess it up if I think this is
one process of a quality gas and

came out but maybe I missed it.

In a landmark publication
that really was at the root

of everything that's happened since the
Institute of Medicine published a book.

All crossing the quality chasm the new
health care system for the twenty first

century it was the time and one of the and
everybody remembers this book because

they said that forty eight to ninety four
thousand Americans are killed every year

for completely preventable reasons and
US health care in US hospitals and

if you read the details it's almost
all has to do with misuse of data.

Made the same and

then Information Technology is is central
to fixing the health care system.

This led directly to Bush's State
of the Union address this.

This is this is the.

The root cause.

So in two thousand and
four Bush said we need to have universal

adoption of electronic health records
by two thousand and fourteen.

But in two thousand and

eight in two thousand and nine a twin
pair of studies published again in

the New England Journal of Medicine
said that if you're looking for

use of an electronic record that actually
has the potential to improve quality.

Four percent of physicians had one and.

I even I was surprised the next year that
hospitals were only at one AM percent.

So nothing much had happened
since two thousand and four.

And the fundamental reason why is
the Bush administration didn't

fund the program they are I knew
the guy they hired to run the O. and C.

and they they talked about using
market forces to make this happen.

The problem is the market forces were
the other direction the way things

were constituted always been
constituted the physicians or

the hospital had paid for
them like trying record.

You know through the very
real pain of implementing it.

And yet

the financial benefits largely went to the
insurance companies and the government.

If it were a financial matter.

So Bush set the goal.

The Obama administration created
incentives that both direct and

indirect incentives that erect in
the Senate as it were some twenty to

thirty billion dollars They've currently
actually paid about twenty two billion.

That would actually pay.

Physicians and hospitals if they
implemented electronic health

records would get into the details of
that you know as well as we move ahead

that's really high tech and
the indirect incentives or

the introduction of outcome based payments
in the Medicare program the accountable

care organizations and those are starting
to happen although we got a few

years to go before that really bites so
those were the goals the strategy

is really in three programs E.H.R.
certification and meaningful use.

And Medicare Medicaid and sent.

Program C.H.R. sort of a huge occasion
is really the domain of the office of

the national coordinator it's a technical
program which I'll quickly describe.

Meaningful Use is kind of a joint thing
and the incentive payments are all

all handled by C.M.S.
because they run Medicare Medicaid.

And then the law that defines
all this is this high tech.

It's part of the stimulus program.

Before I move on to the details
I would give you a quick Georgia

snapshot one of the things that high tech
guys establish regional extension center

programs to help physicians in
small primary care practices these

are pretty much all rural physicians
to give them extra special help to

implement these systems and if we had
time I go on the date of the details

of the New England Journal study them and
several other studies that show that this

is the group that has the most difficulty
implementing H R's it's not surprising.

There are small practices
there in rural areas.

There's nobody to help them.

So the Government's done it and we've
actually done quite well here in Georgia.

There are about four thousand
primary care physicians who

are the target of this program.

Ninety two percent are now
live only a Sherry's.

Fifty three percent have
achieved stage one of

meaningful use I'll describe
what that is in a minute.

They've earned thirty four million
dollars in incentive payments and

here's a case where George
is actually in the top ten.

You don't find that very often.

Unless it's something you don't
want to be in the top ten.

And Georgia Tech has been the largest and

the most successful
subcontractor of this program

the program is actually administered
by Morehouse School of Medicine.

But we've really through our
group in the square been

the major player in making this happen
something Georgia Tech should be proud of.

So what is E.H.R. sort of education.

Well it's a sad a rip.

Pirates placed on the commercial age or
systems.

So.

To qualify for meaningful use
incentive payments so our physician.

Has to buy a certified the A more.

In order to qualify for
meaningful use an incentive payments.

And the standards are in three domains has

the is the system recording certain key
demographic and clinical data elements.

Does it provide tools to measure
improve care quality and

of course to protect the confidentiality
confidentiality integrity and

availability this being key this is them.

This is a this well but
this is HIPAA this is a stand in for

the the age old bugaboo of health I.T.
interrupt ability.

It's the data.

It's one thing to
actually record the data.

Can you get it out.

That's hard.

And you know I can go on with
slides like this all day long.

This is just to give you a flavor so

the vendor is given a test
set of of diagnoses.

This is the real stuff
right from the manual.

They have to supply some system for
specifying the status of each problem.

There is no standard for that and.

They have to have a way of specifying
the date when it's diagnosed and then

the task is typically to change the status
and one of the problems and then look in

all the places in the a charge to see if
that data has been appropriately updated.

So it's a whole series of tests like that
there are four or five companies that have

been set up to be the testing centers and
there are now

several hundred certified E.H.R.'s the
United States that itself is a problem.

We're going to come back to that problem.

Later on if I don't remember a time.

The other key issue of
certification is quality reporting.

Can the system report on or
on certain process and outcome measures.

So a process.

Measure is basically was
the right thing done.

And an outcome measure is did you
receive achieve the desired result.

There are way more process
measures in healthcare than there

are outcome measures.

The poster child for
both is the name of woman A one C. test.

Member when I showed you that New England
Journal study that said both process and

outcomes for advice diabetes were
approved in the safety net practices

that had a charge this is
what they were looking at.

So this is diabetics blood new clothes
bounces around based on what they

actually are size they get whether they
actually take their medications or not.

And what you'd like to know is that
the blood glucose in the patient

shows up at the office after not being
seen for sixty days which you'd like to

know is how well control has a blood
glucose van over the last sixty days.

Well it so
happens that there is a molecule

called in the red blood cells hemoglobin
is found in the red blood cells.

And.

The higher the average blood glucose
like a miraculous thing the higher

the average blood glucose level is over
the last couple of months or so the higher

the hemoglobin A one C. variant of
hemoglobin level is in the cells because.

Higher concentrations of glucose drive the
conversion of hemoglobin A one day one c.

It's all you have to do is measure they
one C. level and effectively you've

got a measure of the average blood
glucose over the last sixty days.

It's hard to believe something
this neat exists but

it doesn't wasn't known when
I was training and seven.

Well seven is the Mayo Clinic's.

Threshold many people use nine but
doesn't know me and so the process.

S. measure is was this actually done.

It's generally recommended diabetic
patients should have an A one C.

level done every year.

What percent of diabetic patients
actually have that not every year.

And I already told you the answer
is fifty percent that's right.

They go fifty percent.

So the process measure is was it
actually done in the outcome measure is

what percent of the dive eight patients
in our practice actually have an A one C.

level at or below seven percent.

So the E.H.R. has to be a report on this.

This is what quality measurements
about it's not rocket science.

So meaningful use is being
implemented in three stages.

Stage two is starting this year.

Stage three was just pushed back
a year to two thousand and seventeen.

Due to political pressure but

actually probably a reasonable
thing to do because this is.

And I'm not going to
read all this to you but

Stage one is really about data capturing
in sharing is this as I showed you

is is the system and the doctor actually
capturing the key data are they.

Measuring reporting on
current clinical quality or

are they using that information to
engage patients Stage two is really more

focused on exchange of data the prescribe
being electronic transmission of summaries

more patient role data and stage three
which it's not clear what stage three

is going to be in theory the focus is on
actually improving outcomes so you go all

the way from capturing ensuring the day to
turn getting improved clinical results.

The way stage one works

is there are measures that have to be
reported and they divide into three

groups I'm not going to get in the way
you can see that clinical quality.

Physicians can choose three of forty one.

So it's a light touch actually
reporting on when to call quality.

Here's examples.

Transmit at least forty percent
of prescriptions electronically

a prescribed how many of you have
a doctor who we prescribe my nose.

The data is overwhelming
on the benefits of the.

Prescribing.

And surprisingly people are actually
more likely to take the medication

if it's the prescribed.

Ten percent more likely probably
because it's so convenient for

the patient they don't have to do
anything and they probably feel guilty.

The knowing it's at the pharmacy so
the better go pick it up.

Engaging patients and families.

So this is a very light stage wind.

They have to be able to ride patients with
a clinical summary of their visit within

three days have to do that for
more than half of their office.

Have to perform show a cast perform a test

to show they could care
coordinate their rights.

We actually have to do it and
then of course privacy and security.

Stage two is basically the same stuff
it's just that the goals are raised and

they're more sophisticated so
I'll give you a couple quick examples.

One area that both are focused
on is smoking status.

So here more than fifty percent.

Of the patients have their smoking
status recorded partly an arduous task.

But in stage two it has
to be eighty percent.

And as we recorded a structured day or

night not just for
you text but structured.

For scribing forty percent in
stage one as I just showed you

more than fifty percent in stage two these
are really not very arduous requirements.

And in this one they
actually have to compare

the prescriptions to a form drug for
me or am I.

We're moving into an area
Accountable Care Act has

drug formularies you're going to read all
kinds of static and probably already have

in the newspapers about this patient
complaining reporter that their drug

isn't covered under the formulary
this is an attempt to manage costs.

Pure and simple.

Do you have a question Dr.

So what will stage three be we
really don't know yet but basically.

Here's some quotes from the latest C.M.S..

Posting the important thing is that under.

Stage three for the first time physicians
who are not meaningful users will have

their Medicare payments reduced
that was in the act all along.

It's just out to you and
the Secretary of State of Health and

Human Services can actually
increase that percent is if he or

she chooses to under the law but
one percent is the big

money in a relatively low margin industry
for a hospital that's a big number.

And all you can read the rest
of the details later but

the this is the current schedule time.

Is all this really necessary.

Well here's a quote from the American
Academy of Family Physicians the largest

group of primary care physicians a country
of the American Academy actually

helped write this law they
are strongly in favor of it but and

they're saying that their doctors only
achieved Standards of Care About fifty

percent of the time so I kept telling
you it's safe answer to any whispers.

So what have they been results what are
the results well indeed there have been

first that physician in terms that I
find this weird that this is a C.D.C..

Study they define any H.R.
as a record this either all or

partially electronic It doesn't matter
what it does but it has to do more

than billing just a billing system doesn't
qualify they do say that a basic E.H.R.

does a list of things that kind of
tracks with meaningful uses I don't know

why they call it a basic H.R.
probably the reason is it doesn't include

clinical decisions support and
more many so in your mind you

should distinguish between a my
preferred term as a passive sure.

You can record all.

Stuff.

It can report on quality that sort of
stuff but it doesn't use the data to try

to improve the quality of care to affect
the physicians clinical decision making.

Whereas an active E.M.R. would do that and
there there are an O. a list of

functionalities that you can do in
an activity are whatever the deafened is.

Seen as you can see that the percent
of physicians with BASIC.

Charge which are a bit more
than basic is dramatically up

this program is actually work.

Now if if you are like me and you've been
in this field since the dawn of time and

I can't even the slide goes out of
the room back in the early seventies and

everybody's been talking about the need
to do this since the early seventy's and

the rationale was clearly stated
in the early seventies and

by you know these numbers don't
track perfectly journal paper.

There's a definitional difference but
it doesn't not very much.

It happened.

And why.

So you want to YOU WANT TO example of
federal policy that's work here it is.

And of course there are differences
around the country by state.

I am not surprised that Washington and
Oregon are very high.

I'm not surprised that Massachusetts
is maybe a mile is extremely high.

Why North Dakota so I probably somehow
system in a place like North Dakota one

health system can and then Georgia's
you know what we could be worse.

You know we're not West Virginia or
Oklahoma Nevada.

Harry Reid should pay attention to his.

And you can go to this site.

This is really a fun site of
yours in those days for dot.

And they've got all kinds of
interactive stuff you can do.

I love playing with this site.

So here's an interactive map of
the country you can hover over George and

see that we're at twenty percent.

And you can actually get it.

This is the.

Prescribing you can get it by county and

you can see the predictable areas of
the state where it's zero to ten percent.

So surprisingly enough despite stuff
that you may have read the C.D.C.

is also surveyed surveyed doctors
about readiness for stage two.

And this is one of the reasons why.

Things have been slowed down a little but
only about thirty percent.

Sort of aren't committed
seventy percent of doctors say.

We're going to do.
Stage two.

Only thirteen percent already though.

What is stage two more specifically well
are a list of the objectives in stage two.

But keep in mind say Sue just beginning
stage one is the current reality.

This one is the one that
we're very excited about and

I was very surprised that
physicians of us provide

patients the ability of you download
transmit their health information.

Remember in the stage one it was to share
clinical summaries of the patients that

could be a print out probably was.

Here they're saying they
must do this electronic.

And there are some percentage of this in
stage two five percent of the patients

who were seen in these reporting
period must have actually done this.

So it isn't just provided
the patients have to use it now.

What's the rationale for

that where there are lots of studies to
show that if doctors actually encourage

patients to do this then like
magic patients will in fact do.

So we're beginning to set the stage for
patients to gauge when.

If you look at these criteria and
some detail you

can go again to the days for diet health
I.T. site these their evidence base this

is the number of clinical studies this is
the percent that are either positive or

mix positive supporting each of the things
this call for in stage two so there.

Just making this stuff up.

And anyone who's studied chronic
disease would would like to list.

Here's the percentage
of cording the C.D.C.

of physicians who currently have the
capability doesn't mean they're doing it

but have the capability
of doing various key.

Factors that are required in stage two.

And again this may surprise a lot of
people and it actually surprises me.

Given what we're going
to talk about a minute.

Only fifteen percent of
doctors in the C.D.C.

survey are either very dissatisfied or
somewhat dissatisfied with their E.M.R..

I'm not sure I actually
believe these numbers but

that's what they are these
are the typical position concerns

it decreases productivity and
hours it takes more time.

It introduces new sources of error.

And maybe because I fail to notice
something or something is recorded wrong.

And a loss of documentation detail this no
words if your particular recording date in

a very highly structured way the nuances
that that you can express in a text no

don't get recorded and only as well
historically there are studies.

To support these views but
this just came out.

It was done.
Late last year.

Iran and A.M.A.
American Medical Association study.

Showed a very good strong
percentage of physicians saying

that he H R's improve the quality of care
and if given the choice only eighteen

percent of physicians on a life trying
health record would go back to paper.

So there is good news.

It's mixed.

But there's good news.

And I would argue that these statistics
really don't tell the whole story.

Now we're sort of moving to
the Bronstein view of the world here but

I'm going to back it up with day.

Here the Domino.

Vendors.

Unsurprisingly epic is at the top
all scripts a clinical work snecked.

And G.E. Healthcare server.

These are the these are in
general the companies that supply

to the health care enterprises.

But if you drill them.

And you look at practice size these
lighter colors or small practices and

these darker colors are larger
practices this tends to be the domain

of the hospital systems and this is
more of the independent practitioner you

see at a striking difference
in in which E.M.R.

is are installed with
amazing charts E M D's and

practices being the most common
choices of small medical practices.

So what are these and here.

Here's the data about what physicians
which systems they actually like and

you see the same systems.

Well that shouldn't surprise you
in a small practice of the doctors

are choosing for themselves.

They're going to pick.

The system they like but here's they
go look at this this is striking.

This is when they're in the top four.

Epic you know and
the others are kind of mediocre and

some of these systems are pretty awful
these what do these four have in common.

Well if you take my mova read my book or.

You'll find out that these are the ones
who are a great deal of attention and

so the innovative things have been
done in the user interface in

the collection of data.

Practices for example uses
machine learning to to learn over

time how physicians treat each kind
of patient and it can actually

anticipate what they're going to document
and make it much more efficient.

But they also have very simple
easy to understand you deserve.

They were all pretty might and
I said if I believe is correct.

All four of these are companies
headed by physicians I'm.

Not saying that's necessarily
good it's interesting.

They put a lot of effort.

In and then here.

Here is the interesting thing in going
back to this chart of the most widely

install systems green means I'm sorry
the yellow here means it was sort of in

the middle range of the position
rankings red means it was in the bottom.

And green means it was in the top.

This is physician ranking the systems and

you can see that the widely installed
systems don't get good physician rankings.

And.

In fact two of the top ones
don't even make the list.

So in my view of the world we have
systems being selected because

the vendors are able to sell
at the enterprise level but

don't necessarily have very good products.

I'm going to skip that one.

I'm almost done.

So there are certain clear
opportunities one is better E.M.R.

as this is an example of
one modernizing medicine.

David crater from Harvard and
him said you've got to go see the system.

And it's nice cool web based
runs on an i Pad highly graphic

they use some variation on machine
learning to learn over time how

physicians practice I don't think it says
that this ticket is what praxis is doing.

Young new entrants into the field.

You know what they're going to get
crushed in the enterprise marketplace.

That's something changes.

More facile exchange of data.

This is direct this was introduced
about four years ago so and

it's gotten some real traction.

Instead of an expensive complicated
centralized engine for exchanging

information why can't doctors just encrypt
it attach it to secure email and send it

to each other with a system calling called
hist handling the technical details.

I won't go into the technical details but

this essentially makes health
information exchange for.

No.

Lighter technologies here
in the last year or two.

You're seeing.

A lot of interest in using restful A.P.I.

is another very civil techniques that
are used everywhere else in the world.

To exchange data and fire is the.

I don't I didn't fire is the term
under which all this stuff.

Folds.

With a lot of talk about fire at this
year's hands meeting is it basically means

that if you need to get data from an E M
R You can just do an arrestable query and

assuming you know you have
them the right to get to it.

It's easy to do simple to do.

There are already companies being built
based on this if you're interested go to

human A.P.I.
SEO a very interesting company

on the West Coast that's aggregating data
from patients at home from devices and

sensors and
making it available using Russell A.P.I..

Advanced query.

This is an O N C an issue of to
basically allow people to query

a Mars again using the web and
certain standards for query.

And so I think of my last line and
this is the latest on C N N issue.

Trying to come up with a simple
easy to implement schema for

clinical decisions support
the basic idea here would be.

That if Jim aying develops.

The world's best algorithm for
making a certain political decision.

People could using these standards sand
patient data to it via web services.

It would crunch and
its recommendations would come back.

This sort of gets out of
the trap of an E.M.R. that really

was never designed to do this stuff which
is what was true of most of just do it.

X. term or
use web services to gain access to it.

So that said I'll be happy
to answer any question.

My graduate students complain
about a lot of stuff.

It's our way to invite you with your
interests to go to the archives.

You know the BROWNBACK I gave
a talk last year from playing.

To personalize medicine that
actually this is what it's about

there's a very interesting study
out of University Indiana Ph D.

thesis where the researcher
worked with the largest

provider of behavioral
health in the country and

was able to develop a machine learning
algorithm that can treat diabetic he

does I'm sorry can treat depression
more cost effectively than doctors.

Better outcomes at lower cost.

And in fact the transoms be heading in
the opposite direction of what you suggest

that these technologies are allowing
patients get far more personalised care

and rather than giving a long answer
if you really should go watch that.

Talk.
Actually I'd love to talk to you

after you watch that.

Because I think there's.

There is a very interesting history there.

That's worth the people
understanding better.

Yes.

All

right.
I mean even my own view and

I'm on thin ice with this kind.

I think maybe we've gone too far in
the protecting paper patient privacy.

Space.

I mean you talk to any researcher
who's tried to get clinical data.

It's a nightmare.

I mean in theory we have a cooperative
friendly relationship with children's and

Raul Bissell and
I have a project it took us six months.

To get the data DIA down if I did.

It's not identifiable that and so

the paranoia levels around
this here at Georgia Tech are.

Serenely I wish Matt was still here.

And you talk to researchers around the
country and then everybody says the same

thing so I think the the barrier to
getting the data is significant enough and

I wonder if it isn't chasing off a lot
of interesting research opportunity.

Not that I'm in favor of just willy nilly.

You know forgetting about privacy and
security but I suspect if we as a society

got a bit more focused on making this
more facile we could make it more facile.

There are proposals by the way
one which I think is old I first

heard it in the seventies but I think
it's gaining some real traction now is.

Look let's just let patients deposit
their data in health information banks

and let them decide who has access to it.

In fact some people are saying and
why not if the data is being used for

purposes that have financial value
such as finding patients for

clinical drug tiles let patients share
in the financial return on their day.

Patients patients surveys show that
patients are inching contributor data for

medical research particular of course if
it's a problem that they've had or or

their family members may have patients
are also very concerned about privacy and

security though so
I mean it's it's a difficult problem.

You're going to talk to any family.

There has a cancer problem or

any other severe long term problem and
you'll discover that.

They tend to become the custody of
the record and take it with them

because they figured out how dysfunctional
the system is what we have a neighbor

down the road from us he
recently had a stroke.

It's done by well and the wife is

just can't stop talking about the number
of mistakes that would have been made

in the hospital I won't say which which
hospital but it does begin with an E..

Had she not been there and

effectively become the custody him of
his care how many wrong medications

morph towse is thinking he would
commit this if this reality from

your age

is why he

writes The whole idea for having
a system capture the revenue that you

think is to give you better care is
really to capture all the revenue with

your problem.

Well that's what that would be P. months
for you and in fact they're spending.

I don't know what the number is it's
a huge number on Epic to achieve

totally coordinate care.

Now the dirty little secret is that
none of these enterprise systems

were actually designed to
do that that they're all.

All relics of another era and
I'm a Piedmont patient myself and

I'm hoping that the next time I have to go
over there every day when I get referred

to a lab or a specialist I won't have to
tell them all about myself because they

don't have any clue who I am which
has been my historical experience and

I suspect happy epic will help quite a bit
with that but I would not as a patient.

Even when epix full fully
implemented pm I would not miss sume

that there is this seamless transfer of
all my data that you are here assume.

Right here.

Yes.

He or

were you living in Indianapolis or pretty
much anywhere in the state of Indiana

you actually could be quite confident
that no matter where you went.

Your health record would be available.

Now why is that not the national model.

Well Indiana is lucky they happen to have
a guy name rage in strife who invented

the front loading dishwasher and at one
time was making forty percent of the front

loading dishwashers in the world
got interested in health care for

his employees and eventually made the
connection to health information exchange

and is Foundation has funded the Indiana
health information exchange.

In a rational country.

We wouldn't be having this discussion.

I mean you talk to anyone from France or
in Israel or

the Scandinavian countries or Australia.

New Zealand.

I mean they don't I did the first one we
got our new Dean you know I met Zvi for

the first time and he couldn't understand
why what I was doing was even worth doing

because in Israel they have one ally
trying to record for the whole country.

That's of course politically
impossible here and.

In a sense it's a good thing that
no decision like that was made

because it would have probably
been a terrible electronic record.

I mean it's not that we have
enormous challenges but

we've made enormous progress.

Pending on what state you're
in it's impossible as you.

Well I wish you were only fifty.

I mean well we've all been
toward in most places.

Is a series of enterprise driven
health information exchange.

So here in Atlanta and we're one of
the worst markets in the country.

It looks like we're going
to have the Emory camp.

We're going to have
the Piedmont Well Stuart camp.

And I don't know that maybe another one or
two before it's over.

Each of whom is using
their electronic record.

As as a vehicle to try to tie together
all the care within their boundaries and

to convince you as a patient that
you should get all your care

within their armed camps.

I don't really think that's
the best solution but

I think that's what we're
going to end up with and

the role of health information Shayne's at
the regional and national statewide level.

Is going to be the interconnect
these ARM cams and

unless something substantial changes I
think that's what we're going to that way.

It is progress.

I mean you and I will probably get more
integrated care at Piedmont than we would

because it really achieve
the national jacket I don't think so.

And there is already concern being
expressed that it's going to drive up

costs because within if one of these
systems becomes really strong and

as has already happened
in other areas of Georgia

prices go up because
there's no competition.

Well I meant what the Affordable Care Act
should have done but

I mean it's politically impossible
is created a single payer system.

And we will and
I actually if you talk to many physicians

they're in favor of a single pairs
as they're having to deal with this

incredible administrative complexity and

a single payer system would have would
lead naturally I think to a more

integrated approach to health I.T.
that's affectively what Kaiser has or

any of these advanced cell systems have
in the data is quite clear that H.M.O.

deliver higher quality cost care
at lower cost and I it's should

be surprising I mean it's not like you
have to do anything really far out there

to reduce cost increase quality you just
have to make people give people a way to

talk to each other make sure everybody's
got the right data at the right place.

The right time.

That's that's all you have to do yes.

Yes

You know I don't think it's
going to have a huge impact on.

Stage two isn't that arduous either.

I think the big issue is just
doctors having difficulty doing

it when the time frames
that have been specified.

But once you've already taken the plunge
and you've you've implemented the M.R..

And you know the handwriting
on the wall that you're

going to happen a meaningful user or
your payments are going to go down and

you probably won't be able to be part of
the private insurance carrier networks.

I think you know they'll pump
the probably some decline but

I don't think it'll be dramatic.

Although States three remains to be
defined in theory stage three is going to

require clinical solution
support which would be

a wonderful thing and would be
a huge boon for academic research.

We'll see how this stage three isn't as
is already a political battle over it

although so far most of the arguments
are over not what we do but

how much time we give so I'm so I'm OK
with given the doctors of the earth too if

we don't lower the bar too much where the
bars are pretty low right now actually.

OK.