[Joint House and Senate Hearing, 108 Congress]
[From the U.S. Government Publishing Office]
S. Hrg. 108-209
TECHNOLOGY, INNOVATATION AND HEALTH CARE COSTS
=======================================================================
HEARING
BEFORE THE
JOINT ECONOMIC COMMITTEE
CONGRESS OF THE UNITED STATES
ONE HUNDRED EIGHTH CONGRESS
FIRST SESSION
__________
JULY 9, 2003
__________
Printed for the use of the Joint Economic Committee
U.S. GOVERNMENT PRINTING OFFICE
90-494 WASHINGTON : 2003
____________________________________________________________________________
For sale by the Superintendent of Documents, U.S. Government Printing Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800, DC area (202) 512-1800 Fax: (202) 512-2250 Mail: stop SSOP, Washington, DC 20402-0001
JOINT ECONOMIC COMMITTEE
[Created pursuant to Sec. 5(a) of Public Law 304, 79th Congress]
SENATE HOUSE OF REPRESENTATIVES
Robert F. Bennett, Utah, Chairman Jim Saxon, New Jersey, Vice
Sam Brownback, Kansas Chairman
Jeff Sessions, Alabama Paul Ryan, Wisconsin
John Sununu, New Hampshire Jennifer Dunn, Washington
Lamar Alexander, Tennessee Phil English, Pennsylvania
Susan Collins, Maine Adam H. Putnam, Florida
Jack Reed, Rhode Island Ron Paul, Texas
Edward M. Kennedy, Massachusetts Pete Stark, California
Paul S. Sarbanes, Maryland Carolyn B. Maloney, New York
Jeff Bingaman, New Mexico Melvin L. Watt, North Carolina
Baron P. Hill, Indiana
Donald B. Marron, Executive Director
Wendell Primus, Minority Staff Director
C O N T E N T S
----------
Opening Statement of Members
Senator Robert F. Bennett, Chairman.............................. 1
Witnesses
Statement of Dr. Mark McClellan, Commissioner, U.S. Food and Drug
Administration................................................. 2
Statement of Dr. Carolyn M. Clancy, Director, Agency for
Healthcare
Research and Quality........................................... 5
Report entitled ``Estimates of the Impact of Selected Health
Information Technologies on Quality and Costs in Inpatient
and Outpatient Settings''.................................. 49
Statement of Dr. Peter J. Neumann, Deputy Director, Program on
the Economic Evaluation of Medical Technology, Harvard School
of Public Health............................................... 23
Statement of Dr. Neil R. Powe, Director, The Welch Center for
Prevention, Epidemiology, and Clinical Research, The Johns
Hopkins Medical
Institution.................................................... 26
Submissions for the Record
Prepared statement of Senator Robert F. Bennett, Chairman........ 37
Prepared statement of Representative Pete Stark, Ranking Minority
Member......................................................... 37
Prepared statement of Dr. Mark McClellan, Commissioner, U.S. Food
and Drug Administration........................................ 38
Prepared statement of Dr. Carolyn M. Clancy, Director, Agency for
Healthcare Research and Quality................................ 45
Prepared statement of Dr. Peter J. Neumann, Deputy Director,
Program on the Economic Evaluation of Medical Technology,
Harvard School of Public Health................................ 53
Prepared statement of Dr. Neil R. Powe, Director, The Welch
Center for Prevention, Epidemiology, and Clinical Research, The
Johns Hopkins
Medical Institution............................................ 56
Statement submitted by AdvaMed, Advanced Medical Technology
Association.................................................... 60
TECHNOLOGY, INNOVATION AND
HEALTH CARE COSTS
----------
WEDNESDAY, JULY 9, 2003
Congress of the United States,
Joint Economic Committee,
Washington, DC
The Committee met, pursuant to notice, at 9:30 a.m., in
room SD-628 of the Dirksen Senate Office Building, the
Honorable
Robert F. Bennett, Chairman of the Committee, presiding.
Present: Senator Bennett.
Staff Present: Donald Marron, Mike O'Grady, Jeff Wrase,
Angela Brimhall, Colleen J. Healy, Melissa Barnson, Wes Yeo,
Rebecca Wilder, Frank Sammartino, John McInerney, and Nan
Gibson.
OPENING STATEMENT OF SENATOR ROBERT F. BENNETT, CHAIRMAN
Senator Bennett. The hearing will come to order.
Perhaps it's appropriate in a hearing dealing with health
care that I have a summer cold.
I apologize for the way I growl this morning, but there's
not much I can do about that. We welcome everyone to today's
hearing on how technology and innovation affect health care
costs.
The United States has a health care financing problem, one
that goes well beyond the budget challenges posed by Medicare.
For many years, our health care spending has grown at a
significantly faster rate than the economy. Projections
indicate that this will continue. Any financial arrangement
where expenses grow significantly faster than income is truly
on very shaky ground.
In other sectors, new technologies usually lead to greater
efficiencies and ultimately, lower costs. Yet, it's unclear
whether the same is true for health care. So what's different
about health care? Is it the technology or the way we pay for
it?
How can we strike the right balance--providing access to
the latest breakthrough technologies, while limiting an open-
ended raid on the public and private treasuries that fund our
health care?
During this hearing, we will explore these issues, bringing
together some of the best minds from the public and private
sectors to help shed some light on this situation.
We should first question whether technology and innovation
have truly added to health care costs, as some claim, or have
reduced health care costs through enhanced efficiency.
And second, we should examine whether new technologies are
disseminated in an efficient and effective manner, and if there
are areas where they are being overused or underused. For
example, some have expressed concern that advanced imaging
technologies may be overused, in part, because of poor
incentives in the payment formulas used by Medicare and other
insurers. At the same time, an article in this week's Health
Affairs, highlights how new technologies may be underused in
treating people who lack health insurance.
We need to find the right balance. We need to judge the
cost-effectiveness of new technologies so that we can properly
fund this critical work, without overpaying and without adding
additional upward pressure on health care spending.
Unlike most of the recent congressional debate on health
care, this hearing is not about Medicare or its coverage of
prescription drugs. However, this issue is crucial to Medicare
and every other health care purchaser that faces the dilemma of
how to add innovative new benefits without setting off an
explosion of health care costs.
[The prepared statement of Hon. Robert F. Bennett appears
in the Submissions for the Record on page 37.]
Senator Bennett. On our first panel, we're privileged to
welcome Dr. Mark McClellan, the Commissioner of the Food and
Drug Administration, who has testified before this Committee
before.
Dr. McClellan, we're delighted to have you back.
Dr. McClellan. Thank you.
Senator Bennett. And Dr. Carolyn M. Clancy, who is the
Director of the Agency for Healthcare Research and Quality. Dr.
Clancy, you're a new appearance here, but we're also delighted
to have you.
Dr. Clancy. Thank you.
Senator Bennett. Congressman Stark is unable to be with us
this morning because of a conflicting schedule. But he has an
opening statement which will be made part of the record at this
point.
[The prepared statement of Representative Pete Stark
appears in the Submissions for the Record on page 37.]
Senator Bennett. We are expecting several other Members of
the Committee to show up as their schedules will permit. And as
they do show up, I will recognize them for an opening statement
or questions as is appropriate.
With that opening and background, Dr. McClellan, again, we
welcome you and look forward to your testimony.
OPENING STATEMENT OF DR. MARK McCLELLAN, COMMISSIONER, U.S.
FOOD AND DRUG ADMINISTRATION
Dr. McClellan. Good morning, Mr. Chairman. I welcome this
opportunity to testify before this very important Committee.
It's a pleasure to see you again and I want to thank you for
your broad interest in health and in issues touching the FDA.
We've talked before about the importance of nutrition and
we're working on ways to help people improve their diet.
In fact, we have an important announcement on this topic
coming later this morning.
I'd glad to be here today to talk with you about the
critical questions you just raised on the effects of
technological innovation in health care on the cost of health
care.
It doesn't take an economist to appreciate that new
technologies often lead to higher costs of medical care.
Millions of Americans are struggling today to afford the rising
cost of what medical technology can do for them.
Many new medical technologies do lower costs, such as drugs
for treating depression that can be less expensive than non-
drug approaches to effective treatment.
Recently, FDA has approved new, simpler tests for HIV
exposure that can be done quickly in a doctor's office, as well
as less costly implantable defibrillators that can reduce the
risk of sudden cardiac death for many patients.
But many technologies do result in increased costs. First,
when a treatment becomes less expensive and safer, more
patients may decide that it's worth the risk and unpleasantness
to get it.
Second, many treatments exist today that do things that
were simply not possible in the past. Patients with heart
disease, cancer, AIDS, arthritis, cystic fibrosis, low birth
weights, and countless other conditions are living longer and
better lives because medical innovation has transformed fatal
illnesses, or illnesses that could only be treated with
supportive measures, into conditions that people can live with
and often live well.
The important fact that Americans are living longer lives
and better lives doesn't show up in any direct way in a
country's national economic accounts, like health care spending
does. But that doesn't mean that these health benefits aren't
worth a lot.
So from an economic standpoint, one key issue is whether
the benefits of medical innovation are rising faster than the
costs.
In recent years, a number of economists and doctors and
other health care experts have addressed this question. It's
hard to answer since it's hard to put a value on better health
and since there are many things besides medical care that
influence health.
Perhaps the best evidence comes from studies of changes in
treatments and associated changes in costs and outcomes for
patients with specific illnesses over time, like patients with
heart attacks or cataracts or depression.
While none of these studies are completely convincing in
themselves, they generally show that medical innovation have
been of great value to the public. That is, the value of the
improvements in health are much greater than the increases in
spending.
Another way to look at this is if you are a patient with
heart disease or at risk for breast or colon cancer with
rheumatoid arthritis or with many other conditions, you'd
generally be much better off with the treatments that you can
get today compared to the inferior treatments you could get a
decade ago, even though in all of these cases, treatment for
your condition is much more expensive than it was a decade ago.
But just because our leadership in medical innovation in
America has added great value in the form of longer and better
lives for millions of Americans, it doesn't mean that we should
just be sitting back and doing nothing.
Just because changes in medical technology have been good
overall doesn't mean that we can't do better. There are lots of
examples of medical treatments used inappropriately or
erroneously or in other ways that add to costs without
providing much, if any, benefits. And in addition, many people
can't afford some of the valuable new treatments and that's a
public health problem, too.
We must work hard to find better ways to increase value in
the development and use of medical technologies. We must work
to keep modern care affordable while still encouraging medical
innovation.
I'm quite concerned about threats to valuable innovation in
health care today. On the one hand, the process of medical
innovation, turning sound ideas from insights in the biomedical
sciences into safe and effective treatments for patients. This
process has become steadily more expensive, more time-consuming
and more uncertain.
That means it's getting much more expensive to get new
technologies to patients.
On the other hand, we are also under more pressure than
ever to find ways to bring health care costs down, and some of
the ideas for reducing health care costs would unfortunately
reduce the financial incentives needed to bring valuable, life-
saving technologies to patients.
We're facing this crisis at a critical time from the
standpoint of medical innovation. The number of new
technologies coming to patients is down.
For example, we got fewer applications for truly new drugs
last year than at any time in the past decade. But this is
happening at the same time as the investment in research and
development by the National Institutes of Health and by the
private sector is higher than ever--over $80 billion, with the
promise of new breakthroughs ahead from understanding the human
genome and many other sciences like genomics and medical
nanotechnology.
If the cost of developing new products that are safe and
effective keep going up, while short-term efforts to control
costs increasingly focus on controlling payment rates, we may
not get more valuable new treatments in the years ahead.
I think there's a better solution, one that means better
health and greater value for medical technology in the years
ahead.
We can take steps today to improve the development and use
of medical technologies and find creative policy solutions that
both support innovation and make health care more affordable,
particularly for those with limited means and great needs.
As part of a new FDA initiative on improving medical
technology announced in January of this year, the FDA is taking
many steps to help foster more efficient innovation, especially
in emerging areas or those with great medical need.
We are working not only to reduce the time for reviewing
new products and determining whether they are safe and
effective. We are also working with partners at the NIH and
with product developers to find ways to make the development
process less costly and more predictable. For example, by
providing clear guidance on what it takes for a product
developer to show that a new treatment is safe and effective.
Lower costs and more certainty in developing new medical
technologies means more safe and effective treatments can reach
more patients faster.
In order to get more value from the medical technology we
use, however, after new technologies are approved, we also need
to work to do more to help doctors and patients use new medical
innovations more effectively. And so, we're working closely
with many participants in health care, including with the
Agency for Healthcare Research and Quality, led by my friend
and colleague, Dr. Carolyn Clancy, who I'm delighted to testify
with today.
We're working with health care organizations to collect
more information, automatically, on potential safety problems
with products after they've been approved.
We're implementing new bar-coding requirements to make sure
the right patient gets the right treatment, avoiding costly
medication errors.
And we're conducting more post-approval studies to develop
better, more up-to-date evidence on safety and effectiveness,
the risks and benefits of medical products after they are
approved.
We're working on a daily med program for physicians using
an electronic version of our product label for physicians that
is easier for them to use to get the treatment information they
need for each patient they're treating. And it can be updated
daily to include the most current information about the risks
and benefits of the drug after it's on the market.
Only by facilitating development of complete, timely, and
easily used information can the FDA help make sure that people
are making the best decisions about their health based on the
best available information.
Mr. Chairman, the FDA is working with AHRQ and with our
partners throughout the Federal Government and the private
sector to promote increased access to high-quality, safe and
effective medical technologies, including drugs, biologics,
devices and combinations of all three.
This is the best way from a public health standpoint to
make health care more affordable and to make sure that we get
the most value for medical technology.
I'm sorry we don't have any more products coming along
sooner for the common cold, but I would appreciate the
opportunity to have my written testimony read into the record
and I'd be very pleased at this point to hear what Dr. Clancy
has to say.
[The prepared statement of Dr. Mark McClellan appears in
the Submissions for the Record on page 38.]
Senator Bennett. Thank you. Your written testimony will be
part of the record.
Dr. Clancy.
OPENING STATEMENT OF DR. CAROLYN M. CLANCY,
DIRECTOR, AGENCY FOR HEALTHCARE
RESEARCH AND QUALITY
Dr. Clancy. Good morning, Mr. Chairman. I'm very pleased to
be here today to discuss the important issues of how we can
facilitate, sustain and promote health care innovation and
ensure that we have a health care system that is affordable.
And I'm also quite delighted to be here with Dr. McClellan.
I wanted to start off by just telling you a little bit
about the Agency for Healthcare Research and Quality (AHRQ).
Our focus is improving the effectiveness, quality, safety
and efficiency of the health care delivery system. So our work
complements that of the Centers for Disease Control, which
focuses more on public health, for example, through the use of
public health ad campaigns.
To improve health care, we focus both on the clinical
content of the care, as well as the systems or settings where
people receive care.
So it's that dual focus that is a unique focus for the
agency.
We contribute to efforts to speed the diffusion of
effective medical breakthroughs. Through effectiveness in cost-
effectiveness research, we can extend the findings of
biomedical research to populations not included in clinical
trials, determine whether patients in daily practice actually
achieve the promising benefits seen in clinical trials, and
identify which people benefit most and least.
So, for example, if a new breakthrough came along for the
common cold, we would be able to help clinicians understand
which patients were most likely to benefit and which patients
might be harmed or not likely to benefit at all.
Our expanding portfolio of implementation research develops
effective strategies to facilitate the rapid adoption of
effective services and technologies.
We also facilitate adoption of new knowledge by putting
into perspective available scientific evidence so that
clinicians and their patients can better assess the importance
of recent breakthroughs, an issue of increasing importance as
new interventions appear almost daily in the media.
So, for example, where the FDA determines that a drug,
biologic or device is safe and that it has an impact, usually
when compared to a placebo, those making coverage decisions and
those making clinical decisions need more information regarding
its relative effectiveness and relative cost--how does it
compare to the other options I have?
For example, our evidence reports and technology
assessments assist Medicare in making coverage decisions of new
clinical interventions.
One area of increasing importance that's relevant to this
discussion is in assessing safe use and minimizing unintended
harm of health care interventions.
While FDA plays a key role in ensuring the safety of drugs,
biologics and devices, their inappropriate use can still lead
to patient harm, and that's an area where our agencies are
collaborating closely.
But there are other innovations in health care, such as new
surgical procedures and medical interventions, or new
applications of existing technology, for which there is no
comparable up-front evaluation of safety.
While some of these innovations offer unprecedented
breakthroughs for some patients, they may also result in
unintended harm, if not used appropriately.
And this unsafe use is both a personal tragedy for
individuals and their families, as well as a big source of
unnecessary costs as clinicians struggle to repair the damage
and as medical liability expenses mount.
This is a growing focus for AHRQ, and the area of drugs,
biologics, and devices is an area where we work closely with
the FDA.
Mr. Chairman, the pace of health care innovation continues
to accelerate. It's increasingly difficult for clinicians and
patients to assess their options adequately.
Many of these developments offer patients the potential for
greatly improving the quality of life for patients, as Mark has
noted. In other cases, the improvements are marginal at best.
Some innovations lead to significantly lower costs, while
others are cost increasing.
The big challenge underlying this is to effectively sort
through the increasing array of clinical care options to
develop objective scientific information so that those making
decisions--policymakers, systems leaders, insurers, employers,
clinicians and patients--can make informed choices.
Whether you favor our current insurance-based system or
favor a more consumer-driven model of care, the need for
objective evidence is compelling and remains constant
throughout.
The resurgence of health care cost inflation, combined with
expected growths attributable to the investments that Mark
McClellan noted in biomedical research, will only accelerate
this demand for objective information.
So I wanted to tell you five ways in which AHRQ can help.
First, AHRQ research identifies what's effective and cost
effective in daily practice.
Experience suggests that new drugs, technologies and
medical or surgical interventions are seldom equally effective
for all types of patients.
For example, will a breakthrough for the treatment of
arthritis tested in clinical trials for patients who only have
arthritis work as well with patients who also have diabetes,
heart failure and hypertension?
Or how well will it work in patients whose racial, ethnic
and demographic characteristics differ from those enrolled in
the original trial?
My written testimony provides two examples that demonstrate
the importance of avoiding simplistic judgments about new
technologies.
In one case, treating middle ear infections in kids, a very
common cause of seeing physicians, we demonstrate the value of
using the low-cost option, generic antibiotics.
In contrast, in a study of the use of very expensive, but
highly effective drugs to treat AIDS, we demonstrated that the
long-run savings that result from the use of these much more
expensive drugs more than warrants their use.
Second, AHRQ research identifies strategies for overcoming
barriers to the use of effective services.
Two weeks ago, you may have seen a lot of headlines about a
study that was published in the New England Journal of Medicine
from the Rand Corporation saying that getting quality of care
in this country was effectively a little bit better than
flipping a coin--about 54 percent of the time is what they
found.
The vast majority of the areas that they measured in
quality of care related to underuse of effective treatments.
Great opportunities for improving health developed through
biomedical research are easily lost if physicians and patients
aren't able to make the best use of that knowledge in every-day
care. And that is a big focus for the agency.
Third, AHRQ facilitates the use of evidence-based medicine.
Developing and synthesizing evidence and objective
information about various clinical options is important. Making
it useful in real time is essential.
In recent years, AHRQ has focused increased attention on
the development of technologies and tools to facilitate the use
of evidence-based medicine.
For example, every year, tens of thousands of patients go
to an emergency department with chest pain and they're worried,
as are their clinicians, that they might be having a heart
attack.
We developed a tool that has now been incorporated into
EKGs that helps clinicians make better decisions and which we
estimate could prevent 200,000 unnecessary hospitalizations and
100,000 unnecessary coronary care unit admissions a year,
saving over $700,000 million annually in costs.
Fourth, AHRQ research assesses the effectiveness of cost
containment and management strategies.
Medicaid pharmaceutical costs are increasing at about 20
percent a year and obviously are the source of great focus and
attention by states right now.
As an example of how our past research was helpful to
today's decisionmakers involved a study of a strategy used by
one of the New England states. And what they did was that they
limited Medicaid prescriptions to three drugs per patient.
Indeed, they saved money on pharmaceutical costs.
The only problem was that they spent more than 17 times
what they saved in unnecessary admissions to emergency rooms,
nursing homes and to the hospital.
So that was of an unintended harm or unwanted aspect of an
intervention that was intended to control costs.
When I actually mentioned this to state legislators in Dr.
McClellan's home state, their eyes got really big because I
think they had been thinking about this strategy as well.
The results from this study led nine other states to change
their policies.
Finally, AHRQ has a role to play in speeding the pace of
evaluation of health care innovation.
One of the critical roadblocks to coverage of innovation
interventions is the lack of solid scientific evidence
regarding their effectiveness, especially in contrast to
existing interventions.
This is often frustrating to those whose creativity leads
to the development of new breakthrough interventions and then
come to realize that they have to get through FDA and CMS
scrutiny, that that's only part of the journey toward seeing
their innovation and widespread use.
I wanted to give you an example of a surgical procedure
because that does not come through the FDA.
We work very closely with the Centers for Medicare and
Medicaid Services and they asked us several years ago to
evaluate a new surgical procedure called lung volume reduction
surgery.
We did an assessment using existing data and found that
some patients benefited enormously, near-miraculously. Others
were harmed.
And what we said to the Medicare program was we can't say
with confidence ahead of time which patients are most likely to
benefit. We think conditional coverage linked to a randomized
clinical trial might be the way to go here.
As a result, the Medicare program turned around and said
that they would only pay for this procedure for the patients
enrolled in a clinical trial.
What happened was that we now know which patients are
likely to benefit and very importantly, we identified a very
high-risk subgroup of patients who are likely to be harmed.
That is to say, they are likely to die from their end-stage
lung disease, much more rapidly, which I think is a very
important contribution to the public's health overall.
There are at least two other ways in which we can improve
how we work as a science partner to promote private-sector
innovation.
First, we want to work closely with industry trade
associations to assist their members who have products moving
to the end of the FDA review process to better understand the
types of studies that will be needed to assess the
effectiveness of their products.
This simple step will facilitate more timely assessment of
health care innovations.
Second, as our existing investments in patient safety come
to an end, we want to expand our focus on human factors
research. This is research that helps us idiot-proof our
technology--for example, making sure that the controls on all
new machines and devices work consistently in the same way that
a pilot, any time that he or she steps into a 747, knows that
the dials are all in the same place.
That way, even if health care professionals are distracted,
stressed or, sometimes, sleep-deprived, they will provide safe
care.
By ensuring that this type of critical information is in
the public domain, we can be a science partner for these
private industries to develop even more effective and safer
health care technologies.
Before I conclude, I just want to say a few words about the
future directions of AHRQ.
We're determined to make the agency even more of a problem-
solving agency. This will entail a greater focus on
implementation research to help overcome barriers in the
adoption of clinical interventions that are both effective and
cost effective.
We've developed closer linkages throughout the research
process between the ultimate customers of our work and our
researchers to ensure that we're addressing their highest
priority challenges.
We're also giving greater emphasis to identifying
strategies for eliminating waste, assuring that evidence-based
information is current and up-to-date, bringing our health care
infrastructure, particularly information technology, into the
21st Century, and redesigning work-flow so that health care
professionals can work more efficiently and effectively. And
finally, evaluating financial and other incentives to encourage
safe, high-quality care.
In conclusion, let me just say that a series of studies
have demonstrated that the timeframe for approval of a research
grant that ultimately leads useful findings to the widespread
diffusion and adoption of those results was, on average, about
17 years.
We consider this timeframe unacceptable.
Now this study did not actually look at the products
developed by the private sector, but I can tell you that there
are many other studies that would suggest that they're probably
not all that much far ahead of the curve.
We're prepared to play an important role in identifying
effective interventions and increasing the pace of their
diffusion.
Thank you very much.
[The prepared statement of Dr. Carolyn Clancy appears in
the Submissions for the Record on page 45.]
Senator Bennett. Thank you. Let me give you a personal
example of the dilemma that we face here when we're looking at
the overall system.
My wife used to jog. She's gotten more sensible in her
later years, and she doesn't do that anymore.
She tore the cartilage in her knee. And went in, got
examined, and was called for surgery. She spent, my
recollection is, a week in the hospital. She has a scar on her
knee as a consequence.
Obviously, the surgery and hospital stay were expensive.
Some years later, she had to have the same kind of surgery
on her other knee. Arthroscopic surgery done as an out-patient
took about an hour. I waited in the waiting room while it was
done and took her home the same day.
Obviously, there was no increase in cartilage injury by
virtue of the invention of the arthroscopic procedure. So you
can't say that the technology stimulated enough new procedures
to take the cost up.
The number of injuries were the number of injuries were the
number of injuries.
So the dramatic cost savings in arthroscopic knee surgery
entering the overall economic picture would indicate that the
cost would come down.
Now, when you look at the fact that costs go up while this
kind of innovation is going on, and there are a number of other
examples like that, you come to the issue that I think you
addressed a little, Dr. Clancy, that there must be some parts
of the system where the cost has gone up exponentially, because
it's not just gone up the 15 percent per year that we're
looking at right now.
In those areas, it has gone up enough more than the savings
to eat up all of the savings and produce on top of the savings
a 15-percent per year increase.
Now, is there ever any prospect in the future that we're
finally going to catch up with this, whether we do it through
evidence-based medicine--which is a great phrase that I like--
or through the analysis, Dr. McClellan, of the benefits
outweighing the costs so that we can continue to justify doing
this?
At some point, will the impact of the--if I can put it in
these terms, the arthroscopic surgery lowering of costs--catch
up with whatever it is that's driving the increase of costs so
that the overall number levels out?
Do either of you have a view of what the next 5 to 10 years
might bring in that regard?
Dr. McClellan. I think there are many examples like the one
that your wife experienced where innovations in medical
technology and capabilities have made it safer and less
expensive to achieve a given treatment, to achieve a given
improvement in outcome for a patient.
In your wife's case, it was repairing some damaged
cartilage.
There are a couple of reasons why, even though that's less
expensive in terms of both the procedure itself and hospital
time and complications, that may not even in itself translate
into lower health care costs overall.
One reason is that when the procedure gets easier to do,
more patients tend to decide that it's worth having it. So that
people with milder injuries, for example, may be more likely to
undergo arthroscopic surgery when they would never consider
going into the hospital for that more miserable week's stay and
major surgery that your wife experienced some years ago.
Senator Bennett. So you're saying that knee or cartilage
injury is a case of elasticity of demand.
Dr. McClellan. If you look at the number of procedures, it
has gone up a lot.
And in fact, picking up on something that Dr. Clancy talked
about earlier in terms of appropriate use, there have been some
studies recently, including the study in the Veterans
Administration, that showed that, in a number of cases,
arthroscopic surgery may be being performed where the benefits
don't outweigh the costs or the risks to the patient.
And that's why developing better information, as Dr. Clancy
and I both emphasized, on risks and benefits of a new
technology for a particular kind of patient, can be very
helpful.
And it's also why the linking up of incentives for using
those technologies in certain patients so that people think
about the costs, might be helpful as well. And that's where a
lot of health policies have been directed in recent years.
There also have been a number of new treatments coming
along for knee injuries that just didn't exist before. It's now
a lot easier to get a knee replacement for people with severe
arthritis who couldn't walk or couldn't walk easily, so that
they can get around and even take up jogging again.
The technology has gotten a lot better.
So, previously, that wasn't very expensive if people just
sat around at home and didn't do much because their knee was
gone.
Now, they can have a much better quality of life in getting
around, but it's added to our health care costs.
All this just goes to show, as you said, that we need to
make sure, or do more to make sure that we're getting a lot of
value out of the new treatments when they're actually used in
practice.
Senator Bennett. So your answer to my overall question is
no, there's not going to be a time in the future when it starts
to level off?
Dr. McClellan. Well, I think a lot of people have been
reluctant, a lot of economists have been reluctant to predict
that because they look back on 50 years of experience of health
care costs growing significantly faster than our overall
economy, and then they look down the pipeline in terms of new
treatments being developed as a result of genomics and other
sciences that are just now starting to have an impact on
medical care.
They see a lot of conditions that are either not treated
today or not treated well--Alzheimer's, many forms of cancer,
other illnesses where people really do have to live with a lot
of disability, if not die, because they suffer from it.
And that gives them reason to think that costs can go up.
On the other hand, I wouldn't be completely pessimistic
that we can't do a lot better if we make the right policy
choices.
There are a huge number of examples, and Dr. Clancy talked
about many of them, of us spending a lot of money on health
care for treatments that don't do much or anything for
patients' health outcomes and may even, in the case of medical
errors and preventable adverse events, make outcomes worse and
add to costs.
And by some estimates, the savings could be in the many
billions of dollars per year.
So if we can find ways to get the new technologies, the
valuable new technologies, moved along, while at the same time
reducing the spending on treatments that don't really do much
for patients, we can have a much more viable health care
system, one that does more for patients.
It may cost more overall. It may cost less. But we'd be
getting a lot more for our money. And that's what I think we
need to focus on.
Senator Bennett. Dr. Clancy.
Dr. Clancy. Yes, I would agree with all of Dr. McClellan's
comments, and we've confirmed this, his observation that if you
make a procedure much easier and less painful, that more people
are likely to want to do it.
For example, we had a research team in place when the new
procedure for gallbladder surgery was introduced, which made an
incredibly miserable operation far more bearable and much more
faster recovery.
Since I'm a physician, but not an economist, unlike my
colleague who is both, I don't have to be quite so embarrassed
about predictions, I don't think.
[Laughter.]
But if I were to look at the population's health and
demographic changes affecting our population, what you see is a
general aging of the population here and in all developed
nations.
I think the great opportunity for savings is two-fold.
One is waste in the health care system and that's a big
focus of the agency's efforts.
The second, though, is really improving the quality of life
for people who now suffer impairments in quality of life.
If you look back over the 20th Century, the huge
achievement was the expansion in life expectancy. What that has
not translated into far enough is improvements in quality of
life in the later years.
So that mantra of adding life to years does make a lot of
sense and I think actually has an enormous opportunity for us
to be able to save money downstream.
What this will mean is a lot more focus on helping people
with chronic illnesses, whether that's arthritis, who benefit
from joint replacements, or people who benefit from new drugs
for treating a variety of conditions.
And I think the other contributor here that will be very
helpful is information technology. I think as more and more
people become engaged as partners in managing their own care,
whether it's diabetes, high blood pressure, or other things,
that will help us actually obtain greater value from our
investments in health care.
Senator Bennett. Let me go to the question of waste. It's
something that you dealt with in your testimony, which again
triggered another personal experience.
I woke up in the middle of the night one night having some,
for me, unique and a little bit strange and ultimately
frightening kinds of symptoms.
Finally, my wife woke up and said, ``Do you think we need
to go to the hospital?'' And I said, ``Yes.''
She drove me to Georgetown Hospital in the middle of the
night. I just presented my Federal employee's card and was very
impressed with how excited everybody got about taking care of
me in the midst of what was an apparent heart attack.
I commented to her about how somebody just off the street
that they'd never seen before was being taken care of. And she
made it very clear, she told everybody I was a United States
Senator.
The head of the cardiac department at Georgetown University
showed up immediately.
It turned out what I had was known as an esophageal spasm,
which has exactly the same kind of symptoms as a heart attack,
particularly for somebody who has never gone through it. But
it's not life-threatening and, indeed, disappears. As mine did.
They thought I was having a heart attack and the reason
that it disappeared is because they put a nitroglycerine tablet
under my tongue. In fact, it had nothing whatever to do with my
condition.
I was in the hospital for, I think, 3 days, on all kinds of
monitoring machines, et cetera.
Finally, they decided after the most extensive and
obviously expensive series of tests, that, no, this wasn't a
heart attack at all.
Now you referred in your testimony to some kind of
technology that can determine that. Right at admission, they
could have patted me on the head and given me a purple pill and
sent me home that same night and everything would have been
fine.
I've had some of those symptoms since, and all I do is take
Prilosec and it goes away.
According to the latest stress test that I had at the
Bethesda Naval Hospital, they said, ``We don't need to see you
for another 10 years. Your heart is as sound and solid as any
we've seen.''
So that raises--you talk about waste being one of the major
problems. We think of waste, fraud and abuse as a continuum,
and there's something sinister about it.
Here was a case of waste where there was nothing sinister
whatsoever, and there was certainly no fraud and there was no
abuse of the system.
They didn't know and they were taking every intelligent
precaution to see to it that I stayed alive. Because if it had
been a heart attack and they had not done the things they had
done, they would have been guilty of malpractice.
What are the prospects of dealing with waste through
technology? The Federal bureaucratic mentality, as I say, is
that waste, fraud and abuse are all a single thing. So we
simply say, we're not going to pay, like the example you gave
of the state legislature saying, well, you can only have three
prescriptions. You may need five or six, but we're not going to
pay because doctors are overusing and overcharging and engaging
in waste, fraud and abuse.
We'll fix that by passing a law that says that you can only
have three. But this ends up costing the system a whole lot
more in another area that doesn't get counted as you're
congratulating yourselves on how much you've brought down your
prescription drug cost.
What are the prospects of dealing with what I would call
from the example I've just given you, benign waste, well-
intentioned waste, through greater technology, bringing down
the cost of that kind of waste through greater technology?
Dr. Clancy. I want to make a distinction between diagnostic
uncertainty, which in your case, even though in retrospect you
could have just gotten your purple pill and gone home, it
sounds to me from how you've described it that they did
everything appropriately. And I don't think anyone would have
wanted them to do one thing less.
Senator Bennett. I wasn't complaining.
Dr. Clancy. Yes.
[Laughter.]
Dr. Clancy. One big source of waste or increased efficiency
that I think is achievable is in the care of people with
chronic illnesses.
There was a study published a couple of months ago, also in
the journal, Health Affairs, that surveyed people in five
countries with various chronic illnesses.
And what you heard consistently--it's interesting, sort of
a global phenomenon--was that many of these folks saw multiple
doctors. They tended to have the same test ordered twice. After
all, I might know that Dr. McClellan has ordered a test on my
patient or my patient might tell me that. But if I can't find
the result or if I'm not really sure, I'm going to order it
again. They also had medication errors and that were lots of
opportunities for miscommunication.
I think that's where investments in information technology
can make a huge difference because I don't have to look for the
result. I can actually just check on it in the computer.
Very recently, the Department of Veterans Affairs has shown
that those types of investments can pay off huge benefits in
terms of improving the quality of care.
So that's one obvious source.
There are lots of opportunities I think to make our health
care systems more efficient and that is a big focus of our
research right now.
Senator Bennett. A central repository of information about
every patient would lend itself to what you've just described.
Dr. Clancy. That would be one model.
Senator Bennett. It raises all kinds of privacy
implications.
Dr. Clancy. Right.
Senator Bennett. And confidentiality of medical records.
And I've spent a lot of time on that issue, too.
The technology exists, I believe, to do what I'm about to
describe. Let's follow that road for a minute.
I step off the curb in New York City, don't look the right
way, and get hit by a taxicab and become unconscious. No one
has the slightest idea who I am or anything about my medical
records, but I am rushed to a hospital.
They find in my wallet a card, which they can put into a
reader somewhere, where my entire medical history is available
on a chip.
All of a sudden, they not only know my name and my social
security number, but they know the level of daily medications
that I take, they know that I've had two hernia operations, an
appendectomy, that my EKG is abnormal, but my heart is not, et
cetera, et cetera.
They know all about me instantly.
And you say you didn't have the information of the latest
test. On that chip, on the card in my wallet, they can know the
date of my latest stress test at Bethesda Naval Hospital, dial
it up somewhere, and instantly get those results.
So that as they work on this unconscious, unknown
individual at Bellevue Hospital or wherever it is I'm taken in
New York City, they have everything in front of them,
everything available to them, and presumably, can then make not
only the best diagnosis, but save huge amounts of money and
give me the right kind of treatment.
The technology to do that exists. Let's set aside the
confidentiality and privacy issue for just a moment and ask
ourselves what would be the economic benefit if that technology
were implemented and everybody carried such a card?
Would health care costs go up or down?
Dr. McClellan. It would certainly help avoid some of the
kinds of duplicative costs of delivering health care that Dr.
Clancy has emphasized exists too often today. And that could
lead to some improvements in costs through less duplicative
tests and higher quality of care for exactly the reason that
you mentioned.
That's a more efficient system.
There are some obstacles to implementing that system.
Senator Bennett. There are a few, yes.
Dr. McClellan. You know, the department has implemented a
strong new privacy regulation. I think it has given people much
more confidence about how their electronic sensitive medical
records are being handled.
There are some other obstacles in terms of standards and
different institutions and organizations store their electronic
data in different ways, using different codes.
Dr. Clancy and the rest of the Department of Health and
Human Services and the rest of the Federal Government have been
involved recently under Secretary Thompson's leadership to try
to get more standardized systems for keeping track of medical
information so that it can be shared across health care systems
effectively.
But there's another type of benefit from using health
information more effectively and that's it can let us learn
more about what treatments are working and which ones aren't,
which ones may actually be causing safety problems and harming
patients in actual use.
It can be difficult for us to get that kind of information
today.
For example, at the FDA, we've long relied on reports from
manufacturers of products to tell us when something is going
wrong. They in turn have to rely on increasingly busy and
harried health care professionals to send information in to
them when something goes wrong with a drug or device.
With the kind of information system that you described,
we'd have a way of capturing automatically in real time or
close to it information that could put us on to an important
safety problem or maybe even an important benefit that's not
well understood for patients. And we can in turn get that
information out to doctors much more effectively by using that
information system going the other direction.
So it should be a two-way street.
So there are a lot of potential benefits there. There are
some obstacles along the way, including confidentiality and
standards and providing the right kinds of incentives for
health care organizations to adopt these modern information
systems.
But there is a tremendous potential there.
Senator Bennett. Well, you get to the question of common
database protocols. I think that's what you're describing here.
Another example.
Intermountain Health Care in Salt Lake City ran a series of
hospitals--still does. And I believe I have these numbers
right. It's been a while since I dredged up this particular
example, but you've triggered it with this testimony.
The question of infections after operations is a serious
question. People go into the operating room. The operation goes
well, but they get an infection.
The standard that was established was 2 percent. If you
could get your infection rate down to 2 percent of the
operations, you were labelled as an acceptable and, indeed,
admirable, kind of operation.
At Intermountain Health Care, they decided that they were
going to experiment a little, not with the patients, but with
the data. And they started checking various things that
happened in the operating room to discover if there was any
correlation between certain things and the elimination of
infection.
I can't remember exactly what they discovered. My memory
tells me that it had something to do with the timing of the
injection, or the introduction of antibiotics, or whatever,
that if they waited past a certain period of time, then there
were infections. And if they did it within a certain period of
time, there were none.
They changed the protocol in the operating room to
correlate with the information they had discovered by virtue of
their research and they brought the infection rate down to 2/
10ths of 1 percent.
Now industry standard would say, if you meet the 2 percent,
you get the seal of approval and everybody accepts that as
being normal.
And they were able to bring that down to 2/10ths of 1
percent.
So, naturally, if you're going to have an operation, you
want to be in one of IHC's hospitals with respect to the
infection problem.
Is there some way that could be devised or adapted to where
you work where that kind of discovery--it's not a research
discovery in the sense that you've got a new drug or a new
device. It's simply a statistical discovery of examining what's
going on and saying, wait a minute. It really matters whether
you do this in the first 20 minutes or the first hour.
To create a central repository of that kind of information
that could then produce a national protocol that says, this is
the way every hospital ought to do it, and take that example
and spread it out over all of the things that can be discovered
that you're talking about, Dr. Clancy.
Respond to that and see if that is something that the
government should be involved in.
Dr. Clancy. Well, I'm very proud to say that we actually
funded the study you're talking about at Intermountain Health
Care.
Senator Bennett. Oh, did you?
Dr. Clancy. [Continuing.] Which identified just how
important it is to get the timing of the pre-operative
antibiotics right.
They also demonstrated that information technology can be a
very important part of reinforcing and making sure that that
happens.
The type of research that you're describing, the systems
research, how do we make sure that what we know works is
actually what happens, is very much a focus of the agency's
work right now.
This fall, we're going to be putting out a big report on
quality of health care, sort of a national report card, if you
will. And in preparation for that, we're beginning to review
all the evidence about what we know works best and are also
hoping to use that as a launching pad for improvements.
So we'll make sure that you get one of the first copies.
Senator Bennett. That's nice to know that there's a report.
But just to pick a city at random, suppose I go into a hospital
in Detroit, where they haven't read the report.
Is there any system for getting the information out other
than we published a report and hope somebody picks it up?
Dr. McClellan. Dr. Clancy emphasized that AHRQ and other
Federal agencies, including us, are trying to get better
information developed so that doctors would have access to the
best and latest information on risks and benefits for a
particular patient of a particular treatment.
And I think that the kind of system that you're talking
about may not come together as just one single global database,
but there are a lot of programs out there that can help doctors
get more accurate information for treating patients.
I'd like to emphasize, though, that that's not enough. As
you emphasized, this is something that happens at the local
level when doctors and nurses and other health professionals
delivering care to individual patients, just having an attitude
and having the support they need to make the right decision at
the right time and avoid errors is something that needs to be
part of the system, part of the environment in which health
care practitioners are functioning.
And having access to information technology can help with
that. But other steps are necessary, too.
For example, one barrier that many institutions cite about
trying to develop that same kind of information so that they
can keep track of why infections are occurring and how we might
prevent them, is concerns about liability.
They're afraid that if they write down what might go
wrong--what went wrong or might have gone wrong for a
particular patient, that's going to end up being held against
them in court.
And I'm pleased to say that there's bipartisan legislation
working its way through Congress, at least the House, which I
know AHRQ and we strongly support that would provide liability
protections that are needed to encourage the environments that
promote safe and effective medical practice.
And also financial incentives matter as well.
Some institutions still today get paid more for treating a
patient for not only the condition that brought him into the
hospital, but for the infection that might keep them there
longer or get them re-admitted to the hospital.
Incentives should be in the right place for getting
patients well and preventing errors in the first place. There
are more supporting things that need to help, that if you get
them in the right place, would help make that kind of national
data that you're talking about be used much more effectively.
Senator Bennett. You're not suggesting that anybody
rejoices or deliberately does things that would cause a patient
to stay in the hospital longer?
Dr. McClellan. Not at all. I'm just pointing out that
financial incentives to help people stay healthy can make a big
difference.
Dr. Clancy. And just to build on that. In your home state
again, at Intermountain Health Care, Brent James has a long and
impressive list of examples where they have improved quality
and the safety of health care and have lost money.
Now they're doing it because it's the right thing to do.
And the reason they've lost money is related to payment policy
and the fact that we pay institutions more for taking care of
sicker patients.
Actually, treating patients more effectively, they have
lost some money and they can provide very clear evidence of
that.
So one of the pieces of this puzzle will indeed be payment
policies to make sure that institutions that do a better job
for a lower cost don't lose. Because for some institutions,
that's not going to be a sustainable approach.
Senator Bennett. How do we deal with that? Back to my
example of my non-heart attack.
They diagnose me instantly as having an esophageal spasm
and they've lost money. I rejoice. They rejoice. How do we get
some kind of financial incentive into the system to do just
what you've described and say to people that if you do it right
and come up with the right diagnosis, you get a bonus of some
kind?
And what are the implications of that because people would
say, oh, this would be great. That is, somebody who is
disreputable would say, this will be great. I'll tell them that
they don't have any real problem. I'll get the bonus for not
having done the other procedure. They'll walk out of here.
They'll have the heart attack. They'll be back and I get two
dips at the ice cream dish on that basis.
Do you have any ideas?
I agree with what you're saying, but do we have any ideas
practically as to how we can do it?
Dr. Clancy. We're getting there. And it's an area of
intense focus for my agency, for CMS, and for other parts of
the department, including FDA.
How do you create the right incentives? At 20,000 feet, we
would all love to pay for quality. We'd pay more for better
quality care.
It's drilling down to make sure that we do that in the
right way.
What I can tell you, Mr. Chairman, is that we recently
developed a summary of the best evidence that we have, short on
how IT information technology can be part of that solution.
And I'd be happy to submit that for you.
Dr. McClellan. I think that, in building on that, there are
a lot of things that can be done to make the kind of care that
you got even more efficient.
First, we need better treatment so that you don't have to
stay in the hospital 3 days to make sure that you don't have
heart disease. It can be done more quickly.
Earlier this year, for example, FDA approved some new
diagnostic tests for the presence of the enzymes that go along
with a heart attack that make it possible to get patients
determined whether they've got a heart attack or not more
quickly.
It's uncommon for someone to stay in the hospital as long
as 3 days to make sure that they don't have a heart attack.
We need better incentives for payments, as you and Dr.
Clancy have mentioned. A lot of people are concerned these days
about the rising amount of costs that people have to pay out of
pocket.
But that has made some people more sensitive to the overall
cost of care that they're getting, doctors and patients, to try
to work together to find ways to keep those total costs down.
And it would be nice to have added incentives as well to
prevent the diseases in the first place. If there are a lot of
steps that people can take to keep them from getting heart
disease in the first place through a good diet, through regular
exercise, good nutrition, that significantly reduces the chance
of developing heart disease and many other chronic diseases
that are extremely costly today in the first place.
That's the kind of health care system that we need. And the
kinds of incentives that we've been talking about would help us
move in that direction.
Senator Bennett. Now you opened the door to another whole
area, which is the possibility through technology to do
screening and thereby be in a position, A, for preventive care
or, B, maybe a subset of A, counseling, where you could not in
the pre-technology age justify the cost of screening tests for
everybody.
You'd have to wait until you have some kind of symptom
before you run the test because the test is so expensive.
When you've got a screening test that is very, very cheap,
you could go into a school, for example, and screen all the
high school seniors and tell 4 percent of them that they are
going to be at risk for this, that, or the other in their
lives, and they have no symptoms yet.
And, presumably, the long-term benefit of identifying those
that are at risk for a variety of reasons, and then treating it
before the symptoms start out would bring down the long-term
societal cost of health care, would it not?
Dr. McClellan. Well, it would certainly help people get
longer and healthier lives for the same, if not less, money.
I think it would be worthwhile from that standpoint to
encourage the development of these technologies.
You mentioned, you're obviously on top of what's going on
in new medical technology. But there are a lot of technologies
coming along as a result of breakthroughs in genomics and
understanding how gene function works that will potentially
allow us to have much more individualized therapy.
So we can tell for people, not only which drugs or medical
treatments may be indicated to prevent diseases or keep them at
bay based on their specific molecular basis of disease, but
this goes beyond medicine as well.
People are increasingly going to have information about
specific changes they can make in their diet and there are
increasingly going to be foods available that are tailored to
people that have particular nutritional needs to help them
prevent diseases.
So there's a lot of potential there for more individualized
high-value medicine. That's not the kind of health care system
that we have now.
Senator Bennett. No.
Dr. McClellan. Those technologies are not yet in place. We
need to think carefully about how the policies that we're
implementing today might encourage or discourage the
development of that potentially better future.
Senator Bennett. Since this Committee has no legislative
authority, we can go anywhere we want. And that's what we're
trying to do with this hearing, is to get an understanding of
what the ideal health care system might be, which we could then
recommend to the committees that have legislative authority.
And of course, underpinning it all is the overall economic
impact.
Let's take an example that we don't think of as technology,
but that's an example of what we're talking about--inoculation.
We routinely inoculate every child in this country with a
variety of shots. Now, we have some problems in some areas of
the country where the parents or guardians, whatever, don't
bring the children in.
I remember we had this debate at the beginning of the
Clinton Administration when they very appropriately said the
Federal Government ought to finance inoculations for everybody.
All of the concern about the people who are left out, the
uninsured, which has become the shorthand name to describe
those who don't have health care. And the government is going
to pay for all this.
And then we discovered, somewhat to our chagrin, that money
is not the problem, that the inoculations are available
everywhere to everyone, and the problem is that the parents or
guardians, if there are some--in many cases, there are
neither--don't bring the kids in to be inoculated.
The technology is there, but in this case, it's not used.
They don't have access. They don't take advantage for a variety
of socio-economic and other reasons.
But that's an example. Let's just set the non-participation
issue aside for just a minute. That's an example of where the
cost has come down so low, that society can afford to fund a
100 percent participation. And we've stamped out smallpox.
We've stamped out a lot of the things that were normal when I
was a little kid growing up.
We've done it universally. It is a form of universal health
care, to pick a phrase.
Could the day come when stepping up from that level to
screening tests and diagnostic examinations would be universal
in the same fashion, and be administered through the school
system and produce the kind of economic benefits that come from
the fact that we no longer have the epidemics of many of these
diseases that have been taken care of through vaccination?
In this case, there wouldn't be a vaccine. There would be a
treatment. There would be a tailored drug, the kind of thing
that you're talking about.
Is that something that we can envision and maybe drive
toward as policymakers down the line? I understand that there
are going to have to be all kinds of cost studies and
examination.
But is that an idea to which we should aspire or is that a
stupid idea that we should forget?
Dr. McClellan. I think it's a great idea to aspire to. But
I do think that it's a long way off. There is a tremendous
amount of research going on now in terms of what kinds of
impacts the latest genomic sciences have for patient care.
But the problem is that we really don't know a lot of the
answers yet. Virtually every pharmaceutical company and biotech
company is now doing extensive testing of all of their
compounds in development on what are called micro-rays--chips
that have literally hundreds, if not thousands, of genes on
them, to see how the genes are up-regulated or down-regulated.
And these are genes that might be involved in disease
processes like cancer, heart disease, or genes that might be
involved in toxicities from drugs, like liver enzymes or
something like that.
So we're getting a lot of information in now. The problem
is we don't have much translational research yet to tie what
happens with these gene expressions to what it actually means
for a patient's outcomes, for impacting the course of the
disease or determining whether or not a treatment would be
harmful to an individual patient.
And that's what I meant when I talked earlier about a lot
of research going on more than ever before in biomedicine that
is moving in this direction of a more individualized, highly
effective health care system.
But we don't yet know, we're not yet there and we're still
a ways away.
One of the main things that we're focusing on at FDA is to
try to make that, what could be a long process and a costly and
uncertain process, more certain and less costly.
But even if we get those technologies developed, and that's
iffy at this point, there needs to be financing mechanisms in
place, incentives in place to encourage the adoption of these
more individualized treatments rather than the one-size-fits-
all policies.
Senator Bennett. Well, we've examined a whole series of
what-ifs here, and I appreciate your willingness to take this
journey with me.
Commissioner McClellan, I understand that you have to leave
at this point.
Dr. McClellan. Thank you.
Senator Bennett. We've probably reached the point of
diminishing returns in our speculation as to what might happen.
Let me thank you both very much for being with us today.
And if you have any additional thoughts that this conversation
may have triggered, we'd appreciate hearing from you and we'd
be happy to make them part of the record.
Dr. Clancy. Thank you. We'd be delighted.
Dr. McClellan. We're going to, obviously, keep working
closely together on many of these issues and would hope to be
able to keep in close touch with you as well.
I've learned a lot from this session and maybe the most
important thing is your good cardiology report.
Glad to know that you'll be up there for quite a while
working with us on this.
Senator Bennett. Mitch McConnell gave us all a scare when
he took his stress test and ended up having a triple bypass.
He's 10 years younger than I am and said, ``You'd better
have one.'' I went to the same place where he had his and they
said, ``you don't need to come back for another 10 years.''
Dr. Clancy. Well, I was going to say, I would agree with
Dr. McClellan's comments and also say that it's really unusual
to be told that we don't need to see you for 10 years. That's
about the highest approval that you could get.
Senator Bennett. Yes. Thank you both very much.
Our second panel will provide further insights on health
care innovation. We're privileged to have Dr. Peter Neumann,
who is the Associate Professor of Policy and Decision Sciences
at Harvard School of Public Health, and Dr. Neil Powe, Director
of the Welch Center for Prevention, Epidemiology, and Clinical
Research at Johns Hopkins Medical Institution.
We have Harvard and Johns Hopkins. The only thing that's
missing is the University of Utah.
[Laughter.]
But at least we have two of the three.
[Laughter.]
We very much appreciate your both being here. We welcome
your thoughts on the challenges. And we'd be happy now to hear
from you in your opening statement, and then continuation of
the dialog that we had with the first panel.
Professor Neumann, let's start with you.
OPENING STATEMENT OF DR. PETER J. NEUMANN, DEPUTY
DIRECTOR, PROGRAM ON THE ECONOMIC EVALUATION
OF MEDICAL TECHNOLOGY, HARVARD SCHOOL
OF PUBLIC HEALTH
Dr. Neumann. Well, thank you very much, Mr. Chairman, for
your invitation to speak before this Committee on the topic of
technology, innovation and their effects on cost growth in
health care.
My name is Peter Neumann. I'm Associate Professor of Policy
and Decision Sciences at the Harvard School of Public Health.
I would like to speak today about how we can better
understand the value or cost-effectiveness of medical
technology.
Broadly speaking, medical technology contributes to growth
in health care expenditures, as we've been hearing.
But as we've also heard, this research says nothing by
itself about the benefit side of the equation. As we consider
medical technology, it is important to address not just how
much medical technology contributes to health costs, but
whether the investments in medical technology are worth the
health benefits produced.
We would all like to get good value for our money when we
pay for new drugs, devices and procedures. How do we get there?
What tools do we have to use and what policy options are
available?
Formal economic evaluation can help us answer these
questions.
The field of economic evaluation of health and medical
interventions has been an active area of research in recent
years. It includes cost-effectiveness analysis, which shows the
relationship between the total societal resources used, the
costs, and the health benefits achieved, the effects for an
intervention compared to an alternative strategy.
Often, a standard metric such as life-expectancy or
quality-adjusted life expectancy, is used as the measure of
health benefits.
In part, with funding from the Agency for Health Care
Research and Quality, my colleagues and I have compiled a list
of over 1500 cost-effectiveness ratios covering a wide variety
of medical technologies and public health strategies in many
disease areas.
More information is available on our website.
These data underscore several important points about the
cost-effectiveness of medical technology.
First, a great deal of information on the topic has become
available to policymakers in recent years. Unlike many
unsupported assertions about the cost-effectiveness of drugs
and other medical technology, these studies quantify costs and
health effects using data and a standard, well-accepted
methodological technique.
Second, according to peer-reviewed articles, many
technologies are indeed cost-effective. Examples include
warfarin therapy to prevent stroke in those with atrial
fibrillation, immuno-suppressive drugs for those with kidney
transplants, and treatment with mood-altering drugs for those
suffering from depression.
These interventions provide good value in the sense that
they produce health benefits for relatively little cost, or may
actually save money for the health care system, despite their
sometimes high pricetag.
Third, cost-effectiveness does not mean cost savings. Over
the years, people have sometimes confused these terms. But
restricting the term cost-effective to cost-saving
interventions would exclude many widely accepted interventions
which do not save money, but are cost-effective in the sense
that their additional benefits are worth their additional
costs.
A related point is that a critical aspect of any medical
technology's cost-effectiveness involves the manner in which
the question is framed. A technology is not intrinsically cost-
effective or cost-ineffective.
It is only meaningful to say that a technology is cost-
effective compared to something else.
A drug prescribed to lower an individual's blood pressure
may in fact be cost-effective compared to the option of no
treatment, but not necessarily when compared to an alternative
intervention such as an intensive program of diet and exercise
or other medication.
Similarly, claims of cost-effectiveness often depend on the
population under investigation.
For example, statin drugs used to lower an individual's
cholesterol have been found to be relatively cost-effective as
secondary prevention in persons with existing heart disease,
but considerably less cost-effective as primary prevention.
Well, does anyone actually use cost-effectiveness analysis?
Logically, cost-effectiveness analysis should be used by
private insurers and state and Federal policymakers. However,
many payers, including Medicare, have shied away from using
cost-effectiveness analysis in coverage and reimbursement
decisions.
But why?
Cost-effectiveness analysis promises to inform decisions
and enhance population health in an explicit, quantitative, and
systematic manner. Medical journals, including the most
prestigious ones, routinely publish cost-effectiveness
analyses.
Furthermore, many other countries have incorporated cost-
effectiveness analysis into their policy decisions.
How do we explain this paradox?
Studies point to a couple of explanations. Some of them
fault the methodology itself. But, in fact, most experts agree
on the basic tenants. Instead, the opposition more likely
relates to the hardened American distaste for explicit
rationing.
This is understandable, perhaps, but still, how do we get
good value in face of this opposition?
I would offer five observations as we look ahead.
First, cost-effectiveness analysis should not be used
rigidly. Leaders in the field have always warned against using
cost-effectiveness analysis mechanically, but experiences teach
us that rigid use of cost-effectiveness analysis will be
resisted.
Expectations for cost-effectiveness analysis should be more
modest. Cost-effectiveness analysis should inform decisions,
not dictate them.
Second, cost-effectiveness analysis will probably not save
money. Cost-effectiveness analysis should not be conceptualized
or promoted as a cost-containment tool, but rather, as a
technique for obtaining better value.
Paradoxically, using cost-effectiveness analysis may
actually increase health spending because it often reveals
under- rather than over-treatment.
Third, how you say it probably matters.
Research shows that physicians understand that resources
are limited, but they are not willing to admit to rationing.
Similarly, health plan managers deny that they ration care,
but admit that their budgets are constrained. These responses
are instructive. It suggests that the term ``cost-
effectiveness,'' may be part of the problem. We might instead
use terms such as ``value analysis'' or ``comparability,''
rather than ``cost-effectiveness analysis'' and ``rationing.''
Context also matters.
Cost-effectiveness analysis may be acceptable to guide
choices on how frequently to screen for certain diseases. It
may not be acceptable to guide choices for those in need of
life-saving treatments.
Fourth, incentives matter.
Debates about the use of cost-effectiveness analysis cannot
be separated from debates about the underlying health system
and the incentives it embodies.
Cost-effectiveness analysis is sometimes opposed because it
is is used centrally by a single decisionmaker. How to
reconfigure incentives in the system is a related but somewhat
separate and still critical challenge.
Fifth, the final message involves the importance of
thinking expansively about applications of cost-effectiveness
information.
Cost-effectiveness analysis should not simply focus on
medical interventions, but more broadly, on interventions to
improve health by reducing environmental exposures, injuries at
home and in the workplace, and motor vehicle accidents.
In closing, let me emphasize that whether medical
technology offers good value is a question that can be best
informed by careful analysis.
I would encourage the judicious use of cost-effectiveness
analysis in the years ahead.
Thank you very much, Mr. Chairman, again for your
invitation and I'd be pleased to answer any questions you have.
[The prepared statement of Dr. Peter Neumann appears in the
Submissions for the Record on page 53.]
Senator Bennett. Thank you very much.
Now am I pronouncing your name correctly, sir?
Dr. Powe. ``Po.''
Senator Bennett. ``Po.'' Very good. Thank you for being
with us and we'd appreciate hearing your testimony.
OPENING STATEMENT OF DR. NEIL R. POWE, DIRECTOR,
THE WELCH CENTER FOR PREVENTION, EPIDEMIOLOGY,
AND CLINICAL RESEARCH, THE JOHNS HOPKINS
MEDICAL INSTITUTION
Dr. Powe. Good morning, Senator Bennett. I'm a general
internist, a clinical epidemiologist, and a health services
researcher. My research has assessed the clinical and economic
impacts of biomedical innovation in medicine.
It examines the impact of new and established technologies
on patients' longevity, functioning, quality of life and, of
course, cost. I've conducted cost-effectiveness studies of
technologies in several areas of medicine and I've attempted to
do so with equipoise.
I've also studied physician decisionmaking and other
determinants of the use of medical technology, including
payers' decisions about insurance coverage for new medical
technologies and the impact of financial incentives on the use
of technology.
New medical technologies include drugs, devices, procedures
and the systems in which we, as medical professionals, deliver
them. They include so-called ``little-ticket'' technologies
which cost relatively little individually, but when used at
high frequency, can become expensive. One such emerging
``little-ticket'' technology is the C-reactive protein
laboratory test for detecting inflammation now being debated as
a useful technology for detection of heart attack risks. ``Big-
ticket'' technologies such as body scans and organ
transplantation have high individual price tags and can
generate high costs, even when used relatively infrequently. In
theory, a new medical technology can increase costs, have
similar costs or decrease costs relative to an existing
standard technology. Evidence to date suggests that much of new
biomedical innovation increases cost to the health care system,
especially in the short-term. ``Little-ticket'' or ``big-
ticket'' technology should not be judged based simply on costs.
The more important question that I'd like to address is what is
the technology's value?
Value is commonly seen as the benefit that's derived
relative to the cost. In theory, a technology can produce
benefit relative to the existing standard if patient outcomes
are better. On the other hand, it can produce no benefit if
outcomes are similar or even produce harm if patient outcomes
are worse. High value occurs when substantial improvement in
patient outcomes occurs at a reasonable cost.
Americans believe in the concept of value and understand
it. For example, they're willing to pay more for many things--a
particular type of clothing, food, service, house, automobile--
because they believe that the utility that's derived from the
purchase is worth the higher price. Cost is a relevant factor,
but value is paramount, so much so that medical technology
needs to be judged in the same way.
Twenty-five years ago, the science of assessing value in
medicine was rudimentary and underdeveloped. Many of the tools
that Dr. Neumann talked about for assessing value were first
applied to health care in the late 1970s and early 1980s. These
include patient outcomes research comprising clinical trials,
evidence synthesis and cost-effectiveness.
These have undergone refinement by researchers at
universities around the country. Much of the work has been
catalyzed and funded by the Agency for Healthcare Research and
Quality. These researchers have sought to create rigorous
standards of high quality research for value science.
Despite the maturation of and demand for the science of
values, its impact has been limited for three reasons.
First, there is an unprecedented number of new technologies
now entering the healthcare marketplace. These include
minimally invasive surgery, as you mentioned, the
transplantation of hearts, lungs, livers, kidneys,
biotechnology drugs, indistinguishable from natural hormones
for patients with congenital or acquired deficiencies, dialysis
therapy for end-stage kidney disease, automatic implantable
defibrillators and cardiac resynchronization devices to bring
life to those with life-threatening arrhythmias and heart
failure.
Knowledge of the structure and function of the genes and
proteins is advancing rapidly and the future will yield even
more promising technologies we never imagined for identifying,
preventing, and treating acute and chronic illnesses.
However, the level of funding for high-quality and unbiased
value assessment pales in comparison to the explosion of new
biomedical innovations.
To the public, payers, and providers, the entry of new
medical technologies into the practice of medicine now seems
like a series of intermittent ``surprise attacks'' on the
pursestrings of American health care. It has been suggested
that less than a fifth of all practices in medicine are
subjected to rigorous evaluation and still less receive an
adequate assessment of the cost consequences in addition to the
clinical consequences.
We are likely to witness a continuing salvo of surprise
attacks in the coming years without adequate funding to do
early, comprehensive, balanced and rapid assessments.
In a study with researchers at AHRQ, I found that medical
directors making coverage decisions for new medical
technologies at private health care plans across our country
are impeded in their decisions because of the lack of timely
effectiveness and cost-effectiveness information. There is
considerable trepidation to decide against covering potentially
useful technology without adequate evidence.
Likewise, there is a concern about making a coverage
decision in favor of a technology that might later be shown to
have minimal benefits at a large cost to society. The
preference of those making decisions about coverage and payment
for technology was for high-quality outcomes research funded by
authoritative government entities.
Early assessments of clinical and economic outcomes could
be accomplished with investment of a small fraction of annual
health care expenditures on value assessments. The payoff would
be substantial.
For example, contrary to relentless, direct-to-consumer
advertising for body CT scans to detect occult disease, my
colleagues and I recently found that screening smokers for lung
cancer with helical CT scans is unlikely to be cost-effective
unless certain conditions are met.
The high number of false positive lung nodules detected by
the scans can potentially lead to more harm from invasive and
costly surgical procedures.
Early assessments such as this, which include primary data
collection, secondary data collection, data synthesis, and
sometimes modeling and forecasting will secure information for
the American public and its policymakers in a timely fashion
needed to prevent premature dissemination of costly technology
with little or no value.
The Agency for Healthcare Research and Quality, as well as
the National Institutes of Health, could act as a focal point
to bring the best team of value researchers in the country to
attack these issues by performing clinical effectiveness
trials, observational studies, cost-effectiveness analyses, and
meta-analyses.
If introduction of some new technologies does not decrease
cost, at least through generation of better and more timely
information, Americans can make sure that what they are
purchasing provides good value for the dollars they spend.
Early assessments are particularly important given rising
numbers and costs of pharmaceuticals, current consideration of
a Medicare prescription drug benefit and use of tiered pricing
arrangements in the private sector to control drug spending.
Tiered pricing is a mechanism to allow consumers choice in
particular drug treatments when they believe one drug has value
over another. However, they must pay more when choosing to use
a more expensive medication.
Placement of a pharmaceutical into a particular tier and
patient decisions to buy and use it are dependent on unbiased
information about the benefits and the costs of the
pharmaceutical relative to the benefits and costs of competing
medications. That is, relative value.
Second, as a corollary, funding for career development of
value scientists needs substantial bolstering to expand the
cadre of people with the capability to perform such research.
Far too few physicians and other health care professionals
and scientists have the necessary training to understand and
produce value science that integrates clinical and economic
issues.
Third, understanding how technologies affect cost and value
involves an understanding of the barriers to decisionmaking for
health care providers. Barriers to optimal decisionmaking can
lead to technologies being overused, underused, or misused.
My colleagues and I performed a study of the factors
affecting physician decisionmaking with regard to adherence to
clinical practice guidelines. We found that there is a process
that must take place for a new technology to become routine
standard practice.
Physicians must be aware that a new technology exists,
agree that it has value, be willing to try it--that is, adopt
it--and then they must adhere to its use.
Lack of awareness leads to underuse. Underuse of an
effective technology can lead to higher expenditures in the
future.
For example, if physicians were not aware that in patients
with diabetes, urine protein screening for detection of occult
kidney disease and application of ACE inhibitors can delay or
prevent expensive dialysis treatment at greater than $50,000
per patient per year for end-stage kidney failure, they might
never employ the strategy in their practice.
Fortunately, methods of communicating new information to
clinicians are improving through rapid summary publications,
clinical practice guidelines by professional societies, and
dissemination through electronic means. Ways for helping them
acquire and assimilate new information are needed.
If aware of a technology, physicians must agree with the
evidence that a technology is effective or safe. If high-
quality evidence on representative patient populations is not
available, physicians may disagree on whether the technology
provides benefit.
We studied how early assessments, released through brief
clinical alerts that were not comprehensive, influenced the use
of carotid endarterectomy. We found that clinicians may
extrapolate research findings to populations without clear
evidence and indications. Value science can provide clear
evidence.
Awareness and agreement are necessary for appropriate use
of technology, but insufficient. Even being aware and with
strong evidence of effectiveness, physicians may not adopt
innovations if there are administrative barriers to its use or
the lack of self-efficacy.
They may also adopt technologies with little benefit if
payment policies that we talked about and heard before
prematurely promote a technology's use.
Financial incentives in payment policy influence both
adoption of and adherence to use of technologies.
Thus, proper use of new technologies means that the
physicians who apply them and the systems into which they are
placed are adequately configured and incentivized to make
optimal use of the technology.
To this end, there's a need for more behavioral and systems
research that studies how biomedical innovation from
laboratories is optimally and rapidly translated into
interventions to improve the health of patients treated at
hospitals and physicians' offices.
In conclusion, biomedical innovation has brought the United
States new, unprecedented medical advances that save and
improve the quality of patients' lives. We need to continue to
encourage biomedical innovation. But we must recognize that for
many health conditions, technologies will bring higher rather
than lower absolute cost.
Cost is relevant, but value is far more important.
We need to protect biomedical innovation and America's
purse by furthering the science of assessing value in medicine.
Strengthening our nation's capacity to perform value
science will help private and public payers in this regard and
provide information that physicians and consumers of medical
technologies need to make decisions about their care.
The American people cannot afford to have technology used
unwisely. A fraction of health care expenditures in the U.S.
should be targeted to the value science of medical care.
Thank you for the opportunity to address you today and I
would be happy to entertain any questions you have.
[The prepared statement of Dr. Powe appears in the
Submissions for the Record on page 56.]
Senator Bennett. Thank you very much, both of you, for your
thoughtful presentations.
You listened to the first panel. Were either of you anxious
to break in with something that you really wanted to say and
straighten out any of the conversation that we had in the first
panel?
Dr. Powe. Well, I'd like to comment on your experience with
the esophageal spasm, which I thought was interesting. Your
question about whether there might be some innovation in the
future, a test in the future that might have prevented the
sequence of events that you went through. And in fact, I think
that there are likely to be technologies that do that.
One of the problems is that there may be 50 technologies
that are tried out before we get it right. And what that means
is that as we experiment and use those technologies, we don't
know what effect each one of them is going to have on the
system.
Senator Bennett. That means high cost at the front end,
but, presumably, you end up with one that means low cost at the
back end.
Dr. Powe. Right.
Senator Bennett. And we're not seeing the low cost yet.
We're still getting all of the high-cost front-end stuff.
Dr. Powe. Right. And while we're trying them out, the
typical situation is not that one technology will supplant
another technology, but that it will add on in the process as
we learn how to use it. Then maybe, in fact, later on, it may
supplant another, our existing technologies.
Senator Bennett. Yes. Did you have anything?
Dr. Neumann. I just wanted to add one----
Senator Bennett. Don't worry about my spasm because it's
fine.
[Laughter.]
Dr. Neumann. OK.
Senator Bennett. But the conversation, generally.
Dr. Neumann. Yes, and I'm very glad to hear that.
Senator Bennett. Yes.
[Laughter.]
Dr. Neumann. I agree with much of what was said earlier
today.
One issue that perhaps was alluded to, but I would just
amplify on, and I think it's an important policy lever, is
funding for value research, cost-effectiveness research.
We spend a lot of money, and a lot of it is well spent, on
biomedical research. We spend really, as Dr. Powe said, very
little on health services research and value research.
Senator Bennett. Maybe part of our problem as we address
this as policymakers, coming through in your testimony and in
the first panel, is that we're looking down a stovepipe.
Let me give you another example out of real life.
Merrill-Lynch hired a doctor to come lecture to all of
their brokers. That's a pretty good contract to have if you're
the doctor. You get to train every Merrill-Lynch broker.
They did that. Every new broker hired at Merrill-Lynch had
a session with this doctor. They said, ``We decided to pay that
cost, an increased training cost, because we suddenly realized
how many of our brokers were dropping dead from stress, having
heart attacks and literally dropping dead. We figured out how
much it cost us to replace them in terms of training,
experience, et cetera.''
A manager of a large Merrill-Lynch office told me--now this
was before the days of online trading and all the rest of this,
this was at a time when everything was done in a particular
office--``If we have a branch manager drop dead from a heart
attack, that's hundreds of thousands of dollars that we have
invested in him''--and it's usually a him. ``And if, with tens
of thousands----
[Pause.]
Senator Bennett. [Continuing.] Excuse me. The spasm may be
fine, but my cold is still here.
``If with tens of thousands of dollars training people on
diet, stress management, et cetera, up front, we can save the
hundreds of thousands that would come from having these people
die, it's worth it.''
Now, we are focusing entirely in public policy discussions
on the cost of health care premiums, the cost of Medicare. And
I think the government is dealing with the cost of Medicare in
exactly the wrong way by simply saying that we will arbitrarily
pay only X-percentage of this.
But that's another debate and I said at the outset, we're
not going to get into that debate here.
But I think when we're talking about value and costs, we're
saying the country as a whole and the economy as a whole is
better off, even if we're seeing an increase of 15 percent per
year in cost, because of the increase in productivity, the
increase in contribution, et cetera, of the lives that are
saved. And we don't figure that into the conversation.
Now it does get figured in, wearing your economist's hat
for just a minute--you both say you're not economists, but the
Commissioner was.
Let's look at health care cost as a percentage of GDP, and
say if the health care cost does not grow more rapidly than the
GDP grows, we're fine, because we're getting the benefit of
increased GDP.
Now, as soon as you do that, you're at 3 percent. And
nobody's going to bring health care costs down to 3 percent.
But if we could find some calculation that says the
contribution to GDP is 12 percent per year by virtue of what we
get, then we could say, society as a whole can justify this
kind of an increase every year.
The problem for the employer is, the individual employer,
he's getting hit with that 15 percent compound every year and
he does not see the benefit in his employee pool because most
of the benefit is coming for the retired. And he says, I can't
sustain this any longer.
Just react to that and give me what you think the real
value to society is from--to pick a number for sake of
conversation, a 15-percent per year compounded increase in
health care costs, which is enormous in terms of the burden
that it puts on employers--and taxpayers.
Dr. Neumann. Right. Well, you raise many very good points.
I guess I would say a few things.
One, I think we in the academic community, maybe society at
large, feel frustrated because we see increases in health care
spending. We see the 15 percent. We see the 14 percent of GDP.
And there's a feeling that perhaps we're not getting the
benefits, the health gains we should be getting, or perhaps we
don't have the tools to measure it very well.
I think some research that Dr. McClellan referred to by
economists in recent years have begun to document better, that
in fact, the additional spending on health is resulting in
measurable gains that are worth paying for, gains in terms of
increased life expectancy, gains in terms of decreased
disability rates, and so forth.
Now, even if we can get ourselves there, there's still the
other problem you referred to, which is employers now feeling
that they are getting those gains, that the gains are coming to
retirees later in life and so forth.
I think that's a real problem and challenge.
Part of it may be a measurement issue, that in fact,
productivity is growing and the employers genuinely are getting
returns and we just haven't been able to measure them very
well.
But part of it may be kind of a structural issue, that they
are being asked to pay for health gains that occur later in
life.
And that's a challenge.
Dr. Powe. I would agree that some of the benefits or value
is coming in terms of longevity and improved quality of life.
And at least for the working population in terms of increased
productivity--and they've been measured in many types of the
studies that we've talked about.
The issue of the pressure from retiree health expenditures,
I think, is a vivid illustration of what Dr. McClellan was
talking about, about incentives not being aligned correctly
within our society with regards to health care.
So I think that's going to take an alignment of incentives
in order to have everything work in concert to address that
issue.
Senator Bennett. Do either of you have an opinion as to
whether or not Federal laws and regulations distort the
creation or the use of new technologies and innovations in
health care? Or do they encourage?
Dr. Neumann. I guess I would say two things.
One, they probably do both and there are probably certain
incentives in the system that encourage and some that
discourage.
One issue that I've been doing some thinking and writing
about is Medicare and cost-effectiveness. Medicare hasn't
formally incorporated cost-effectiveness analysis into its
decisionmaking process, despite attempts to do so, and even
attempts to offer proposed regulations that would allow it to
do so.
It doesn't seem to be a statutory issue. That is to say,
Medicare's statute says that it will pay for reasonable and
necessary services in some categories.
That seems to allow it to use cost-effectiveness analysis.
It has never been able to get there because of larger issues,
fear of rationing and so forth.
That may have to change with legislation eventually.
Dr. Powe. I think that our regulations and laws actually
strike an appropriate balance, checks and balances within the
system.
In some sense, we have patent regulation that promotes
biomedical innovation because it provides a period of time
where competitors cannot come in and sell particular products.
So I think that that promotes biomedical innovation and
it's probably a necessary thing that we have.
But, on the other hand, then we have an approval process
for drugs that the Food and Drug Administration on the back end
says that you have to show efficacy and safety.
So that there's checks and balances there.
On the coverage side, whether things are paid for, I think
that things are a little bit more helter skelter, both in
regards to public and private coverage of new medical
technologies.
And we don't really have a uniform system of criteria for
doing that. I think that some of the points that we're trying
to make is the use of value science to actually help in that
process to develop the kinds of criteria that should be used to
cover technologies.
Senator Bennett. Getting back to one of the questions that
probably I should have asked the first panel, but you have just
as much expertise in here.
One of the statements that is made is that once a provider
invests in new equipment--say he buys an MRI machine--he then
feels he has to use it, even if, medically and diagnostically,
it isn't necessary because he has to get his money back by
running as many tests as possible through it and charging.
Then another provider decides he has to buy an MRI machine
if he's going to compete and you have overcapacity and then,
ultimately, overuse.
Again, let me describe a brief experience that I had when I
was on the campaign trail the first time and visiting a
hospital. I'm a businessman by background. I innocently asked
these people if they had an MRI machine. This is when MRIs were
relatively new.
And they looked at me like I was an idiot--which I was--and
they said, ``Well, of course. We have to have an MRI machine.''
And I said, ``What's your usage?'' And they said, ``30 percent
of its capacity.''
And, being a businessman, I said to them, ``Well, the
marketplace is trying to tell you something. It's trying to
tell you that you don't need an MRI machine. Maybe we ought to
do something at the Federal level to change the anti-trust laws
in such a way that you could make a deal with the hospital down
the street. If you're only using their MRI machine at 30
percent, then you could contract that you'd send your stuff
there.''
Again, they looked at me like I was a bigger idiot and
said, ``The marketplace demands that we have an MRI machine.''
And I said, ``No. If the marketplace was demanding that you
have an MRI machine, you'd be getting 100-percent
utilization.''
We talked past each other on this issue until the light
suddenly went on in my mind, that when they were talking about
the marketplace and their customers, they were talking about
insurance companies. When I was talking about the marketplace,
I was talking about customers--that is, patients.
Insurance companies told them, ``We will not approve
patients being sent to your hospital if you don't have an MRI
machine.'' So they really didn't care whether it was utilized
ever, as long as it was on the premises so they could certify,
we have an MRI machine.
When I was talking about the marketplace, selling things to
customers, I was talking about the number of customers walking
through the door. The patients walking through the door weren't
their customers. The insurance company paying the bills was
their customer.
Now, do we have, in fact, an overcapacity built in by
virtue of the third-party payer system and then a sense of we
have to get our money back by running tests through this
machine that is part of the escalating health care situation
created by technology?
Dr. Powe. Well, that's a tough question to debate.
Senator Bennett. That's why the staff came up with it.
[Laughter.]
Dr. Powe. I guess one solution would be to do away with
insurance companies and have patients pay. But I think that
that would cause even greater problems by having many people
pay out of pocket.
There are safeguards that insurance companies do on the
other hand in terms of making sure that the populations they
serve receive the very best care and can do that because they
have large purchasing power and can actually influence the
quality of care that providers provide.
So I'm not sure that we ought to go to a system in which
patients would pay directly out of pocket like we do in other
areas, other sectors of the economy.
Dr. Neumann. I would agree with Dr. Powe, especially with
the notion that that's a very tough question.
But I guess I would say two things.
One, in some sense it's an evidence question, it seems to
me, and Dr. Clancy talked about this. As we move as a culture,
as medical establishment toward evidence-based medicine, one
hopes that tests that are ordered will be tied to studies that
shows that they're appropriate, that they work for the patients
they're being given to.
And part of the problem that you identified might be
addressed with better evidence.
But another issue, and perhaps more important issue, is the
structure of the marketplace and the incentives that it
embodies.
Why the situation exists in the first place, that the
hospital felt the need to buy the MRI in this case. And that's
a very complicated question.
We create insurance because it solves a problem. But it
also creates some other problems.
We can talk about establishing the structure of the
marketplace and the incentives, changing incentives, but that's
a very large debate.
Senator Bennett. Yes, and it's a debate I'll have in
another forum at another time.
But, Dr. Powe, I'd be happy to talk to you about this. I
think getting rid of the third-party payer for routine activity
would actually be beneficial.
I think insurance ought to be for catastrophic events.
Health is the only place where insurance pays for routine
activities. If I have car insurance, it doesn't pay me for
changing the oil. It only pays me when I'm in a wreck.
And I think health insurance should only pay me when I have
a serious health problem. I should not necessarily have to file
an insurance claim for a routine kind of test. But that's
another debate for another day.
Dr. Powe. In fact, I think we're evolving to a system in
that way. I think most people when they see a physician have to
pay something out of pocket today in terms of coinsurance and
deductibles.
Senator Bennett. Yes.
Dr. Powe. So the cost is not entirely free for non-
catastrophic services today.
Senator Bennett. Do either of you know the extent to which
Medicare or the veterans health program, as well as private
insurers, are already using some kind of cost-benefit analyses
to make judgment calls?
And if so, how do they do it? Or is this a brand new idea
that still hasn't caught on?
Dr. Neumann. Well, for the most part, Medicare has not.
There are a couple of exceptions over the years. They've added
some preventive services that were informed by cost-
effectiveness analysis. But they haven't formally incorporated
into their coverage process.
Despite some attempts to do so, they haven't been able to
for various reasons that probably have more to do with politics
and fears of rationing than they do with statutory limitations.
My sense is that the VA does use it on occasion to inform
decisions about which drugs go on formulary, which drugs go on
first-line treatment, second-line treatment, and so forth.
But, again, my sense is it's limited there as well.
Senator Bennett. What about private insurers?
Dr. Neumann. Well, that's I think a difficult question to
answer in the sense that surveys--and Dr. Powe has done some of
these--surveys of health plan managers, medical directors, that
ask them, ``Do you use cost-effectiveness analysis?'', often
yield a response, ``No, we don't use it.''
We look up clinical evidence and base our decisions on
clinical evidence.
I think if you drill down a bit, though, it becomes clearer
that cost-effectiveness evidence and other economic information
does inform decision.
And in some sense, perhaps they're not willing to admit it
because they're afraid of admitting to the rationing. And in
some sense, I think it's almost an indirect piece of evidence
that they use because they've read a journal article or they
adhere to a clinical guideline that has, in fact, used cost-
effectiveness evidence.
Senator Bennett. Dr. Powe.
Dr. Powe. I would concur with Dr. Neumann. The entities
that you mentioned are making value judgments. They may not
formally be using cost-effectiveness analysis in the formal
sense that researchers might, but they are making value
judgments and using the components and the logic of the science
of value in making those judgments.
They may not call it cost-effectiveness analysis as such.
Senator Bennett. Well, ultimately, who should decide
whether the additional cost of a new drug or a new medical
device is worth it? The provider? The drug company? The
government? NIH?
Dr. Powe. I think we all should.
Senator Bennett. When everybody decides, then nobody
decides.
Dr. Powe. Right. But I think individuals have to decide
when they cost-share in medical care, so they have to know the
value of the treatments that they might pay for.
I think that the Medicare program needs to know the value
of the treatments that they're paying for. I think private
insurers need to know the value.
So I think we all are in this together. We do it for
different things. But what will help us all is better
information on what value technology has.
Senator Bennett. A vote has just started, and we've been
going for 2 hours. I would love to continue this dialog, but I
think we probably will close.
Let me thank you both again for your being here and your
willingness to share your expertise, and invite you, if either
of you come across anything that you think would help inform
the issue that we're addressing here, to send it on in to the
Committee.
Again, thank you all.
Dr. Neumann. Thank you.
Dr. Powe. Thank you.
Senator Bennett. The hearing is adjourned.
[Whereupon, at 11:30 a.m., the hearing was adjourned.]
Submissions for the Record
=======================================================================
Prepared Statement of Senator Robert F. Bennett, Chairman
Good morning and welcome to today's hearing on how technology and
innovation affect health care costs.
The United States has a health care financing problem that goes
well beyond the budget challenges posed by Medicare. For many years,
our health care spending has grown at a significantly faster rate than
the economy, and projections indicate that this will continue. Any
financial arrangement where expenses grow significantly faster than
income is truly on very shaky ground.
In other sectors, new technologies usually lead to greater
efficiencies and lower costs, yet it is unclear whether the same is
true for health care. What's different about health care? Is it the
technology or the way we pay for it?
How can we strike the right balance--providing access to the latest
breakthrough technologies, while limiting an open-ended raid on the
public and private treasuries that fund our health care?
During this hearing, we will explore these issues, bringing
together some of the best minds from the public and private sectors to
help shed some light on this situation.
We should first question whether technology and innovation have
truly added to health care costs, as some claim, or have reduced health
care costs through enhanced efficiency.
Secondly, we should examine whether new technologies are
disseminated in an efficient and effective manner, and if there are
areas where they are being overused or underused. For example, some
have expressed concern that advanced imaging technologies may be
overused, in part because of poor incentives in the payment formulas
used by Medicare and other insurers. At the same time, an article in
this week's Health Affairs highlights how new technologies may be
underused in treating people who lack health insurance.
We need to find the right balance. We need to judge the cost-
effectiveness of new technologies, so that we properly fund this
critical work, without overpaying and adding additional upward pressure
on health care spending.
Unlike most of the recent congressional debate on health care, this
hearing is not about Medicare or its coverage of prescription drugs.
However, this issue is crucial to Medicare and every other health care
purchaser that faces the dilemma of how to add innovative new benefits
without setting off an explosion of health care costs.
On our first panel, we are privileged to welcome Dr. Mark
McClellan, the Commissioner of the Food and Drug Administration, and
Dr. Carolyn M. Clancy, the Director of the Agency for Healthcare
Research and Quality (AHRQ).
Our second panel will provide further insights on health care
innovation. We are privileged to have Dr. Peter Neumann, Associate
Professor of Policy and Decision Sciences atHarvard School of Public
Health, and Dr. Neil Powe, Director of the Welch Center for Prevention,
Epidemiology, and Clinical Research at Johns Hopkins Medical
Institution.
We welcome each witness's thoughts on the challenges facing health
care today. I want to thank Ranking Member Stark for his interest and
help in organizing this hearing and in bringing these distinguished
experts before the Committee. I ask all of you to join me in a
bipartisan spirit as we engage in this important task.
__________
Prepared Statement of Representative Pete Stark,
Ranking Minority Member
Thank you, Chairman Bennett. I would like to commend you for
holding this hearing on ``Technology, Innovation, and Health Care
Costs.'' It's an important topic that requires serious inquiry.
Finding the right balance between cost consciousness and allowing
access to new, potentially life-saving, innovations is crucial.
Progress and technology often extend and improve lives, such as an MRI
that provides early detection of a tumor or new surgical techniques for
cataract treatment. Yet in too many cases, the latest technology simply
becomes a profit center for hospitals when other, less costly,
treatments would serve patients equally as well.
Smart utilization can spread the benefits of new technology without
substantially increasing health care costs. I believe that providing
the highest quality health services should be our goal--a goal that
cannot be compromised. As we've seen with HMOs, it is too easy to deny
patients access to appropriate care in the name of cost cutting.
Analysis of cost-effectiveness must be mindful of the needs and
interests of the patient.
New drugs and medical devices are not the only advances we need.
Better use of information technology would not only improve care, it
could save lives. An estimated 44,000 to 98,000 Americans die each year
because of medical errors, according to an institute of Medicine study.
This is unacceptable and unconscionable. Many medical errors are
attributed to poor handwriting and other sloppy mistakes. Storing
medical records on IT systems would prevent many of these mistakes--and
deaths--as well as allow for the easy transfer of records when a
patient switches doctors or visits a specialist. The technology is
available, but it is not being fully used.
Cutting-edge medical technology may as well be science fiction for
the 41 million Americans without health care--people without the means
to utilize innovative, and often, preventative treatments available to
those with insurance. Among the uninsured, illnesses and deaths that
may have been avoided if they had access to new technologies for the
treatment of just three conditions--heart attacks, cataracts, and
depression--cost our society more than $1 billion a year. The inequity
in access to health care prevents health outcomes from being as
universally successful as they could be.
Thank you Mr. Chairman and I look forward to the testimony of our
witnesses.
__________
Prepared Statement of Mark B. McClellan, M.D., Commissioner of Food and
Drugs, U.S. Department of Health and Human Services
Good morning Mr. Chairman and Members of the Committee. I am Dr.
Mark B. McClellan, Commissioner of Food and Drugs, and I welcome this
opportunity to testify before the Committee today. As we enter the 21st
century, America leads the world in developing and commercializing new
medical innovations and technologies. From information technology to
biotechnology to materials science, United States (U.S.) scientists and
high technology workers are making new discoveries and developing new
products every day that are steadily improving the quality of our
lives. This progress is critical to our health and our economic
prosperity.
Innovations resulting from breakthroughs in science and technology
fuel economic growth. According to the Department of Commerce, the
information technology sector accounts for just seven percent of all
businesses in the U.S. economy, yet between 1996 and 2000, it drove 28
percent of the overall U.S. real economic growth and created jobs at
twice the pace of other sectors. These jobs paid twice as much on
average as well. Many leading economists now believe that new
discoveries in information technology led to investments over the last
couple of decades that helped account for the historic surge in
economy-wide productivity growth in the 1990s.
background
While all economists appreciate the contribution of such economic
growth to the well-being of the U.S., there is often less appreciation
of the contribution of innovations in biomedical technology. A primary
reason is that technological change in medicine brings benefits in
addition to direct economic gains, including increased longevity,
improved quality of life, and less time absent from work. These
benefits are not taken into account in standard measures of aggregate
economic output. If a country had real gains in its overall health, but
not in its material well being (most often measured by per-capita
income) the national income accounts would not change, even though
those accounts are often thought to measure the well being of a
population.\1\ In addition, the direct economic and public health
benefits of developing important new medicines often takes considerable
time to be realized. If a high-technology firm invents a better memory
chip, the time to get that innovation into products sold in the U.S.
could potentially be as short as a matter of weeks or months.
Regardless of how promising a drug or other new treatment appears in
the laboratory or even in animals, it must undergo extensive clinical
trials before it can be approved as safe and effective for market
introduction.
---------------------------------------------------------------------------
\1\ ``Measuring the Health of the United States Population'',
Brookings Papers on Economic Activity, Microeconomics, 1997, 217-272
(with Elizabeth Richardson).
---------------------------------------------------------------------------
In recent years, economists have tried to quantify the value of
biomedical innovation to society. Some economists actually estimate
that the value of the longer and better lives that have resulted from
translating new biomedical knowledge into steps to prevent and slow
diseases is worth literally many trillions of dollars in better health.
In particular, the value of biomedical innovation to the U.S. equals
the value of innovation in all other sectors of the American economy
combined.\2\ Even with the benefits of new medical technology, the fact
remains that technological innovation is a major source of increase in
real per-capita medical spending in the U.S. Innovations in medicine
can reduce spending on medical care. For example, treatments ranging
from effective care for depression to laser eye surgery are much less
expensive than in years past. But many new technologies result in
increased costs, and in some instances the net effect of overall
technological change has been to raise health care expenditures. First,
when a treatment becomes less expensive and safer (fewer
complications), more patients may decide that a treatment is worth the
risks and unpleasantness. In the early 1980s, relatively few seniors
had cataracts removed because the procedure required an unpleasant
hospital stay, often had complications, and yielded imperfect results.
Today, thanks to improvements in technology, millions more seniors with
more modest visual impairment find that modern cataract surgery
improves their lives. Second, many treatments exist that do things that
simply were not possible before, such as allowing many patients to
survive previously fatal or impairing diseases. Americans spend much
more on transportation today than they did a century ago because of
innovations in transportation ranging from automobiles to airplanes,
allow people to go places they simply could not before. Similarly,
patients with heart disease, cerebrovascular disease, cancer,
arthritis, AIDS, and countless other conditions are living longer and
better lives because of medical innovations that transformed fatal
illnesses or illnesses that could only be treated with comfort measures
into manageable conditions.
---------------------------------------------------------------------------
\2\ Cutler, David M. and Mark McClellan. ``Is technological change
in medicine worth it?'' Health Affairs; September/October 2001: 11-29.
Grabowski, H., J. Vernon, J.A. DiMasi. ``Returns on research and
development for 1990s new drug introductions.'' In ``The Cost and Value
of New Medicines in An Era of Change.'' PharmacoEconomics 2002; 20
(Suppl. 3):11-29.
---------------------------------------------------------------------------
The increased spending on health care does not necessarily reflect
negatively on technological change. While many studies attribute a
large share of the age- and price-adjusted growth in per capita medical
spending in recent decades to technological innovation, a key issue is
whether the benefits of innovation are rising faster or slower than the
costs.
This important question is difficult to answer. It depends on our
ability to determine the value of output from the health services
sector, and putting a value on a longer life or a higher quality of
life is hard to appraise. Nonetheless, a limited number of studies have
attempted to aggregate the medical value of new innovations across the
whole health care economy in general and the drug industry in
particular. Even with these studies, it can be difficult to sort out
whether the observed improvements in health are from medical
technology, or from other factors that may influence health outcomes,
such as higher incomes, improved public health measures, or changes in
behavior as a result of greater biomedical knowledge. To try to
identify the net value of medical technology itself, several studies
have attempted to measure the value of specific kinds of innovations. A
number of studies have examined outcomes for specific illnesses, such
as heart attacks and depression, where the impact of specific changes
in technology can be examined more closely. While none of these studies
are completely convincing in themselves, they generally show that
medical innovation has greatly increased value, that is, the value of
the improved health is far larger than the increase in spending.\3\
---------------------------------------------------------------------------
\3\ Lichtenberg, Frank R. ``Are the Benefits of Newer Drugs Worth
Their Cost? Evidence From the 1996 MEPS,'' reprinted from Health
Affairs, Vol. 20, No. 5, September/October 2001. Kleinke, J.D. ``The
Price of Progress: Prescription Drugs in the Health Care Market,''
reprinted from Health Affairs, Vol. 20, No. 5, September/October 2001.
Lichtenberg, Frank R. ``The Effect of New Drugs on Mortality from Rare
Diseases and HIV.'' NBER Working Paper W8677, December 2001.
---------------------------------------------------------------------------
The reasons are quite intuitive. Individuals are living longer and
better lives, because our nation is making real progress in the quality
of medical care for many conditions. While the achievements of health
improvements in past decades have been impressive, recent progress in
genomics, proteomics, nanotechnology, information technology, and many
other fields promise even greater improvements in our lives in the
years ahead.
We achieved the improvements of the last few decades without a
sophisticated science of genomics--the human genome was sequenced in
just the last few years. Genomically-based drugs, and gene and tissue
therapies based on genomic sciences, are making up a growing number of
the new drugs entering clinical trials. We also achieved our recent
progress without the new science of proteomics, and an increasingly
sophisticated understanding of how gene and protein expression interact
to cause disease in individual patients. We also did it without a new
generation of increasingly powerful biomedical tools based on the
latest information technology that can enable sophisticated systems for
supporting effective medical decision-making. These additional tools
increase the future potential for more effective, more targeted, even
individualized medical treatments that can cure or at least slow or
halt disease progression.
impact on health care costs
As health care costs have gone up, it is increasingly important to
make sure we are realizing the full value of the new medical
technologies that we create. Maximizing our public health gains and our
economic gains from new medical technology also requires that we
encourage high value innovations and also realize more value from the
products that we use. This is important for the future, because while
the cost of new medical technologies may continue to rise, the
potential benefits of new treatments could grow even more dramatically.
We must find better ways to increase value, to keep modern care
affordable, while still encouraging medical innovation. With these
unprecedented technological achievements have also come unprecedented
concerns about the total spending on healthcare and, in particular,
about the rising spending on these new medical technologies. Many worry
that, even if these new technologies come along, they will not benefit
because they will not be able to afford the high cost. While we need to
take new steps to address the problem of health care affordability, we
need to do it carefully. We must address this issue in a way that will
not risk the tremendous potential for public health and economic
benefits from continuing medical innovation by putting significant new
limits on the payments or the intellectual property protections of
innovative treatments that have made it through an increasingly long
and costly development process.
In particular, there is concern about the threats to innovation
because the process of medical innovation--of turning sound ideas from
insights in the biomedical laboratory sciences into safe and effective
products for treatments--has steadily become more costly. Getting a
product into general use is an increasingly lengthy and costly business
and fraught with significant risk.\4\ Some estimates put the total cost
of developing a novel drug at more than $800 million, and by all
estimates it has increased substantially in the past decade.\5\ Too
often, the process is unpredictable, and may take years of hard work
with high costs for product testing and developing reliable production
lines.\6\
---------------------------------------------------------------------------
\4\ DiMasi, J.A. ``Uncertainty in drug development: approval
success rates for new drugs'' Chapter 20 in Clinical Trials and
Tribulations, 2d Edition, revised and expanded. Cato AE Sutton L Cato A
III, editors. New York: Marcel Dekker, 2002:361-77.
\5\ DiMasi, J.A., Hansen, R.W., Grabowski, H.G. ``The Price of
Innovation: New Estimates of Drug Development Costs.'' Journal of
Health Economics, 2003 Mar; 22(2):151-85. Kaitin KI, Ed. ``Post-
approval R&D raises total drug development costs to $897 million.''
Tufts Center for the Study of Drug Development Impact Report 2003 May/
Jun;5(3).
\6\ Horrobin, David F. ``Modern Biomedical Research: An Internally
Self-Consistent Universe with Little Contact with Medical Reality?''
Nature Drug Discovery; February 2003:151-154.
---------------------------------------------------------------------------
Many people involved in the development of new medical technology
believe the slowdown in drug approvals is likely to be only temporary.
Currently, the National Institutes of Health (NIH) is completing a
five-year doubling of its budget, to more than $27 billion. Less well
known is that spending on research and development by pharmaceutical
companies worldwide has also doubled since 1995 and now is estimated to
be more than $54 billion. R&D spending by biotech and medical device
companies is also rising. The impact of these investments in research
is already becoming evident in the form of more investigational new
drugs (INDs) under development than ever.
But if the impact of information technology on the economy is any
guide, it may require a decade or more of increased investments in
order to have a real impact on productivity--on how much output we get
as a result of these inputs. And it could take much longer, because of
the unusual length and uncertainty of the product development process
in health care. At this point in genomics, for example, scientists are
still primarily gathering information, sorting out patterns, and only
starting to understand what the turning on or off of hundreds of genes
by a new drug means for whether it is safe and effective in patients.
The increase in the time and cost of product development has already
been associated with a decline in the number of truly new drugs and
biological treatments being approved by FDA. Last year, FDA approved 21
new molecular entities (the truly new drugs) down from 44 such entities
in 1996. And FDA approved 12 new biological license applications
(BLAB), down from 27 BLAB in 1998. The decline in products approved is
not the result of FDA rejecting more applications; it is directly
related to a decline in the number of new applications for drugs and
biologics coming in to the Agency, and it is a worldwide phenomenon.\7\
---------------------------------------------------------------------------
\7\ CDER 2002 Report to the Nation: ``Improving Public Health
through Human Drugs.'' May 13, 2003.
---------------------------------------------------------------------------
traditional approaches versus new technologies
While there are and no doubt will continue to be traditional
``blockbuster''-type drugs in development that may bring important
public health benefits to many millions of patients, breakthroughs in
genomics, proteomics, and other new fields of molecular biology also
hold great promise for truly individualized drug therapy in which
diagnostic tests and novel drug delivery mechanisms guide the use of
medications, turning heterogeneous diseases like cancer and heart
disease into distinct types of pathologies that appropriately require
distinct therapeutic approaches. Other new technologies are breaking
down the traditional barriers between drugs, tissues, and devices,
including products in development that are combinations.
Translating the new biomedical sciences into these new kinds of
treatments for patients requires major new investments, and it seems
plausible that such investments may take many years to reach fruition.
It should not be surprising that we haven't yet seen the huge increase
in biomedical investment of the past decade, and especially the last
few years, turn into more and more valuable medical products for
patients. But the fact remains that developers of biomedical products
are not producing drugs particularly faster than they were before all
these innovations came along. From a public health standpoint, with
millions of Americans suffering from diseases that may be curable or at
least manageable in the not too distant future, we cannot afford to
wait many more years for all these investments to become valuable
products.
On the research and development side, it's possible that the costs
and uncertainty of developing new treatments could keep rising. It's
easy to see how this could happen: there are not many more obvious drug
targets left to exploit, and developing genomics- and proteomics-based
therapies remains very costly. So far, genomics has mainly added steps
at the front end of the development process, through microarray testing
of gene responses, and has not reduced the costs of clinical research
significantly. On the policy side, there is intense pressure to make
health care more affordable, and so policymakers may focus only on
reducing medical costs in the short run--which, if not done properly,
could reduce the incentive to incur these high and rising development
costs. This combination of rising costs of product development and
pressures to control costs rather than increase value is not a good one
for keeping the United States at the forefront of biomedical
innovation, and more importantly it's not a good combination for
affordable and high-quality, innovative health care for our population.
Instead, a more effective approach would involve bringing costs down by
reducing the high cost and uncertainty of developing new medical
treatments, and taking more steps to help patients and doctors use them
effectively after they are approved by FDA.
potential policy solutions
We can take steps today to improve the development and use of
medical technologies, and find creative policy solutions that both
support innovation and make healthcare more affordable, particularly
for those with limited means and great needs. There are many ways to do
this, but above all, we need to increase value in the process of
developing and using new medical technologies. To these ends, a key
element of FDA's new strategic action plan is efficient risk
management. In all of FDA's major policies and regulations, the Agency
is seeking to use the best biomedical science, the best risk management
science, and the best economic science to achieve its health policy
goals as efficiently as possible.
The enormous growth in research investment has required the Agency
to deal with more complex and innovative products in development than
ever before. As discoveries made in the laboratory are flowing into the
medical products consumers are using, it means that the Agency is
challenged to upgrade its own science to keep pace with this new
innovation and the growing sophistication of manufactures. As part of a
new FDA initiative on improving medical innovation announced in January
2003, the FDA is taking specific steps to help foster more efficient
innovation, especially in emerging areas or those of great medical
need. The initiative has several elements that are described below.
Need for Performance Measures
One element of this plan is the development of ``quality systems''
for the Agency's review procedures. The idea is to build on FDA's
professional staff expertise to identify and apply best management
practices internally to the review processes. This includes using peer
review programs coupled with more empirical data for drug and device
reviewers to exchange ideas and use each other's experience to learn
about best practices. A key part of this effort is developing
performance measures that the Agency's experts believe are related to
the goal of approving safe and effective treatments as efficiently as
possible.
FDA is also working to develop new guidance documents that can
bring more predictability to regulatory process. These are in a
tradition of FDA documents that serve as roadmaps for drug and device
developers, offering guidance on how to structure studies to prove that
new treatments work. These new documents represent an enhanced effort
to combine internal expertise with input from outside experts to make
sure that are regulatory methods are up to date in important areas of
technology development. Some of the guidance will focus specifically on
diabetes, obesity, and cancer. Despite all the innovation that has
already occurred, these are therapeutic areas that remain underserved
by effective treatments and that have promising technologies under
development today.
Developing New Guidance for New Areas
The Agency is also developing guidance in new areas of technology
development, including pharmacogenomics, novel drug delivery systems,
and cell and gene therapy. In each of these cases, the Agency expects
to learn something from outside experts in the open process of
developing them. For example, FDA is setting up a ``research
exemption'' program for product developers as well as academic experts
to share data on pharmacogenomic results, such as microarray studies,
that may be useful for predicting clinical benefits and risks and thus
reducing the costs of demonstrating safety and effectiveness. This kind
of information can also be used to increase the value of a new medicine
by allowing doctors to target drugs to patients most likely to derive a
clinical benefit or least likely to suffer a rare side effect. The goal
in all of these endeavors is to use the new regulatory standards to
reduce the time and cost of product development and to ensure that the
Agency's regulatory procedures are current at the same time. We hope
this will lead to earlier and broader access to new treatments.
Rapid Access to Generic Drugs
Supporting the development of safe and effective new treatments is
one of the most important ways that FDA can promote the public health.
But when appropriate patents have expired, we need to facilitate
broader access through lower-cost generic drug alternatives. Generic
drug manufacturers produce medications that are just as safe and
effective as their brand counterparts. Yet the prices of generics are
generally much lower. A generic version of a $72 average brand-name
prescription costs about $17. With more brand-name medications coming
off patent--more than 200 of them in the next few years--and with ever-
improving scientific knowledge and public awareness about the benefits
of generic drugs, the health and economic benefits of using generic
drugs are constantly growing.
Encouraging rapid and fair access to more affordable generic
medications is one of FDA's major priorities. FDA is proposing new
resources to enable us to implement major reforms in its generic drug
programs to reduce the time it takes to get a generic drug approved.
Right now, it takes well over a year and a half on average to approve a
new generic medication and we think we can significantly improve. In
addition, the Agency recently finalized a generic drug final rule that
would expedite and increase access to more affordable generic drugs by
limiting the ability of innovator drug companies to receive multiple
extensions that delay entry of generic competition. This final rule is
projected to save American consumers $35 billion dollars over the next
10 years. Furthermore, this rule makes changes to the patent listing
process that are also designed to improve generic competition.
Revised Good Manufacturing Practices
Another application of the principle of efficient risk management
to reduce medical costs and improved outcomes is in improving the way
that medical products are manufactured. These guidelines are referred
to as good manufacturing practices (GMPs), and these GMP regulations
for drugs have not been updated in 25 years. Meanwhile, best practices
in manufacturing technologies and methods have undergone significant
progress over that time, particularly in other high-tech industries.
For example, the semiconductor industry also has a very low tolerance
for impurities and inaccuracies in production. When its production
processes were lagging because of high costs and too many errors that
industry helped invent the ``six sigma'' production methods. Through
continuous quality improvement, those methods achieved enormous
improvements in production cost and quality, and they have since been
widely adopted in manufacturing industries.
But continuous quality improvement in manufacturing hasn't been the
subject of as much attention in the pharmaceutical industry, even
though many experts on manufacturing processes believe that large
savings in production costs could be realized while maintaining very
high standards for purity and accuracy. FDA wants to make sure that
regulations are encouraging such progress, not standing in the way. The
Agency is working on a program for developing new GMPs based on the
latest science of risk management and quality assurance. The new
standards would be designed to encourage cost-reducing and precision-
enhancing innovation in manufacturing and technology, and to ensure
that all three FDA medical centers use consistent and up-to-date
methods, including inspectors specializing in particular types of
production methods.
In addition to substantial savings in the development and
manufacturing of safe and effective medical products, there are many
more opportunities to increase the value of the medical products FDA
regulates after they are approved and maximize their public health
benefits. By making better information available to patients and
doctors about the benefits and side effects of new medical
technologies, and by taking other steps to help doctors and patients
avoid errors and adverse events, people can realize more value from
these products by making better decisions about when to utilize them
for maximum advantage.
Prevention of Medical Errors
Approved medical products, while safe and effective when used as
intended, can be involved in costly and potentially preventable adverse
events, including medical errors. A November 1999 report of the
Institute of Medicine (I0M), entitled ``To Err Is Human: Building a
Safer Health System,'' focused a great deal of attention on the issue
of medical errors and patient safety. The report indicated that as many
as 44,000 to 98,000 people die in hospitals each year as the result of
medical errors. About 7,000 people per year are estimated to die from
medication errors alone--about 16 percent more deaths than the number
attributable to work-related injuries.\8\ Preventable errors and
complications involving prescription drugs alone are also responsible
for billions of dollars in additional health care costs each year, in
addition to all of the unnecessary suffering. The IOM report estimates
that medical errors cost the Nation about $37.6 billion each year;
about $17 billion of those costs are associated with preventable
errors. About half of the expenditures for preventable medical errors
are for direct health care costs. That's too much money that would be
better spent on proper care.
---------------------------------------------------------------------------
\8\ To Err Is Human: Building a Safer Health System. Institute of
Medicine Press, 2000.
---------------------------------------------------------------------------
FDA has a role in helping to avoid these costly errors by
supporting the development and use of safer health care systems;
systems that help health professionals avoid errors and deliver higher
quality care. The majority of medical errors do not result from
individual recklessness, the report says, but from basic flaws in the
way the health system is organized. Stocking patient-care units in
hospitals, for example, with certain full-strength drugs (even though
they are toxic unless diluted) has resulted in deadly mistakes. And
illegible writing in medical records has resulted in administration of
a drug for which the patient has a known allergy.
To help mitigate these risks, earlier this year FDA proposed a
universal bar coding system for prescription medications and blood
products. Coupled with barcode readers and electronic medical records,
bar codes on drugs are expected to reduce the rate of medication errors
that occur at the stage of dispensing and administering medications by
half or more. Bar codes can help make sure that the right patient gets
the right medication in the right dose at the right time, and soon a
standardized system of codes will be built in to all drug packaging.
Based on the published relationships between hospital admissions and
adverse drug events, FDA has estimated that of 372,000 preventable
adverse drug events per year in hospitals, bar code identifiers on drug
products could be expected to avoid about 22 percent of these events.
Over 20 years, FDA expects more than 413,000 fewer adverse drug events
because of bar coded products. The average annual benefit of avoiding
these events is $3.9 billion dollars in patient pain and suffering and
direct treatment costs.\9\ FDA's work on standards has another benefit.
According to the hospital industry and many health care purchasers,
standard bar codes will speed the adoption of electronic health
information systems by hospitals and other healthcare organizations,
because the standardized codes increase the payoff from having
electronic systems.
---------------------------------------------------------------------------
\9\ Bates, D.W., D.J. Cullen, N Laird, L.A. Peterson, S.D. Small, D
Servi, et al. ``Incidence of Adverse Drug Events and Potential Adverse
Drug Events: Implications for Prevention.'' The Journal of the American
Medical Association. July 5, 1995. Classen, D.C., S.L. Pestonik, R.S.
Evans, J.F. Floyd, and J.P. Burke. 1997. ``Adverse Drug Events in
Hospitalized Patients: Excess Length of Stay, Extra Costs, and
Attributable Mortality.'' The Journal of the American Medical
Association. January 22/29. Jha, A.K., G.J. Kuperman, J.M. Teich, L
Leape, B Shea, E Rittenberg, et al. 1998. ``Identifying Adverse Drug
Events.'' The Journal of the American Medical Association. May.
---------------------------------------------------------------------------
Even with the best available data, drugs are sometimes found to
have adverse effects that could not have been predicted or uncovered in
any feasible clinical trial. Most of these subtle or rare problems,
such as liver toxicities, that occur in a small number of people and
most become apparent only after drugs have been used in real-world
patient populations for some period of time. The Agency must have
effective systems in place to detect such problems, so that preventable
adverse events are identified, and better ways can be found to prevent
these events.
As part of this effort, the Agency is working on developing
information technology tools that will allow it to link into the
electronic; medical records of large healthcare institutions and
organizations, and automatically scan medical records for combinations
of new drugs and clinical endpoints such as blood test results that
might contain harbingers of trouble. The idea is to use modern
information technology to acquire information on associations between
adverse events and use of a medical product that might warrant focused
further investigation. FDA wants to have systems in place that allow us
to be proactive in collecting this clinical information, rather than
continuing to rely primarily on vigilant doctors and FDA's voluntary
adverse event reporting systems.
Safety and Efficacy Studies for Approved Medical Products
More studies of the safety and effectiveness of medical products
after they are approved can be very helpful for learning more about the
risks and benefits of medications in special populations and can help
guide more informed medical decisions. For example for a new cancer
drug that recently gained accelerated approval, the National Cancer
Institute is funding so-called ``Phase 4'' studies to confirm clinical
benefits and help assess longer-term risks. These efforts to use modern
information systems and post-approval studies can add substantially to
the body of knowledge about which patients are most and least likely to
benefit from an approved treatment, in turn leading to higher-value
treatment decisions.
Better Informed Consumers
FDA is also working to encourage more effective, high-value use of
medical treatments by helping patients and health professionals get
access to the latest and best information on risks and benefits. For
all that improving medical technology can do, it is much less than
people can do through their own choices to improve their health. From
encouraging better guidance to patients in pharmacy labels, to clearer
guidance on communicating risk and benefit information in direct to
consumer advertising, to new enforcement initiatives against dietary
supplement manufacturers who make health claims without scientific
foundation, to food labeling that better discloses diet-disease
information, FDA is undertaking new efforts to help consumers make
better-informed decisions about how to use their health care dollars.
In one recent example, FDA is working on a DailyMed program for
physicians, so that a redesigned electronic product label that can be
updated daily to include the most current information about a drug
after they are already on the market. Only by facilitating access to
complete, timely, and easily used information available to consumers
and health professionals can FDA help to make sure that people are
making the best decisions about their health based on the best
available information.
conclusion
Medical innovation is a difficult and complex process, but one that
can bring great value to patients. This long and difficult process is
also a delicate one that requires the right mix of incentives,
safeguards, and effective regulation to make sure people can derive the
maximum benefit from safe and effective new medical technologies. Only
by adopting policies that protect the incentives to develop new drugs
and medical devices, and reward cost-effective medical practice and the
most high value use of new technology, will we continue to realize the
full benefits of these innovations. As described in this testimony, at
FDA, as Commissioner of Food and Drugs, I am working to implement
numerous policies, initiatives, and regulatory improvements that
reflect these critical needs in order to promote increased access to
high quality, high value, safe and effective medical products,
including drugs, biologics, devices and combinations of all three.
I appreciate the opportunity to provide this testimony and I would
be pleased to respond to any questions.
__________
Prepared Statement of Carolyn Clancy, M.D., Director,
Agency for Healthcare Research and Quality
Good morning, Mr. Chairman and Members of the Committee. I am very
pleased to be here today to discuss the important issues of how we can
facilitate, sustain, and promote health care innovation while we ensure
that we have a health care system that is affordable. As my testimony
will indicate, I believe that the work of the Agency for Healthcare
Research and Quality (AHRQ) is critical to achieving these goals and
complements the important work of the National Institutes of Health
(NIH) and the Food and Drug Administration (FDA) and supports decision-
making by the Centers for Medicare and Medicaid Services (CMS).
ahrq's role
Let me begin with a few words about where AHRQ fits within the
Department of Health and Human Services. The basic and biomedical
research supported by the NIH serves as the foundation for many of the
advances in the prevention, diagnosis, and management of disease and
impairment. Its work greatly expands the realm of possible public
health and clinical interventions. While the Centers for Disease
Control and Prevention (CDC) takes the lead on public health,
community-based interventions often led by state and local health
departments or public service media campaigns to improve health, AHRQ
focuses on the role of clinical care and the health care delivery
system.
AHRQ's mission is to improve the effectiveness, quality, safety,
and efficiency of healthcare services that patients receive. What is
unique about our mission is that it encompasses both the evaluation of
the effectiveness and quality of clinical services and the most
effective and efficient ways to organize, manage, and safely deliver
those services. As the Institute of Medicine report To Err Is Human
made clear, this dual focus--on services and systems--is critical to
improving health care.
AHRQ contributes to efforts to speed the diffusion of effective
medical breakthroughs. Our research can extend the findings of
biomedical research to populations not included in clinical trials,
evaluate the effectiveness and cost-effectiveness of interventions to
determine which populations benefit most, and develop effective
strategies to facilitate their rapid adoption. We also facilitate
adoption of new knowledge by putting into perspective the available
scientific evidence so that clinicians can better assess the importance
of recent breakthroughs.
In the area of drugs and devices that have received FDA approval,
AHRQ focuses on their effectiveness (especially in comparison to
existing options) and cost-effectiveness. We complement FDA's focus on
the safety of drugs, biologics, and devices, with our focus on their
safe use in daily practice. In the context of this hearing, this role
is especially important. The harm that can result from inappropriate
use of otherwise safe drugs, biologics, and devices is not only a
tragedy for the patients involved but adds to health care inflation
through the costs involved in attempting to repair the damage and
related increases in medical liability expenditures. As a result, I am
delighted to report that Dr. McClellan and I are developing an
increasingly strong partnership between FDA and AHRQ in these areas.
However, innovations in health care are not limited to drugs and
devices but may also include new surgical procedures, new applications
of existing technology, information technology or communications
advances. Moreover, while some of these innovations offer unprecedented
breakthroughs for some patients they may also result in unintended harm
if not used appropriately. AHRQ's role, then, is to provide the best
evidence regarding how to match specific services to patients' needs
and preferences to promote the best possible outcomes.
Finally, we serve as a science partner for efforts by the Centers
for Medicare and Medicaid Services to improve the effectiveness,
quality, and safety of services they support and improve the ability of
beneficiaries to make more informed health care choices. Prior to our
1999 reauthorization, we were required by law to make recommendations
to CMS on coverage decisions. Today, upon request, we undertake
technology assessments and other research activities to objectively
synthesize all existing evidence on the effectiveness of medical
interventions under consideration for coverage by CMS. We do not make
recommendations.
health care innovation and health care costs
Mr. Chairman, America has a track record for health care innovation
that is the envy of the world. The Administration and Congress in
partnership have done much to accelerate and sustain that record
through their commitment to biomedical and health care research. As a
result, the pace of innovation has accelerated, the number of
scientific journals and published research studies is exploding, and
reports of scientific breakthroughs appear almost daily.
Many of these developments offer the potential for greatly
improving the quality of life for patients; in other cases the
improvements are marginal at best. In some cases, innovation leads to
the same or even higher quality of care at significantly lower costs
while other innovation is cost increasing. The underlying challenge,
therefore, is to effectively sort through the increasing array of
clinical care options to develop objective scientific information so
that those who make decisions--policymakers, systems managers,
insurers, purchasers, clinicians, or patients--can make informed
choices. The ultimate goal is to ensure that they can get real value
for their health care dollar. Each of us may make different decisions
as we weigh the evidence. My Agency's role is not to make those
judgments. It is to develop and synthesize the evidence regarding
health care interventions so that, whether you favor the current
insurance-based system or favor a more consumer-driven model of health
care decisionmaking, objective credible scientific information--on
effectiveness, cost-effectiveness, and benefits (including downstream
cost savings)--is available to inform those decisions.
The need for such information has never been more compelling.
Moreover, the resurgence of health care cost inflation at a time of
increasingly constrained resources, both in the public and private
sector, will only accelerate the demand for proof that we are getting
real value for the health care dollars that we spend. Because our
research focuses on both the effectiveness and cost-effectiveness of
health care services as well as ways to improve the effectiveness,
efficiency, and safety of the ways we deliver and use health care
services, AHRQ is uniquely positioned to develop this type of
scientific evidence.
how ahrq can help
Let me suggest five broad areas in which AHRQ can assist in sorting
through the array of new health care innovation and help to speed the
adoption of effective interventions.
First, AHRQ research identifies what is effective and cost-
effective in daily practice. Experience suggests that new drugs,
technologies, and medical or surgical interventions are seldom equally
effective for all types of patients. Will a breakthrough for the
treatment of arthritis, tested in clinical trials with patients who
only have that affliction, work as well in patients who not only have
arthritis but are also taking medications for diabetes, congestive
heart failure, and hypertension? Or how well does it work in patients
whose racial, ethnic, and demographic characteristics differ from those
in the clinical trial? Consider two examples from our research, one
demonstrating the value of using the low-cost option; the other
demonstrating the value of investing in much more expensive
pharmaceuticals.
The first example, treatment of otitis media (middle ear
infection), is the most frequent reason for administering antibiotics
to children. Over-prescribing increases the chance for adverse
reactions, leads to the development of bacterial resistance, and
increases expenditures. AHRQ supported researchers found that the use
of the less expensive generic antibiotics resulted in the same or lower
failure rates. They concluded conservatively that substituting low cost
antibiotics for only half of the expensive antibiotic prescriptions
would have saved Medicaid nearly $400,000. This research has led to the
development of guidelines by the American Academy of Pediatrics
recommending less-expensive antibiotics and to a metric used to
accredit health plans.
By contrast, in some cases, costly new interventions can reduce the
long-term use of other health care resources. AHRQ research
demonstrated that new, more costly anti-retroviral therapy for treating
AIDS patients is both effective and cost-effective. The increased
expenditures for those drugs are much less than the savings in
inpatient, outpatient, and emergency room costs. Overall annual costs
per patient were reduced from $20,300 to $18,300. If extrapolated to
the approximately 335,000 adults receiving care for HIV infection in
1996, over $500 million will be saved in HIV related healthcare.
Second, AHRQ research identifies strategies for overcoming barriers
to the use of effective services. Great opportunities for improving
health, developed through biomedical research, are easily lost if
physicians and patients are unable to make the best use of the
knowledge in everyday care. These wasted opportunities are apparent
daily in the under use of effective interventions and continued
reliance upon outmoded approaches to patient care, which in turn
contributes to the ever-increasing cost of care and avoidable loss of
lives. By conducting and supporting research that focuses on their
effective use, and working with clinicians and health care
organizations to assure that this information is accessible when
decisions are made, AHRQ ensures that Americans reap the full rewards
of basic research and medical innovation.
For example, NIH-supported research identified the potential of
warfarin, a blood thinner, to reduce the risk of stroke in patients
with atrial fibrillation. But physicians seldom prescribed warfarin for
their patients. AHRQ-supported researchers concluded that warfarin was
effective in daily practice, identified the reasons that physicians
were reluctant to use warfarin, and developed a program of providing
warfarin that would have an expected annual net savings of $1.45
million per 100,000 people aged 65 years or older, of whom 6,000 would
be expected to have atrial fibrillation. Using this knowledge, Medicare
Peer Review Organizations implemented projects to increase
anticoagulation, and 28 projects in 20 states had a 58-71% increase,
with a projection of 1,285 strokes prevented. The findings of this AHRQ
funded study were influential in the development of guidelines by the
American College of Physicians, American Heart Association, American
College of Chest Physicians, and the Joint Council of Vascular
Surgeons. Based on this work, United HealthCare has included use of
anticoagulation therapy for patients with atrial fibrillation in the
profiling of its 262,000 physicians.
Third, AHRQ facilitates the use of Evidence Based Medicine. In
recent years AHRQ has focused increased attention on the development of
technology and tools to facilitate the use of evidence-based medicine.
For example, each year tens of thousands of patients who go to an
emergency department worried that their chest pain is being caused by a
heart attack, are inappropriately sent home, inappropriately
hospitalized, or suffer because of delay in treatment due to an
inconclusive electrocardiogram (EKG). These delayed or missed diagnoses
have serious implications for patient survival or impairment rates,
hospital costs and subsequent malpractice lawsuits. An increasing
number of EKGs are now equipped with special software developed by AHRQ
research that improves diagnosis by predicting the likelihood of
whether chest pain is the result of a heart attack. The software could
prevent 200,000 unnecessary hospitalizations and more than 100,000
coronary care unit admissions a year and save roughly $728 million a
year in hospital costs if implemented in half of the hospitals
nationally. Soon-to-be-published research estimates that improved
accuracy of diagnosis that results from use of this predictive tool
could reduce malpractice costs nationally by $1.2 billion per year.
Approximately 600,000, or 15 percent, of the 4 million Americans
who develop pneumonia each year are hospitalized. Because of the lack
of evidence-based admission criteria and the tendency to overestimate
the risk of death, many low-risk patients who could be safely treated
outside the hospital are admitted for inpatient care. An easy-to-use
method developed by AHRQ-supported researchers accurately predicts
which pneumonia patients can be safely treated at home, which costs 10
to 15 times less than hospital care for pneumonia. The findings from
this study also suggest that hospitals could reduce pneumonia hospital
stays in many cases by 1 day without adversely affecting patient
health. Criteria were developed to assist physicians with determining
when patients could be discharged safely.
Fourth, AHRQ research assesses the effectiveness of cost
containment and management strategies. With Medicaid pharmaceutical
costs increasing 20% per year, States are considering and implementing
a variety of cost containment strategies. An example of how our past
research can be helpful to today's decisionmakers involves a study of
an initiative by a New England legislature to limit Medicaid
reimbursement to three prescriptions per month. AHRQ concluded that the
strategy back-fired. Increases in utilization costs were 17 times
greater than the savings in drug expenditures. The result was that the
state abolished the prescription cap, and another 9 states have also
changed their policies based on this research.
AHRQ research has also demonstrated that 85% of women with pelvic
inflammatory disease, the leading cause of infertility, can be safely
and effectively treated as outpatients, and developed an evidence-based
approach to identify which nursing home patients require
hospitalization for possible pneumonia and which can be treated at the
nursing home This approach not only saves the cost of a hospitalization
but also helps frail, elderly patients avoid the risks of experiencing
additional hospital complications.
Fifth, AHRQ's role in speeding the pace of evaluation of health
care innovation. AHRQ's 1999 reauthorization directed us to serve as a
science partner for public and private sector efforts to improve
quality and urged us to continue our efforts, begun in the mid-1990s,
to speed the pace of the evaluation of health care innovations.
One of the critical roadblocks to coverage of innovative
interventions is the lack of solid scientific evidence regarding their
effectiveness, especially in comparison with existing interventions.
While the FDA determines that a drug, biologic, or device is safe and
that it has an impact when compared to placebo, those making coverage
decisions, including clinicians and patients, still need more
information regarding its relative effectiveness and relative costs.
Similarly, promising biomedical research breakthroughs face a similar
test. This is often frustrating for those whose creativity leads to the
development of promising new technologies as they come to realize that
passing FDA scrutiny is only part of the journey toward seeing their
innovation in widespread use.
While these constraints are not of AHRQ's making--and are certainly
not unique to the public sector; the private sector takes technology
assessment seriously as well--we have begun, and will continue, our
efforts to facilitate the speed of this process. For example, when
Medicare asked us to evaluate the effectiveness of lung-volume
reduction surgery, we concluded that there was insufficient evidence to
reach a determination at that time. But we pointed out to Medicare the
potential for developing the evidence through an innovative process of
conditional coverage--in which Medicare would pay for the procedure in
selected institutions, provided the surgeons and patients agreed to the
collection of outcomes data. This resulted in a partnership between
Medicare, the National Heart Lung and Blood Institute, and AHRQ to
assess the procedure. As a result of this study, we now know which
patients are likely to benefit, and very importantly, a subgroup of
patients who experienced increased mortality as a result of the
procedure were identified so that avoidable and unintended deaths can
be reduced.
Similarly, AHRQ has revamped its ability to provide Medicare with
much more timely scientific advice, in as little as two weeks for brief
assessments of the volume of available evidence to full-scale
technology assessments that might take a year. These time frames
reflect as significant improvement in our ability to serve Medicare
more effectively.
There are at least two other ways in which we can serve as a
science partner for private sector innovation. First, most technology
assessments conclude that there is a lack of credible scientific
studies from which to judge whether a technology is effective or
ineffective. We are prepared to work with industry trade associations
to assist their members, who have products moving to the end of the FDA
review process, to better understand the types of studies that will be
needed to assess the effectiveness of their products. This simple step
would make a significant contribution to facilitating timely assessment
of health care innovation.
Second, in future years, as existing patient safety grants end, we
will want to expand our focus on human factors research. As one wag
commented, human factors research helps us to ``idiot proof our
technology.'' More accurately, this research helps us to develop
controls for our technologies so that they remain easy to program even
by a harried, stressed, distracted, sleep-deprived health care
professional. One example is the infusion pumps, used to administer
fluids to patients through their veins, that are often involved in
patient safety adverse events. Human factors research would help us to
understand approaches for reducing inadvertent errors in programming
these pumps. As we expand our support for human factors research within
our patient safety portfolio we will want to work with industry to
ensure that we are targeting the critical questions that will improve
the safety and quality of the products they design in the future. By
ensuring that this type of critical information in the public domain,
we can be a science partner for their efforts to develop even more
effective and safe health care technologies.
ahrq's new direction
Mr. Chairman, before concluding, I would like to say just a few
words about the future direction of AHRQ. As you know, I have been
serving as Acting Director since March, 2002 and Director now for five
months. During that time, our senior staff and I have undertaken a top
to bottom review of our procedures and processes to determine how we
can better fulfill the mandate of our 1999 reauthorization legislation
to serve as a science partner for public and private sector efforts to
improve quality.
We are determined to make AHRQ a ``problem solving'' agency. This
entails a greater focus on ``implementation research'' that is designed
to develop strategies for overcoming barriers to the adoption of
clinical interventions that are both effective and cost-effective. We
need to be more pro-active in closing the gap between what we know is,
effective and cost-effective in health care and what is done in daily
practice.
We have developed closer linkages, at every stage of the research
process, between the ultimate customers of our work and researchers, to
ensure that we are addressing their highest priority challenges. In the
public sector, we are beginning to work more closely than ever before
with Medicare, Medicaid, the Community Health Centers, the Federal
Employee Health Benefit Program, and the Departments of Defense and
Veterans Affairs.
We also will be giving greater priority to identifying strategies
for eliminating waste, assuring that evidence-based information is
current, bringing our health care infrastructure, especially
information technology, into the 21st century, redesigning workflow so
that health care professionals can work more efficiently and
effectively, and evaluating our financial and other incentives to
ensure that we encourage safe, high quality care.
conclusion
Mr. Chairman, in conclusion, let me note that one study
demonstrated that the time frame from the approval of a research grant
that ultimately yields useful findings to the widespread diffusion and
adoption of those results was at least 17 years. That time frame is
unacceptable. AHRQ is committed to playing its role in developing the
scientific evidence for identifying effective interventions sooner and
increasing the pace of their diffusion.
This concludes my formal testimony. I will be happy to respond to
any questions.
______
estimates of the impact of selected health information technologies on
quality and costs in inpatient and outpatient settings
executive summary:
Healthcare Information Technology has the potential to improve the
quality, safety and efficiency of healthcare by helping health care
professionals make the best decisions and by assuring that those
decisions are implemented as intended. This potential value will be
realized in better adherence to clinical protocols, utilization of
clinical decision support, reduction in medical errors, cost reductions
and improved access to healthcare information.
In order to estimate and put in context the relative value of these
high-impact HIT functions AHRQ compiled the following analysis. The
relative impact on quality, cost and net savings for selected in/
outpatient HIT functions is outlined below:
------------------------------------------------------------------------
Impact on Cost/Net
HIT Function Impact on Quality Savings
------------------------------------------------------------------------
Computerized Physician Order Decrease rate of Total annual
Entry (inpatient). serious med error savings range
by 55%; decrease from $7 to $14
rate of potential billion
adverse drug (nationally)
events by 84%.
Clinical Decision Support Decrease ordering Decrease
Technologies. of drugs that pt. antibiotic cost
is allergic to; by -$200 per
decrease in hospitalization;
orders for wrong lower cost of
(ineffective) hospital care
meds. ($26,315 v
$35,283) and
shorter hospital
stays (10 v 12.9
days)
Automated Medication Dispensing Significantly One hospital
Systems (inpatient). fewer missed realized savings
doses of drugs (- of $1.28 million
16.9%). over 5 yrs.
Bar Coding Technologies......... 75% decease in Annual national
errors caused by savings of $15.3
administration of billion.
wrong meds; 93%
reduction in
errors from wrong
med to wrong pt..
E-Prescribing in Physician Decreased One study
Practices. medication demonstrated
errors; Improved reduced pharmacy
physician costs of $1.15
efficiency. PMPM; 30%
decrease in
physician to
pharmacy phone
calls.
Computerized Physician Order Eliminate 2 $27 billion
Entry (outpatient). million adverse savings in
drug events; medication
Avoid 1.3 million expenses
office visits and (nationally)
190,000
hospitalizations.
Electronic Medical Records 34% reduction in Reduced Spending
(Primary Care Settings). adverse drug by $44 billion
events; 15% per year: Savings
decrease in drug of $86,400 per
utilization; 9% provider over a
decrease in five yr. period.
unnecessary lab
utilization.
------------------------------------------------------------------------
estimated net savings:
Our analysis demonstrates potential savings ranging in the tens to
hundreds of billions for these few high value functions.
computerized physician order entry in inpatient settings
Description. Computerized physician order entry (CPOE)
systems allow physicians to submit orders for medications and
laboratory tests using an online system. The technology includes
algorithms that prompt physicians about possible drug-drug
interactions, drug allergies, and the need to order certain laboratory
tests to measure whether a medication is effective. A study by Bates et
al., conducted at Massachusetts General Hospital and Brigham and
Women's Hospital in Boston, Massachusetts, compared the rates of
adverse drug events before and after implementation of CPOE.
Impact on Quality. The study showed that use of CPOE in
those hospitals reduced the rate of serious medication errors resulting
in patient injuries by 55 percent, from 10.7 events per 1000 patient-
days to 4.86 events. The rate of potential adverse drug events--that
is, errors that did not result in an injury--decreased by 84 percent.
The study found that the rate of errors in ordering of medications fell
by 19 percent, the rate of errors in transcription of orders fell by 84
percent, the rate of errors in dispensing of medications fell by 68
percent, and the rate of errors in administration of medications fell
by 59 percent.
Another study by Teich et al. (2000) examined the CPOE
system at Brigham and Women's Hospital and Massachusetts General
Hospital as well. The study compared physician prescribing practices
before and after implementation of the CPOE system. It found that use
of CPOE occurred contemporaneously with an increase in adherence with
certain clinical guidelines that were programmed into the CPOE system.
Use of computerized guidelines was associated with an increase in the
use of a recommended drug, while use of a dose selection menu was
associated with a decrease in variation in drug dosages among similar
patients. The proportion of doses that exceeded the recommended maximum
dosage decreased from 2.1 percent to 0.6 percent, while the display of
a guideline for administration of a particular drug increased the
proportion of orders that complied with the guideline. Each of those
results was statistically significant.
Impact on Cost. An earlier study by Bates and colleagues
found that the annual cost of preventable adverse drug events at
Brigham and Women's Hospital was $2.8 million. A 17 percent reduction
in preventable adverse drug events was observed in this study, which
would equate to annual savings of $480,000 for that hospital.
Estimated Net Savings. Implementing computerized patient order
entry in all hospitals in the U.S. could reduce the rate of preventable
adverse drug events by 17 percent, avoiding 656,800 preventable adverse
drug events per year.
The additional cost of treating a preventable adverse drug event
has been estimated at $5,857. Thus, the savings from averted
preventable adverse drug events could total $654 million per year.
A study of the implementation of CPOE in a hospital with 726 beds
found that annual savings for that hospital were between $5 and $10
million. If this savings can be extrapolated to the over 1 million
hospital beds in the U.S., total annual savings would range from $7 to
$14 billion.
The first year cost of implementing a CPOE system in an individual
hospital ranges from $2,480 to $15,000 per bed, while the ongoing cost
of maintaining the system ranges from $870 to $1500 per bed. Amortizing
the initial costs over 20 years at 7 percent interest, the national
costs of implementing CPOE systems in each of the 1 million hospitals
in the U.S. could range from $1.1 to $2.9 billion. (Amortizing those
costs over 5 years would equate to annual costs of $1.5 to $5.1
billion, while amortizing those costs over 10 years would equate to
annual costs of $1.2 to $3.6 billion.)
clinical decision support technologies
Description. LDS Hospital in Salt Lake City, Utah,
implemented a clinical decision support system that assists clinicians
in choosing a course of antibiotic and anti-infective therapy for
patients in the intensive care unit. The decision support system uses
information about the patient's diagnosis, white-cell count, body
temperature, and information from pathology and microbiology reports to
recommend a course of anti-infective therapy for identified and
potential pathogens. The system also considers information about drug
allergies, drug-drug interactions, and costs in choosing a recommended
course of therapy. A study of the system was reported by Evans et al.,
1998.
Impact on Quality. The study found that the system was
associated with a significant reduction in orders for drugs to which
patients had reported allergies (from 146 to 35 during the previous
two-year period), reduced excess drug dosages (from 405 to 87), and
reduced antibiotic-susceptibility mismatches (from 206 to 12). The
average number of days of excessive drug dosage was significantly
reduced (from 5.9 to 2.7), as was the number of adverse events caused
by anti-infective agents (from 28 to 4.) Each of those results was
statistically significant.
Impact on Costs. The study found that patients who
received the recommended anti-infectives had lower costs of anti-
infective agents ($102 vs. $340 for those in the preintervention
period), lower costs of hospital care ($26,315 vs. $35,283 for those in
the pre-intervention period), and shorter hospital stays (10 days vs.
12.9 days for those in the pre-intervention period.)
automated medication dispensing systems in the inpatient setting
Description. Automated medication dispensing systems
replace the existing manual systems used in many hospitals to dispense
a 24-hour supply of each patient's drugs to nurses on the floor. The
automated system is connected to the pharmacy computer system, so that
orders for new prescriptions are transferred electronically to the
automated dispenser. The automated system stores and dispenses most of
the medications that nurses administer to patients, while automatically
billing for the drugs used. A study of the use of an automated
dispensing system at the University of California, San Francisco
Hospital was reported by Schwarz et al., 1995.
Another study of an automated dispensing technology in a 600-bed
teaching hospital in Dallas, Texas, was conducted by Borel and Rascati
(1995).
Impact on Quality. The University of California study
found that after implementation of the automated dispensing system,
there were significantly fewer missed doses of drugs. The number of
reported medication errors decreased for the surgical unit, but
increased for the coronary intensive care unit.
The Texas study found that before implementation of the automated
dispensing system, the medication error rate was 16.9 percent, while
after implementation of the system, the error rate dropped to 10.4
percent. (Most errors consisted of administering a drug at the wrong
time.)
Impact on Costs. The authors estimate that the automated
dispensing system could save the hospital $1 million over five years if
all personnel time savings could be translated into reductions in
staffing. The cost of the automated dispensing system for 330 acute
care beds and 48 critical care beds was $1.28 million over five years.
The savings of $2.08 million over 5 years was attributable to decreased
labor costs for pharmacists, pharmacy technicians, pharmacy billers,
and nurses.
bar coding technologies
Description. Bar code technologies replace traditional
data entry. Bar codes similar to those utilized in many other
industrial sectors allow the quick accurate linkages between component
parts of a complex process. For example, a patient's ID bracelet with a
bar code is scanned and compared against a similar code in a medication
dispensing unit prior to medication delivery. Another example is the
usage of bar codes to conduct inventory in a hospital pharmacy. Both
these examples allow for faster entry of information with fewer errors.
Impact on Quality. A review of the use of bar code
technologies was conducted by Bridge Medical, Inc. The Colmery-O'Neil
Veterans Affairs Medical Center, a division of the Eastern Kansas
Health Care System, developed proprietary Bar Code Medication
Administration software. In 2001, the health system reported a
medication error rate of 3.0 incidents per 100,000 units dispensed,
compared with 21.7 incidents per 100,000 units in 1993, the last year
in which a manual medication system was used. The health system
experienced a 75 percent decrease in errors caused by administration of
the wrong medication; a 62 percent decrease in errors caused by
incorrect dosing, a 193 percent improvement in errors related to giving
drugs to the wrong patient, and an 87 percent decrease in errors
related to administering drugs to patients at the wrong time.
Impact on Cost. FDA expects their proposed bar coding
rule, once fully implemented, to lead to 12.8 fewer adverse drug events
per hospital, a national reduction of 84,200 (23% less).
Full implementation of this rule would lead to annual net savings
of about $190 million in hospital treatment costs, roughly $29,000 per
hospital. This considers an average additional treatment cost of $2,257
per adverse drug event associated with errors occurring at bedside. The
average start up costs for a hospital is $369,000, and, after
installation, the average annual operating costs are predicted to be
$312,000 per hospital.
The annual societal benefit from avoiding medication errors is
about $2.3 million per hospital, an estimated benefit of $15.3 billion
nationally. Approximately 2,400 mortalities and 1,600 permanent
disabilities would be avoided each year.
When both treatment and societal savings are combined, annual
reductions per hospital would be $2,329,000. Considering start up costs
(amortized over 20 years at 7%) and annual operating costs, the net
annual benefit is likely to be $1,983,000 per hospital.
e-prescribing in physician practices
Description. E-prescribing technologies allow physicians
to submit prescriptions to pharmacies electronically. The technologies
eliminate problems associated with hand-written prescriptions and
incomplete orders, and also allow physicians to check potential drug
interactions at the time the prescription is ordered. Advocates of e-
prescribing believe it is capable of improving patient safety,
improving adherence to formularies, and increasing online access to
patient information and decision support resources. Quantum, Inc., a
physician practice management company in San Antonio, Texas,
implemented an e-prescribing system sold by Allscripts, Inc., in 1998.
Another example includes the Tufts Health Plan and AdvancePCS
implementation of an e-prescribing technology called PocketScript. The
technology which can be used remotely on Personal Digital Assistants or
even Blackberries was introduced to 100 physicians' offices in
Massachusetts. Finally, in another study Gandhi and colleagues (2002)
compared rates of medication errors and adverse drug events in two
physician practices that used electronic prescribing technologies with
two practices that used traditional hand-written prescribing over a
six-week period.
Impact on Quality. The Cap Gemini Ernst and Young studied
the Quantum/Allscripts implementation and found the system improved the
practices' efficiency and increased use of generic drugs by about 4
percent. In survey conducted following the Pocketscript implementation,
35 percent of physicians reported patient care benefits due to the
ability to check drug interactions and prescription accuracy. The
Gandhi study found that the practices that used electronic prescribing
had fewer violations of prescribing rules and fewer medication errors,
but the rates of preventable and non-preventable adverse drug events
were not significantly different. The main types of errors were related
to identifying medication-related symptoms and inappropriate drug
choice. Computerized ordering checks would have prevented only one-
third of the preventable adverse drug events that occurred.
Impact on Cost. One of the Quantum physician practices in
which the technology was used experienced savings of $1.15 per member
per month in pharmacy costs, for a total of $69,000. Increased
operational efficiency contributed to an additional $12,000 in savings
for that practice. Pocketscript technology improved operational
efficiency for the practices. It reduced phone calls between physician
practices and pharmacists by 30 percent, and saved nearly one hour per
pharmacist in a typical day.
ambulatory computerized physician order entry
Description. Computerized Physician Order Entry (CPOE)
systems in the ambulatory (or outpatient) setting allow physicians to
submit orders for medications, immunizations, lab tests, radiology
studies, nursing interventions, and referrals. A key component of CPOE
in the ambulatory setting is clinical decision support, which gives
physicians tools for diagnosing and treating patients while avoiding
medical errors. Clinical decision support, one of the most important
attributes of CPOE, essentially gives the physician access to a bank of
medical knowledge at the point and time of care. A review of CPOE in
ambulatory settings was conducted by the Center for Information
Technology Leadership. It included a literature review, interviews of
vendors, and an expert panel meeting.
Impact on Quality. The review found that nationwide
adoption of advanced CPOE systems in the ambulatory setting would
eliminate more than 2 million adverse drug events, and over 130,000
life-threatening adverse drug events. In addition, nationwide use of
CPOE would avoid nearly 1.3 million physician office visits per year,
and more than 190,000 hospitalizations per year.
Impact on Cost: The study estimates that nationwide use of
CPOE in the ambulatory setting could save nearly $27 billion in
medication expenses each year. Those savings include switches from
brand to generic drugs, switches from more expensive to less expensive
drugs within the same therapeutic class and more appropriate drug
utilization. Of that total, savings of more than $2 billion would be
achieved through averted hospitalizations from prevented adverse drug
events, while $10 billion of savings would come from reduced radiology
costs and nearly $5 billion in reduced laboratory costs.
Estimated Net Savings: The Center for Information
Technology Leadership estimates that implementing advanced CPOE systems
in the outpatient setting would eliminate over 2 million adverse drug
events per year, and would avoid nearly 1.3 million physician visits,
190,000 hospital admissions, and over 130,000 lifethreatening adverse
drug events per year.
Nationwide adoption of advanced CPOE systems in the outpatient
setting would avoid about $44 billion per year in health care spending.
That savings would consist of savings on medications (60%), radiology
services (24%), laboratory services (11%) and avoided adverse drug
events (5%).
The cost of adopting advanced CPOE systems that include ambulatory
electronic medical record systems is over $29,000 per provider in the
first year, and about $4000 per provider in subsequent years. If those
costs were applied to each of the over 473,000 office-based physicians
in the U.S. and amortized over 20 years at 7 percent interest, the
annual cost of implementing an advanced CPOE system across the U.S.
would be $2.2 billion. (Amortizing those costs over 5 years would
equate to annual costs of $2.7 billion, while amortizing those costs
over 10 years would equate to annual costs of $2.4 billion.)
electronic medical records in primary care settings
Description. Partners Healthcare System in Boston,
Massachusetts, internally developed an electronic medical record that
replaces paper medical charts. The system aggregates a patient's
complete medical record--including physician notes, lab test, radiology
results, immunization records and a host of other data elements--into
an electronic version. The record, up to date and secure, is then
available to providers either at the patient's primary point of care
(physician office) or via secure linkage, at other sites of care (ER,
specialist, etc . . .). A cost-benefit analysis of the electronic
medical record was conducted by Wang et al., 2003.
Impact on Quality: The authors estimated that the
electronic medical record was associated with a 34 percent reduction in
adverse drug events, a 15 percent decrease in drug utilization, a 9
percent decrease in laboratory utilization, and a 14 percent reduction
in radiology utilization.
Impact on Cost. The study found that the electronic
medical record had net financial benefits of $86,400 per provider over
a five-year period. Savings in drug expenditures made up one-third of
that amount, with the remainder of savings attributable to decreased
radiology utilization, decreased billing errors, and improved charge
capture.
__________
Prepared Statement of Peter Neumann, Sc.D., Associate Professor of
Policy & Decision Sciences, Harvard School Of Public Health
Thank you very much Mr. Chairman for your invitation to speak
before this committee on the topic of technology, innovation, and their
effects on cost growth in health care.
I would like to speak today about how we can better understand the
value or cost-effectiveness of medical technology.
Broadly speaking, medical technology contributes to growth in
health care expenditures.
But this research says nothing by itself about the benefit side of
the equation. As we consider medical technology, it is important to
address not just how much medical technology contributes to health
costs, but whether the investments in medical technology are worth the
health benefits produced.
We all would like to get good value for our money when we pay for
new drugs, devices, and procedures. How do we get there? What tools do
we have to use, and what policy options are available? Formal economic
evaluation can help us answer these questions.
The field of economic evaluation of health and medical
interventions has been an active area of research in recent years. It
includes cost-effectiveness analysis, which shows the relationship
between the total resources used (costs) and the health benefits
achieved (effects) for an intervention compared to an alternative
strategy. Often a standard metric such as life-expectancy or quality
adjusted life-expectancy is used as the measure of health benefits.
In part with funding from the Agency for Health Care Research and
Quality, my colleagues and I have compiled a list of over 1500 cost-
effectiveness ratios, covering a wide variety of medical technologies
and public health strategies in many disease areas. More information is
available on our website www.lispb.harvard.edu/cearegistry.
These data underscore several important points about the cost-
effectiveness of medical technology. First, a great deal of information
on the topic has become available to policymakers in recent years.
Unlike many unsupported assertions made about the ``cost-
effectiveness'' of drugs and other medical technology, these studies
quantify costs and health effects using data and a standard, well-
accepted methodological technique.
Second, according to peer-reviewed articles, many technologies are
indeed cost-effective. Examples include warfarin therapy to prevent
stroke in those with atrial fibrillation, immunosuppressive drugs for
those with kidney transplants, and treatment with mood-altering drugs
for those suffering from depression. These interventions provide good
value in the sense that they produce health benefits for relatively
little cost, or may actually save money for the health care system.
Third, cost-effectiveness does not mean cost-savings. Over the
years, people have sometimes confused these terms. But restricting the
term cost-effective to cost-saving interventions (where equal or better
health outcomes is implied) would exclude many widely accepted
interventions, which do not save money but are ``cost-effective'' in
the sense that their additional benefit are worth their additional
cost.
A related point is that a critical aspect of any medical
technology's cost-effectiveness involves the manner in which the
question is framed. A technology is not intrinsically cost-effective or
cost-ineffective. It is only meaningful to say that a technology is
cost-effective compared to something else. A drug prescribed to lower
an individual's blood pressure may in fact be cost-effective compared
to the option of no treatment, but not necessarily when compared to an
alternative intervention, such as an intensive program of diet and
exercise, or another medication. Similarly, claims of cost-
effectiveness often depend on the population under investigation. For
example, statin drugs used to lower an individual's cholesterol have
been found to be relatively cost-effective as secondary prevention in
persons with existing heart disease, but considerably less cost-
effective as primary prevention.
Does anyone actually use CEA? Logically, cost effectiveness
analysis should be used by private insurers and state and federal
policy makers. However, many payers, including Medicare, have shied
away from using CEA in coverage and reimbursement decisions.
But why? Cost-effectiveness analysis promises to inform decisions
and enhance population health in an explicit, quantitative and
systematic manner. Medical journals, including the most prestigious
ones, routinely publish CEAs. Furthermore, many other countries have
incorporated CEA into their policy decisions.
How do we explain this paradox? Studies point to a couple of
explanations. Some of them fault the methodology itself. But in fact,
most experts agree on the basic tenets. Instead, the opposition more
likely relates to the hardened American distaste for explicit
rationing. This is understandable, perhaps. But still, how do we get
good value in face of this opposition?
I would offer five observations as we look ahead.
CEA should not be used rigidly. Leaders in the field have always
warned against using CEA mechanically, but experiences teaches that
rigid use of CEA will be resisted. Expectations for CEA should be
modest. CEA should inform decisions not dictate them.
CEA will not save money. CEA should not be conceptualized or
promoted as a cost containment tool, but rather as a technique for
obtaining better value. Paradoxically, using CEA may actually increase
health spending, because it often reveals under--than over treatment.
How you say it matters. Research shows that physicians understand
that resources are limited but they are not willing to admit to
rationing. Similarly, health plan managers deny that they ration care
but admit that their budgets are constrained. These responses are
instructive. It suggests that the term ``cost-effectiveness'' may be
part of the problem. We might instead use terms such as ``value
analysis'' and comparability, rather than cost-effectiveness analysis
and rationing.
Incentives first. Debates about the use of cost-effectiveness
cannot be separated from debates about the underlying health system and
the incentives they embody. The technique is sometimes opposed if used
centrally. But reconfiguring the incentives facing providers and
patients is challenging and critical.
Think broadly across sectors. A final message involves the
importance of thinking expansively about applications of CE
information. CEAs should not simply focus on medical interventions but
more broadly on interventions to improve health by reducing
environmental exposures, injuries at home and in the workplace, and
motor vehicle accidents.
In closing let me emphasize that whether medical technology offers
good value is a question that can only be informed by careful analysis.
I would encourage the judicious use of cost-effectiveness analysis in
the years ahead.
Thank you very much.
Table 1: Selected Cost-Effectiveness Ratios
------------------------------------------------------------------------
Cost per QALY
Interventions ratio (US $2002)
------------------------------------------------------------------------
Onetime colonoscopic screening for colorectal cancer Cost-saving.
at 60-64 yrs old vs. no screening in women over 40
years old...........................................
Cost-saving Chemoprevention with tamoxifen vs. $1,800
surveillance in 40-year-old women with high-risk
breast cancer 1/2 mutations.........................
Drug treatment vs. no treatment in stage I $4,800
hypertensive patients: men, age 80..................
High-dose palliative radiotherapy vs. best supportive $13,000
care in patients with advanced non-small-cell lung
cancer..............................................
Combined outreach for the pneumococcal and influenza $13,000
vaccines vs. no new outreach program in persons aged
65 years old and older never vaccinated with
pneumococcal vaccine and/or not vaccinated for
influenza in the last year..........................
Screening for diabetes mellitus vs. no systematic $22,000
diabetes mellitus screening in all ndividuals age 35-
44..................................................
Driver side air bag vs. no air bags in driving $30,000
population (and passengers).........................
Bypass surgery vs. medical management + aspirin over $35,000
5 years in ischaemic heart disease patients.........
Automated external defibrillators on large-capacity $36,000
aircraft, selective training vs. no automated
external defibrillators, attendants with basic life
support training in patients experiencing cardiac
arrest onboard US commercial aircraft during a 12
month period........................................
Coronary artery bypass graft surgery vs. percutaneous $99,000
transluminal coronary angioplasty (PTCA) in 55-yr.
old men with 3-vessel coronary artery disease and
type A lesions with severe angina and normal
ventricular function................................
Intensive school-based tobacco prevention program vs. $5,300-650,000
status quo (Current average national tobacco
educational practices) in every 7th and 8th grade in
the U.S.............................................
MRI + dynamic susceptibility contrast-enhanced (DSC) $530,000
magnetic resonance imaging (MRI) vs. head computed
tomography (CT) scan only in patients presenting for
the first time to an Alzheimer's Disease center/
clinic..............................................
Triple therapy with zidovudine, lamivudine, and $850,000
indinavir for all exposures vs. the current United
States Public Health Services (USPHS) post-exposure
prophylaxis guidelines in health care workers dosed
to known HIV+ blood.................................
Surgical strategy vs. Medical strategy in 45 year old $1,900,000
men with severe esophagitis.........................
------------------------------------------------------------------------
Source: Harvard School of Public Health Cost-Effectiveness Registry,
2003. www.hsph.harvard.edu.cearegistry.
Cost-Effectiveness of Underutilized Interventions in the Medicare
Population
------------------------------------------------------------------------
Percent
Cost-Effectiveness Implementation in
Health Intervention ($/QALY) Medicare
Population
------------------------------------------------------------------------
Influenza vaccine............... Cost saving....... 40-70
Pneumococcal vaccine............ Under $1OK/QALY... 55-60
Beta blocker treatment after Under $1OK/QALY... 85
myocardial infarction.
Mammogram....................... Under $20K/QALY... 75 (depending on
age)
Colon cancer screening.......... Under $20K/QALY... 20-40 (depending
on age)
Osteoporosis screening.......... Under $20K/QALY... 35
Antidepressant medication Under $25K/QALY... 40-55
management.
Hypertension control............ Under $50K/QALY... 35
------------------------------------------------------------------------
Source: Harvard School of Public Health, 2003.
QALY=quality-adjusted life year.
Note: The estimates in this table are intended to provide a rough guide
to cost-effectiveness and implementation. However, study methodology
for estimated cost-effectiveness often varies across analyses.
Moreover, cost-effectiveness may depend on factors such as the age and
gender of the population, and the particular screening and
technologies used.
__________
Prepared Statement of Neil R. Powe, M.D., MPH, MBA, Professor of
Medicine, Epidemiology and Health Policy & Management,
Johns Hopkins University
Good morning Mr. Chairman, Senators and Representatives. I am Neil
R. Powe, MD, MPH, MBA, Professor of Medicine, Epidemiology and Health
Policy & Management at Johns Hopkins University in Baltimore, Maryland.
I direct the Welch Center for Prevention, Epidemiology and Clinical
Research, an interdisciplinary research center of the Johns Hopkins
School of Medicine and Bloomberg School of Public Health. I am a
general internist, clinical epidemiologist and health services
researcher. My research has assessed the clinical and economic impacts
of biomedical innovation in medicine. It examines the impact of new and
established technologies on patients' longevity, functioning, quality
of life and costs. I have conducted cost-effectiveness studies of
technologies in several areas of medicine and have attempted to do this
with equipoise. Among the technologies I have studied are kidney
replacement therapies such as dialysis and transplantation,
biotechnology medications such as recombinant human erythropoietin,
cardiac revascularization procedures, imaging tests for lung and heart
disease, laboratory testing for periodic screening, laser therapies,
vascular procedures to prevent stroke and minimally invasive surgery. I
have also studied physician decision making and other determinants of
use of medical technology including payers' decisions about insurance
coverage for new medical technologies and the impact of financial
incentives on the use of technology.
New medical technologies include drugs, devices, procedures and the
systems of care in which we, as medical professionals, deliver them.
These include so called ``little ticket'' technologies which cost
relatively little individually, but when used at high frequency, can
become expensive. One such emerging ``little ticket'' technology is the
C-reactive protein (CRP) laboratory test for detecting inflammation now
being debated as a useful technology for detection of heart attack
risk. ``Big ticket'' technologies such as ``body scans'' and organ
transplantation have high individual price tags and can generate high
cost even when used relatively infrequently. In theory, a new medical
technology can increase costs, have similar costs or decrease costs
relative to the existing standard. Evidence to date suggests that much
of new biomedical innovation increases cost to the health system,
especially in the short run. ``Little ticket'' or ``big ticket'',
technology should not be judged based simply on costs. The more
important question that I would like to address is ``what is a
technology's value''.
Value is commonly seen as the benefit that is derived relative to
the cost. In theory, a technology can produce benefit relative to the
existing standard if patient outcomes (effectiveness and/or safety) are
better; on the other hand it can produce no benefit if outcomes are
similar, or even produce harm if patient outcomes are worse. High value
occurs when substantial improvement in patient outcomes occurs at a
reasonable cost. Americans believe in the concept of value and
understand it. For example, they are willing to pay more for many
things--a particular type of clothing, food, service, house or
automobile--because they believe that the utility (happiness,
satisfaction, health, well-being) that is derived from the purchase is
worth the higher price. Cost is a relevant factor, but value is
paramount. So much so, that medical technology needs to be judged in
the same way.
Twenty-five years ago, the science of assessing value in medicine
was rudimentary and underdeveloped. Many of the tools for assessing
value were first applied to health care in the late 1970s and early
1980s. These include patient outcomes research comprising clinical
trials, evidence synthesis (including meta-analysis) and cost
effectiveness analysis. At that time it was uncertain how these tools
would fare in assessing health care. They have undergone refinement by
researchers at universities across the country. Much of this work has
been catalyzed and funded by the Agency Healthcare Research and
Quality. These researchers have sought to create rigorous standards of
high quality research for value science. Teams of clinicians,
epidemiologists, health services researchers, health economists and
others are involved in assessing value. Despite the maturation of and
demand for the science of value, its impact has been limited for three
reasons.
First, there is an unprecedented number of new technologies now
entering into the healthcare marketplace. These technologies earn the
admiration of the world and are made possible from continual progress
in biomedical science. They include minimally invasive surgery,
transplantation of hearts, lungs, kidneys and livers, biotechnology
drugs indistinguishable from natural hormones for patients with
congenital or acquired deficiencies, dialysis therapy for end stage
kidney disease, automatic implantable defibrillators and cardiac
resynchronization to bring life to those with life threatening
arrhythmias and heart failure. Knowledge of the structure and function
of genes and proteins is advancing rapidly and the future will yield
promising technologies we never imagined for identifying, preventing
and treating acute and chronic diseases in an aging population. For
example, genetic tests are now in the making for early detection of
breast cancer, Huntington's disease and Alzheimer's disease. However,
the level of funding for high quality and unbiased value assessments
pales in comparison to the explosion of new biomedical innovations.
To the public, payers and providers, the entry of new medical
technologies into the practice of medicine now seems like a series of
intermittent ``surprise attacks'' on the pursestrings of American
health care. It has been suggested that less than a fifth of all
practices in medicine are subjected to rigorous evaluation and still
less receive an adequate assessment of the cost consequences in
addition to the clinical consequences. We are likely to witness a salvo
of ``surprise attacks'' in the coming years without adequate funding to
do early, comprehensive, balanced and rapid assessments. In a study
with researchers at the AHRQ, I found that medical directors making
coverage decisions for new medical technologies at private healthcare
plans across our country were impeded in their decisions because of
lack of timely effectiveness and cost-effectiveness information.\1\
There is considerable trepidation to decide against covering
potentially useful technology without adequate evidence. Likewise there
is concern about making a coverage decision in favor of a technology
that might later be shown to have minimal benefits at a large cost to
society.\2\ The preference of those making decisions about coverage and
payment for technology was for high quality outcomes research funded by
authoritative government entities.\3\
---------------------------------------------------------------------------
\1\ Steiner CA, Powe NR, Anderson GF, Das A. Technology coverage
decisions by health care plans and considerations by medical directors.
Medical Care. 1997; 35:472-89.
\2\ Boren SD. I had a tough day today, Hillary. New Engl. Jour. of
Med. 1994; 330:500-2.
\3\ Steiner CA, Powe N, Anderson GF, Das A. The Review Process and
Information Used by Health Care Plans in the United States to Evaluate
New Medical Technology. Journal of General Internal Medicine 1996;
11:294-302.
---------------------------------------------------------------------------
Early assessments of clinical and economic outcomes could be
accomplished with investment of a small fraction of annual healthcare
expenditures on value assessments. The payoff would be substantial. For
example, contrary to relentless, direct-to-consumer advertising for
body scans to detect occult disease, my colleagues and I recently found
that screening smokers for lung cancer with helical CT scans is
unlikely to be cost-effective unless certain conditions are met.\4\ The
high number of false positive lung nodules detected by the scans can
potentially lead to more harm from invasive and costly surgical
procedures. We have performed similar cost-effectiveness studies to
guide decision making for detection of mild thyroid gland failure using
thyroid stimulating hormone (TSH) laboratory tests and use of cardiac
ultrasound devices in patients with stroke showing what tests have
substantial value.\5,\ \6\ Early assessments such as these, which
include primary data collection, secondary data collection, data
synthesis, modeling and forecasting would secure information for the
American public and its policymakers in the timely fashion needed to
prevent premature dissemination of costly technology with no or little
value. The Agency for Healthcare Research and Quality as well as the
National Institutes of Health could act as the focal point to bring the
best teams of ``value researchers'' in the country to attack these
issues, by performing clinical effectiveness trials, observational
studies, cost-effectiveness analyses and meta-analyses. If introduction
of some new technologies does not decrease costs, at least through
generation of better and more timely information, Americans can make
sure that what they are purchasing provides good value for the dollars
they spend.
---------------------------------------------------------------------------
\4\ Mahadevia PJ, Fleisher LA, Frick KD, Eng J, Goodman SN, Powe
NR. Lung cancer screening with helical computed tomography in older
adult smokers: a decision and cost-effectiveness analysis. Journal of
the American Medical Association 2003; 289:313-22.
\5\ Danese MD, Powe NR, Sawin CT, Ladenson PW. Screening for Mild
Thyroid Gland Failure at the Periodic Health Examination. A Decision
and Cost-Effectiveness Analysis. Journal of the American Medical
Association 1996; 276:285-292.
\6\ McNamara R L, Lima JAC, Whelton PK, Powe NR. Echocardiographic
Identification of Cardiovascular Sources of Emboli to Guide Clinical
Management of Stroke: A Cost-effectiveness Analysis. Annals of Internal
Medicine 1997; 127:775-787.
---------------------------------------------------------------------------
Early assessments are particularly important given rising numbers
and costs of pharmaceuticals, current consideration of a Medicare
prescription drug benefit and use of tiered pricing arrangements in the
private sector to control drug spending. Tiered pricing is a mechanism
to allow consumers choice in particular drug treatments when they
believe one drug has value over another. However, they must pay more
when choosing to use a more expensive medication. Placement of a
pharmaceutical into a particular tier and patient decisions to buy and
use it are dependent on unbiased information about the benefits and
costs of the pharmaceutical relative to the benefits and costs of
competing medications, i.e. relative value.
Second, as a corollary, funding for career development of ``value
scientists'' needs substantial bolstering to expand the cadre of people
with the capability to perform such research. The AHRQ and the NIH
could amplify training programs focused on preparing and assuring
experienced value scientists to perform this function, just as the AHRQ
and NIH have support training of biomedical scientists who innovate.
Far too few physicians and other health care professionals and
scientists have the necessary training to understand and produce value
science that integrates clinical and economic issues.
Third, understanding how technologies affect cost and value
involves an understanding of the barriers to decision making for health
care providers. Barriers to optimal decision making can lead to
technologies being overused, underused or misused. Physicians are
responsible for most of the decisions in medicine and therefore the use
of medical technologies. My colleagues and I performed a study of the
factors affecting physician decision making with regard to adherence to
clinical practice guidelines.\7\ We found there is a process that must
take place for a new technology to become routine, standard practice.
Physicians must be aware that a new technology exists, agree that it
has value, be willing to try it (adopt) and then, they must adhere to
its use. Lack of awareness leads to underuse. Underuse of an effective
technology can lead to higher expenditures in the future. For example,
if physicians were not aware that in patients with diabetes, urine
protein screening for detection of occult kidney disease and
application of angiotensin converting enzyme inhibitors can delay or
prevent expensive ($>50,000 per year) dialysis treatment for endstage
kidney failure, they might never employ this strategy in their
practice. Fortunately, methods of communicating new information to
clinicians are improving through rapid summary publications (Up To
Date, ACP journal club), clinical practice guideline production by
professional societies and dissemination through electronic means. The
continued proliferation of technology will be even more challenging for
physicians to keep abreast of new technology. Ways for helping them
acquire and assimilate new information are needed.
---------------------------------------------------------------------------
\7\ Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud PA, Rubin
HR. Why Don't Physicians Follow Clinical Practice Guidelines? A
Framework for Improvement. Journal of the American Medical Association,
1999; 282(15):1458-1465.
---------------------------------------------------------------------------
If aware of a technology, physicians must agree with the evidence
that a technology is more effective or safe. If high quality evidence
on representative patient populations is not available, physicians may
disagree on whether the technology provides benefit.\8\ We studied how
early assessments, released through brief clinical alerts that were not
comprehensive influenced the use of carotid endarterectomy.\9\ We found
that clinicians may extrapolate research findings to populations
without clear evidence and indications. Value science can provide clear
evidence.
---------------------------------------------------------------------------
\8\ Cruz-Correa M, Gross CP, Canto MI, Cabana M, Sampliner RE,
Waring JP, McNeilSolis C, Powe NR. The Impact of Practice Guidelines in
the Management of Barrett's Esophagus: A national prospective cohort
study of physicians. Archives of Internal Medicine 2001; 161:2588-2595.
\9\ Gross CP, Steiner CA, Bass EB, Powe NR. The Relation Between
Pre-publication Release of Clinical Trials Results and the Diffusion of
Carotid Endarterectomy. Journal of the American Medical Association
2000; 284(22):2886-2893.
---------------------------------------------------------------------------
Awareness and agreement are necessary for appropriate use of
technology but insufficient. Even being aware and with strong evidence
of effectiveness, physicians may not adopt innovations if there are
administrative barriers to its use or lack of self-efficacy (i.e.
belief in their ability to use the technology to improve outcomes).
They may also adopt technologies with little benefit if payment
policies prematurely promote a technology's use. Financial incentives
in payment policy influence both adoption of and adherence to use of
technologies. We found that providers responded to financial incentives
in payment policy for a biotechnology product (recombinant
erythropoietin) used to treat the profound anemia associated with
kidney disease.\10,\ \11\ Under a fixed, per case payment system,
administered doses of this medication were less than optimal to achieve
the maximal benefit. Changes in payment policies by the Centers for
Medicare and Medicaid Studies were necessary to assure that Medicare
spending was leading to maximal value for recombinant erythropoietin.
Thus, proper use of new technologies means that the physicians who
apply them and the systems into which they are placed are adequately
configured and incentivized to make optimal use of the technology. To
this end, there is a need for more behavioral and systems research that
studies how biomedical innovation from laboratories is optimally and
rapidly translated into interventions to improve the health of patients
treated at hospitals and physicians offices. The AHRQ can play a role
in this regard.
---------------------------------------------------------------------------
\10\ Powe NR, Griffiths RI, Anderson GF, de Lissovoy GV, Watson AJ,
Greer JW, Herbert RJ, Whelton PK. Medicare Payment Policy and
Recombinant Erythropoietin Prescribing for Dialysis Patients. American
Journal of Kidney Diseases 1993; 22:557-567.
\11\ Powe NR. Prescription Drugs in Medicare and the ESRD Program.
Seminars in Nephrology 2000; 20(6):535-5.
---------------------------------------------------------------------------
A final issue affecting cost and value is whether new technologies
supplant older ones and whether technology induces more demand. New
tests do not always replace older ones.\12\ For example, CRP testing is
a new test that could be routinely adopted for assessing heart attack
risk. But it is unlikely to substitute for other tests such as
cholesterol and diabetes testing. Similarly, ambulatory blood pressure
monitors are unlikely to substitute for traditional office-based blood
pressure monitoring. Minimally invasive surgery is an example of a
technology that may induce persons who would otherwise not have a
surgical procedure to undergo an operation. Although these technologies
may not substitute for older traditional tests and may induce further
expenditures through wider use, they may provide health value.
---------------------------------------------------------------------------
\12\ Eisenberg JM, Schwartz JS, McCaslin FC, Kaufman R, Glick H,
Kroch E. Substituting diagnostic services. New tests only partly
replace older ones. Journal of the American Medical Association 1989;
262:1196-200.
---------------------------------------------------------------------------
In conclusion, biomedical innovation has brought the United States
new, unprecedented, medical advances that save and improve the quality
of patients' lives. We need to continue to encourage biomedical
innovation. But we must recognize that for many health conditions,
technologies will bring higher rather than lower absolute costs. Cost
is relevant, but value is far more important. We need to protect
biomedical innovation and the America's purse by furthering the science
of assessing value in medicine. Strengthening our nations' capacity to
perform value science will help private and public payers in this
regard and provide information that physicians and consumers of medical
technologies need to make decisions about their care. The American
people cannot afford to have technology used unwisely. A fraction of
health care expenditures in the U.S. should be targeted to the value
science of medical care.
Thank you for the opportunity to address you today. I would be
happy to entertain any questions you may have.
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]