First of all,
let me thank
you for the chance to be here and to present some thoughts to you. As I
looked
at the list of keynote speakers - Steve Ballmer and Colin Powell and
Mike
Leavitt, among others - I was reminded that this is fast company, and
I'm
genuinely
honored to be invited here and sharing this stage with people of such
accomplishment.
I'm not an expert on eHealth matters. This room is full of men and
women who
are experts, who know vastly more about the field than I do, who deal
with
issues relating to it daily, who are much more conversant with the
latest
efforts and issues. I want to do two things in my presentation this
morning;
update you on the
I believe the reason I was invited to come here is that I co-chair the
State
Alliance for e-Health with Jim Douglas, the Governor of Vermont. This
project
has just begun - our first meeting was in January - and has been
organized
by the
NGA - the National Governor's Association - with the help and support
of
HHS. The
NGA is a bi-partisan organization - in this case a Democratic governor
co-chairs
the
The charge to Governor Douglas and me is to analyze major issues that
states
should consider as they engage in electronic health information
exchange. The
We are starting out with three Taskforces: One for Health Information
Protection, one considering aspects of Health Care Practice, and one
considering Communication and Data Exchange. We spent a good part of
our first
meeting listening to presentations from experts and working with our
Commission
to structure some specific initial charges to the three Taskforces.
The Privacy Taskforce is now charged with looking at the major state
health
privacy laws with an eye to how they affect the ability to achieve a
workable
sharing of information. It is then charged with making recommendations
as to
how to address such issues. It will work with the Health Information
Privacy
and Security Collaboration that 33 states and
The Practice Taskforce is charged with looking at licensure laws and
regulations and how they affect
eHealth, and to bring us an overview of the liability issues that may
arise. I
would add that these issues are important not only to interoperability,
but
also to a national response to a pandemic or major disaster.
The Communications and Data Exchange Taskforce is charged with looking
at large
state coverage programs - Medicaid, state employees, and state public
health - and
seeing how they might further facilitate and use health information
exchange.
We also asked them for a presentation on current state progress in the
creation
of exchange networks. A part of the reason that Jim and I were asked to
be the
initial chairs was the breadth and success of the work that is ongoing
in
The
This State Alliance has limited time and budget, and is not intended to
be the
last word on the subject by far. I do believe that if we can keep our
Commission and its taskforces focused in some well-defined areas we can
make
progress, and we are both committed to making a success of this effort.
As I said at the outset, I'm not an eHealth expert. Where I can claim
some
expertise is in the art of getting things done; getting the right
things done.
I've had the experience of starting a company as the sole employee and
growing
it to several thousand. And I've spent the last decade and a half in
the public
sector, trying to figure out how to actually make things happen under
that set
of rules.
I want to talk with you this morning about getting things done, and I
want to
challenge you to think about how we get things done in using and
exchanging
electronic health information in some new ways.
I'm going to challenge you to simplify.
This is important stuff. Information technology has the capacity to
vastly
improve the quality and control the cost of care in our nation. And yet
we
underused it terribly; health care is still a cottage industry in many
respects, and unless we bring that cottage industry into the modern
era - not in
its medical technology but in its organization - it is going to bite
us.
The frontier in modernizing is not technical, we have too many
technical tools,
not too few. The frontier is in how to actually make something happen
in the
vast sector of our economy that health care has become. There are
twenty
thousand very smart and well educated people at this conference in
Put plainly, my thesis this morning is, "Enough with the grants, enough
with
the conferences, enough with new paradigms, enough with the pilot
projects;
this is good stuff; how do we actually get health care professionals
and
organizations to use it?"
This shouldn't come as a surprise to anyone: the task of leadership is
almost
never problem solving, but rather finding common ground on which
diverse ideas
and interests can progress.
eHealth is important for many obvious reasons, but let me offer also
one less
obvious one. In this new century, we have as a nation what seem to me
to be a
whole new class of problems: eHealth, the health care system as a
whole, or to
take an entirely different sector, things like energy independence in a
global
economy. These are a new class of issues because they are of a
complexity and
an entanglement with the economy we haven't seen before. There are some
things
that government is good at; solving these new problems is not yet one
of them.
We must change that, though, because solving these problems needs the
standing
and strength of government to get the right things done.
The less obvious reason that eHealth is especially important is that it
is
likely one of the most tractable of these problems, and I believe that
in
progressing there we might point the way to tackling some of the even
more
difficult ones that await us.
Before we begin, let's step back for a moment and look at the argument
for
eHealth.
We have a very expensive health economy today; 16+% of GDP as we speak,
on its
way to 20% and more, still something like 50 million Americans who lack
insurance. The next highest nation, with universal coverage, is at I
believe
11% and growing more slowly. There are still some companies and some
unions and
some advocates who couch the problem with a straight face as simply one
of who
pays, but they are getting fewer and farther between. The cost of our
health
care system has become a significant competitiveness issue for the
American
economy, and will increasingly drive jobs overseas.
Our system is first class in terms of convenience, but very middle of
the road
in results; our health indicators are in many cases worse than those in
economies that spend far less. This is
We clearly have to do at least two things:
First, we have to alter the economics of health care. Any student who
has
finished Economics 101 in college can predict what will happen if an
anonymous
third party pays for incidents of service, not for results.
And second, we have to forge a system that is much more centered on the
individual. My mother is 86, is in reasonable health for someone of
that age,
and has six separate doctors who not only communicate rarely but on
occasion
countermand each other.
Altering the economics and centering the system on the
patient - defragmenting
it. The reason that eHealth is so important is that you can't do these
things
without it. Organizing and communicating health information simply and
to each
part of the complex system of health care is an essential tool that our
country
needs in its toolbox to succeed.
It is time to move eHealth out of the laboratory and into the real
world. To do
that, there is going to have to be some central direction and driving
and our
federal government is about the only place that can happen.
Many of you in this room remember as I do when President Kennedy spoke
to the
nation, and set an audacious goal of putting a man on the moon before
the end
of the decade. We all swelled with pride and purpose. Now suppose he
had added, "And to make this happen I'm going to let a thousand flowers
bloom; I'm
announcing a grant program to fund research on how to do this and pilot
programs
in states and on campuses across our great nation." We would have all
been
suitably deflated, and would turn our attention to other things.
The point here is that to have a chance of success we have to move
beyond this
tinkering around the edges approach of grants and pilots, make some
choices,
simplify our focus, and move ahead. This is something that likely only
the
standing of the federal government can accomplish.
How do we do this? I want to outline for you three things that I
believe we
could do to move this effort forward.
First, we need to establish a set of simple standards that can serve as
a
toolbox. I want to use the internet as a model.
Second, we need to reduce the size of the landscape that we are working
with;
to concentrate on one area and work through the problems there. I want
to
suggest the management of prescription medications - and ePrescribing
in
particular, although there are certainly others.
Third, we need to require the integration of information technology
into the
basic day-to-day operation of the practice of medicine. The challenge
is to
move beyond early adopters and those willing to experiment and into the
real
mainstream. Without this penetration, we have nothing. In Tennessee, if
Vanderbilt wants to
incorporate sophisticated IT in its healthcare practices, as it does,
that is
great; what I really need though, is Family Practitioner Dr. Jones who
is
using paper
records up in Goodlettsville to be a part, and there we're not even
close yet.
I'll suggest some ideas on how this might be jump-started.
First, to establish a simple set of general standards that can serve as
a
toolbox.
I've given one other speech on the subject of eHealth in my life, about
a year
ago, much shorter and to a much smaller but similar audience back in
Since I'm my own speechwriter on this subject, I did so, and what I
found was a
vastly more complex, less specific and less stable set of standards
than I was
advocating. I talked to some practicing systems people, and they told
me that
while the general outlines offered by HL-7, for example, were very
useful, the
state of the art was that you pretty much developed an information
exchange for
a specific version of the standards and a specific set of users, and
that the standards
generally did not work well across different versions or with systems
with
which you had not had previous conversations.
Let me say that there has been progress since last year, and I
acknowledge that
the HITSP standards published last fall are a real step in the right
direction
in providing an overarching encyclopedia of standards. However, the
basic
description of high complexity still stands. I counted 10 separate
standards
documents in the laboratory results reporting section alone; 24 in the
biosurveillance section.
As one person engaged in health IT put it to me, "The good thing about
standards is that there are so many of them."
In contrast, just think for a moment how simple and highly specific the
internet protocols really are. If your email server wants to send an
email
message to another, it connects using the universal TCP/IP protocol to
a
specific port, port 25, on the other computer. The other computer says,
"2, 2,
0, Space, Its Domain Name".
Your computer then says, "H, E, L, O, Space, Its Domain Name." and the
other
computer replies, "2, 5, 0, Space, Hello"
You then say... and so on.
It is an almost absurdly simple and specific way of communicating, but
that
standard serves up billions of email messages a day, it now easily
accommodates
secure messaging, and most important, it allows two computers anywhere
in the
world to communicate arbitrarily complex documents with no prior
arrangements.
I've just described a communications standard - SMTP; now consider a
content one:
World Wide Web pages. When Tim Berners-Lee invented the World Wide Web,
wrote
down a very simple tool for describing structured documents and for
extending
them using those links we all click on. Again, that tool - HTML - is
almost
absurdly simple; you can teach a willing 14 year old in an hour how to
make up
simple web pages from scratch with a text editor. The first web page
ever
written will display just fine in my new copy of Internet Explorer 7.
In
setting down a first simple standard, he didn't try to anticipate
Amazon, or
Google, or iTunes. He didn't try to involve every likely user of such a
web in
the design process. He didn't even try to specify the security
techniques that
now make online banking and shopping possible; that came later. He
invented
some simple tools that allowed his creation to take its first steps,
and with
those simple tools a million others quickly took his creation places
he - no
one - could ever have planned or imagined in advance.
I'm certainly aware that there are in this room professionals from the
various
standards bodies, others who have committed a great deal of time and
energy
working with them, and yet others who have financially supported this
activity.
I'm not being critical, I truly believe what you are doing is valuable.
In software terms though, it's as if you're being asked to write very
detailed
specifications for Version 5.0 of something before Version 1.0 is up
and
running. As the years go on, we'll get to much more complexity, to
Version 5.0
and 6.0 and beyond, just like the internet did, but it will be built on
and
grounded in what we have learned and invented in the real world and not
from
trying to divine the future or involve every known interest at the
outset. We
need to simplify and execute: simple standards; simple communications
protocols, and simple tools for structuring data, Version 1.0 out the
door and
into the customer's hands.
The internet is a great model here, let's learn from it.
So my first point is that we need some much simplified standards to use
as a
toolbox. My second is that we need to pare the immense problem of
eHealth down
to one that mere mortals can attempt; to cross the river hopping from
rock to
rock.
The information associated with health care is complex in their own
right, and
the health care sector is enormous. We have to break this problem down
into
manageable pieces to have any chance of success. To me, that would take
the
form of picking an area of health care and pushing forward to implement
a
solution widely in that area.
While some might see a narrowing of focus as a retreat from the large
goals
that have been widely articulated, I don't: I see it as real progress.
Success
in a complex environment almost always results from establishing a
beach-head;
get something working, get something established and then expand its
reach.
Success is contagious.
There are several areas that we might consider as a start.
We've established in the
But the area that is most intriguing to me as a beachhead is
ePrescribing, for
several reasons.
It offers meaningful improvements in the quality and cost of care;
reducing
errors, helping ensure compliance, and attacking fraud.
It is conceptually simple; you deal with a limited, known universe of
products
and transactions.
It cuts across the entire spectrum of patients, doing limited, simple
things
for nearly 100% of patients rather than delving much more deeply and
encountering much more complexity with 2% or 5%. You touch nearly all
patients,
you touch nearly all doctors.
It raises and requires practical solutions for many of the issues of
confidentiality and patient identification that await us.
It forces us to confront the challenges of widespread adoption; to make
it work
we need to sell a lot of down-to-earth practitioners who have no
inherent
interest in eHealth except as it positively affects their practice and
their
livelihood.
I've described two ideas: simpler, more stable and more specific
standards for
data exchange, and establishing a working beachhead. The underlying
idea here
is keep planning and thinking about the future, but to simultaneously
get
Version 1.0 working, widely used, gain experience, and learn from that
to
extend it. It doesn't have to be perfect or comprehensive any more than
DOS 1.0
or Word 1.0 was perfect or comprehensive.
The third principle centers on how we incorporate what we do into the
mainstream, day-to-day business of delivering health care. Aside from
early
adopters and true believers, how do you get ordinary, mainstream health
providers to use this technology, how do you establish a reason
powerful enough
to compel large numbers of them to invest their time and resources and
participate.
The frontier in getting practitioners to join is likely the physician
practice,
and doctors and nurses are notoriously conservative and difficult to
penetrate
with new technology that doesn't directly affect their practice. They
are on
the whole reluctant to change behaviors they've learned over the years,
and
reluctant to invest in things that don't have tangible short term
returns.
If we want to do what I have argued for today - let's say specifically
to
get an
ePrescribing 1.0 system up and running - we have to establish some
clear
incentives to do so. Health care is just too big and diverse to have
any hope
of depending on a voluntary model, or on arguments about patient care
or what
happens if there is another Katrina. Our entire health care crisis is
like one
slow-moving and prolonged Katrina; a glacier rather than a hurricane,
harder to
see it happening, but just as powerful and just as dangerous.
As with everything, one of the easiest and best incentives is the
incentive
that you don't get paid unless you do it. The U.S. Department of Health
and
Human Services is easily the biggest purchaser of pharmaceuticals in
the
nation; suppose they mandated that at some time certain in the future
they
would no longer pay claims where they couldn't take the claim and go
electronically with our simple and universal protocol and find the
specific
instance they were paying for in whatever database system we agreed to
use.
In other words if I am DHHS, I say, "What I am buying from you with
respect to
filling a prescription is now threefold: the drug itself, the
professional and
business expense of dispensing it, and the placement of appropriate
clinical
and administrative information regarding it in an agreed-upon and
widely
accessible location."
If DHHS were to require this, the Blues and major commercial insurers
and
intermediaries would quickly do the same; the benefits would be
enormous. As to
maintaining the databases themselves, I believe there would be no lack
of
volunteers to set up such databases; once you have really simplified
and standardized
the interactions, it isn't even that expensive. The framework for this
exists
in what drug chains are already doing. Even for those who don't use a
third
party payor, who pay for drugs themselves, they still overwhelmingly
buy them
at pharmacies where computers are integrated into the workflow, and it
is
simple to incorporate their information as well.
If such a strategy were to be adopted, there would be areas that need
cleaning
up in the law. That is an area that our NGA Committee is looking at,
and may be
able to contribute to.
So we might begin establishing a clean and simple database working from
the
payor's side. How do we now get the providers to use it? And how do we
move the
input of data upstream to them as well, with all the attendant
advantages in
completeness and accuracy?
IT is widely used in individual hospitals today because it is
integrated into
the workflow; you can't order a service any other way. How do we
integrate into
the workflow in other places like a doctor's office?
Here's one idea: use the power of authoritative "best practices".
Suppose that an authoritative body - it could be the accrediting
bodies,
for
example - specified that it was a "best practice" for a provider where
possible
to familiarize himself or herself with the pharmaceuticals now and
recently
being taken by the patient. That would in fact be a "best practice",
and one
which it would be reasonable to encourage as such - if there were a
practical way
to do it. We've been talking about practical ways to do this for the
past ten
minutes.
Many practitioners will try to correspond to best practices where they
can as a
professional responsibility. The defining of a "best practice" in this
regard
also engages the power of defensive medicine in a constructive way for
a
change - if a "best practice" is not followed, that failure can be an
element in
potential litigation - your malpractice carrier is going to want you to
do it as
well.
There are suddenly multiple good reasons to play: it's good medicine,
it's practical
and easy, and if you don't it could cost you.
How would a practice actually get this kind of information easily and
inexpensively?
Again, inventing as we go, imagine there is a web-based service that a
staff
member could access that, with proper credentials and patient
protections,
would provide a page summarizing the recent drug history of the
patient. It's
done via the web so that there is no issue of software to be learned or
maintained or bought, it uses something virtually everyone has - a
computer with
a browser. That page comes in a printable version, the printed sheet is
attached to the chart along whatever else the practice does for a
visit, and
all the information is right there for the provider at a glance.
That simple sheet of paper would catch an amazing number of medication
errors,
of duplications, and of fraud. It could alert physicians to new
information - the
Vioxx issue, for example. It could recommend best practices.
I'm not in the least concerned with who might provide such a service or
even
that it would cost the practice anything: if I were the CEO of a
pharmaceutical
company or a payor, I would do almost anything to be the provider of
that
sheet.
When the practitioner goes to write a prescription, he or she will
still likely
want to do so longhand - a PDA in every doctor's shirt pocket is still
a
long
time off. It's not a big step to have new and refill prescriptions
written on
that piece of paper, and to have them entered back on that computer by
the
office staff, where you can get an immediate check for interactions,
for
correct dosages, for compliance with an insurer's formulary, and so on.
Refills
are already managed in the back office as a practical matter, and
making the
refill process more effective and efficient would be a great start.
The sheet goes into the plain old paper medical record, the good
practice
requirements are addressed, compliance is documented and defensive
medicine
does some good for a change.
I've been inventing ideas on the fly this morning, and have talked in
very
concrete terms. I talked about specific data protocols, one specific
application - ePrescribing, and even specific pieces of paper in the
medical
office.
That may seem like an oddly detailed and technical discussion for a
Governor to
be making in an address like this. But I spoke in specifics because
that always
seems like a useful way of getting down to business. For those of you
who write
a lot, you know how much the act of putting actual words onto paper
clarifies
and makes concrete your thinking.
As I said at the outset, I acknowledge that this room is full of people
with
much more depth and knowledge of this subject than I have. There are
plenty of
people here who could take any of these ideas I've served up and
improve them
or show how some alternative is better.
Behind this presentation however are a few underlying principles from
an
experienced practitioner of the art of getting things done; a few
thoughts as
to how we might break the logjam in this field and start moving
something that
can truly improve health care into full being.
I do ask you to consider those. What I ask you to take away from this
and
consider is not all these details and invention, but rather three
principles:
Whenever you propose simplification and slimming down, you open
yourself to the
criticism that you're naive. If you're in my field - politics - you
open
yourself
to the criticism that you're taking the politician's approach of
putting
complex issues on a bumper sticker. That is not what I am doing.
I have found all my life that it is harder to simplify than to make
things
complex, harder and vastly more rewarding.
If you take nothing else from this morning, I want you to leave here
believing
that there is great power in simplicity and focus.
Jim Collins from Stanford wrote a management book a few years ago
called "Good
to Great"; I suspect a good many people in this room have read it. He
followed
it up about a year ago with "Good to Great in the Social Sectors". In
that
follow up he asks the question "Based on the return to the public
shareholder,
what is the most successful company in
People guess Microsoft, or GE, or maybe Amazon or Google or some other
high
tech startup. The answer is Southwest Airlines. That basic, simple,
low-cost
airline in a generally terrible business has economically outperformed
every
high-flyer, every conglomerate, every merger, every exciting New Thing
in
If there ever were a testament to the power of simplicity and focus,
well-executed, it is Southwest Airlines. Limited routes and cities, one
kind of
airplane, no reserved seats, peanuts, no interline baggage transfer,
you can
only buy your ticket from Southwest. And they represent not the best
performance of any airline, but of any company in
I hope I've given you this morning some things to think about. Thank
you once
again for the honor of being here, and thank you for all you are doing
and will
continue to do to make better the care and the lives of our fellow
Americans.