[Senate Hearing 112-878]
[From the U.S. Government Publishing Office]
S. Hrg. 112-878
MEDICAL DEVICES: PROTECTING PATIENTS
AND PROMOTING INNOVATION
=======================================================================
HEARING
OF THE
COMMITTEE ON HEALTH, EDUCATION,
LABOR, AND PENSIONS
UNITED STATES SENATE
ONE HUNDRED TWELFTH CONGRESS
FIRST SESSION
ON
EXAMINING MEDICAL DEVICES, FOCUSING ON PROTECTING PATIENTS AND
PROMOTING INNOVATION
__________
NOVEMBER 15, 2011
__________
Printed for the use of the Committee on Health, Education, Labor, and
Pensions
Available via the World Wide Web: http://www.gpo.gov/fdsys/
----------
U.S. GOVERNMENT PRINTING OFFICE
87-884 PDF WASHINGTON : 2014
-----------------------------------------------------------------------
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) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001
COMMITTEE ON HEALTH, EDUCATION, LABOR, AND PENSIONS
TOM HARKIN, Iowa, Chairman
BARBARA A. MIKULSKI, Maryland MICHAEL B. ENZI, Wyoming
JEFF BINGAMAN, New Mexico LAMAR ALEXANDER, Tennessee
PATTY MURRAY, Washington RICHARD BURR, North Carolina
BERNARD SANDERS (I), Vermont JOHNNY ISAKSON, Georgia
ROBERT P. CASEY, JR., Pennsylvania RAND PAUL, Kentucky
KAY R. HAGAN, North Carolina ORRIN G. HATCH, Utah
JEFF MERKLEY, Oregon JOHN McCAIN, Arizona
AL FRANKEN, Minnesota PAT ROBERTS, Kansas
MICHAEL F. BENNET, Colorado LISA MURKOWSKI, Alaska
SHELDON WHITEHOUSE, Rhode Island MARK KIRK, Illinois
RICHARD BLUMENTHAL, Connecticut
Daniel E. Smith, Staff Director
Pamela Smith, Deputy Staff Director
Frank Macchiarola, Republican Staff Director and Chief Counsel
(ii)
C O N T E N T S
__________
STATEMENTS
TUESDAY, NOVEMBER 15, 2011
Page
Committee Members
Harkin, Hon. Tom, Chairman, Committee on Health, Education,
Labor, and Pensions, opening statement......................... 1
Burr, Hon. Richard, a U.S. Senator from the State of North
Carolina....................................................... 15
Franken, Hon. Al, a U.S. Senator from the State of Minnesota..... 18
Casey, Hon. Robert P., Jr., a U.S. Senator from the State of
Pennsylvania................................................... 19
Bennet, Hon. Michael F., a U.S. Senator from the State of
Colorado....................................................... 21
Mikulski, Hon. Barbara A., a U.S. Senator from the State of
Maryland....................................................... 24
Blumenthal, Hon. Richard, a U.S. Senator from the State of
Connecticut.................................................... 26
Hatch, Hon. Orrin G., a U.S. Senator from the State of Utah...... 30
Hagan, Hon. Kay R., a U.S. Senator from the State of North
Carolina....................................................... 32
Witness--Panel I
Shuren, Jeffrey, M.D., J.D., Director of the Center for Devices
and Radiological Health, Food and Drug Administration, Silver
Spring, MD..................................................... 3
Prepared statement........................................... 5
Witnesses--Panel II
Hall, Ralph F., B.A., J.D., Professor of Practice, University of
Minnesota Law School, Minneapolis, MN.......................... 36
Prepared statement........................................... 38
Challoner, David R., M.D., Vice President for Health Affairs,
Emeritus, University of Florida and Chair, Committee on the
Public Health Effectiveness of the FDA 510(K) Clearance
Process, Institute of Medicine of the National Academies,
Gainesville, FL................................................ 51
Prepared statement........................................... 52
Curfman, Gregory D., M.D., Executive Editor, New England Journal
of Medicine, Boston, MA........................................ 54
Prepared statement........................................... 56
ADDITIONAL MATERIAL
Response to questions of Senator Bennet by David R. Challoner,
M.D............................................................ 70
(iii)
MEDICAL DEVICES: PROTECTING PATIENTS AND PROMOTING INNOVATION
----------
TUESDAY, NOVEMBER 15, 2011
U.S. Senate,
Committee on Health, Education, Labor, and Pensions,
Washington, DC.
The committee met, pursuant to notice, at 2:34 p.m. in Room
SD-G50, Dirksen Senate Office Building, Hon. Tom Harkin,
chairman of the committee, presiding.
Present: Senators Harkin, Franken, Merkley, Casey, Bennet,
Mikulski, Blumenthal, Hagan, Hatch, and Burr.
Opening Statement of Senator Harkin
The Chairman. Good afternoon. The Senate Committee on
Health, Education, Labor, and Pensions will come to order. This
is the third hearing we have convened as part of our ongoing
process to authorize the FDA user fee legislation. Today we
examine the FDA's regulation of medical devices.
As the sponsor of the American's Disabilities Act, now 21
years old, I recognize that these devices often enable
individuals to live their lives to the fullest. It's hard to
imagine modern medicine functioning without them, and countless
patients who have had their lives changed for the better by
medical devices. Accordingly, it is essential that we encourage
the continued development and improvement of medical devices,
and that efficient regulatory processes get these innovative
devices to patients as quickly as possible. However, there is
no virtue in getting devices to patients quickly if the devices
don't work or, worse, if they cause injury or death.
I think most Americans would be alarmed if they understood
the current process we use to approve most medical devices.
People probably imagine that for every moderate or high-risk
device--certainly anything that's implantable in the human
body--that experts at FDA examine clinical data, and conclude
that the device has been demonstrated to be safe and effective.
But that's not what happens. Most devices are cleared by FDA
through a process in which a device must merely show to be
``substantially equivalent'' to another device that's already
on the market, even if that device was substantially equivalent
to a previous device, and that previous device was
substantially equivalent to a previous device--and on and on
and on and on.
This process gets devices to patients more quickly, but
sometimes with catastrophic consequences for patients.
Recently, for example, all-metal hip implants were cleared
through the 510(k) process, in many cases without clinical
data. As it turns out, when the metal ball rubs against the
metal socket, tiny metal particles can wear off, cause damage
to the bone and tissues surrounding the implant, and the metal
ions can get into the bloodstream and cause problems in the
heart, nervous system and thyroid gland. Today, there are
around a half a million Americans walking around with a
dangerous hip implant, and no great options. Do they have
surgery to have implant removed, or keep it and risk becoming
another victim of the rush to get these products to market?
There's been a great deal of talk lately about promoting
innovation, and I'm all for that, especially when innovation
leads to the creation of jobs and even entirely new industries.
But promoting innovation doesn't just mean, willy nilly,
getting products to market so that device companies can make a
profit. In a recent article, one of our witnesses today, Dr.
Gregory Curfman, made the point that true innovation is not
just the matter of getting products to market quickly, but also
of ensuring that they are safe and effective. A device is only
a worthy innovation if it works, and it doesn't hurt people.
Speeding medical devices to market without adequate data and
testing might be good for business in the short-term. It might
even create some jobs in the short-term. But if the device is
faulty, patients will pay the price, business will be hurt, and
those jobs will disappear. As Dr. Curfman noted in his article,
and I quote: ``Our regulators should not be in the business of
creating jobs in the manufacture of dangerous devices.'' That
is not a good business model at all. I want to do everything
possible to help U.S. manufacturers to create innovative and
safe devices, and get them to market as expeditiously as
possible. To that end, the FDA must strike the appropriate
regulatory balance.
At a minimum, FDA should reserve its streamlined 510(k)
process for devices that are truly of moderate risk. Any high-
risk device should be required to submit a premarket approval
application. Over 20 years ago, in the Safe Medical Devices Act
of 1990, Congress made it clear that FDA should use its
premarket approval process for high-risk class III devices, or
it should reclassify them to a lower-risk category. Despite
this direction from Congress, high-risk devices continue to
slip by this requirement.
If we're going to retain a system in which devices can be
cleared based on substantial equivalence to predicate devices,
we need to create assurance that the predicate device is safe,
and works. We need to follow cleared devices throughout their
lifecycle so that we know how they perform in the real world.
That way, when a follow-on device seeks approval based on a
predicate, we know something about how that predicate worked,
and we will be more confident that ``substantial equivalence''
tells us something about safety and effectiveness. Certainly,
if a device turns out to be dangerous, if it's withdrawn or
recalled for safety reasons, it is absurd to continue to allow
that dangerous device to be used as a predicate for later
products.
So, I intend to work with FDA and my colleagues on this
committee to strengthen and improve FDA's postmarket
authorities, so that we all can have more confidence in the
510(k) system's ability to ensure patient safety.
This afternoon, we'll hear from several expert witnesses
who approach this important issue from a variety of
perspectives. I thank you all for being here, and I look
forward to your remarks.
We have two panels. On panel one we have just one witness,
Dr. Jeffrey Shuren, who is the Director of the FDA Center for
Devices and Radiological Health.
Dr. Shuren became the Director in January 2010. He
previously served as Acting Center Director beginning in
September 2009. Dr. Shuren is both a medical doctor and a
lawyer, having earned his medical degree at Northwestern
University, and his law degree at the University of Michigan.
He has had a long and distinguished career at FDA. He's
held a variety of policy positions at the agency, including
Acting Deputy Commissioner for Policy Planning and Budget,
Associate Commissioner for Policy and Planning, Special Counsel
to the Principal Deputy Commissioner, Assistant Commissioner
for Policy, and a Medical Officer in the Office of Policy.
In the last 18 months, he's led FDA's effort to improve
both the performance of the Center for Devices for Radiological
Health, and the 510(k) review system. We're pleased to have him
here today. We welcome you here, Dr. Shuren. And without
objection, your full statement will be made a part of the
record. And if you could sum it up in, oh, 5, 7 minutes, or
so--we'd certainly appreciate it.
Dr. Shuren, please proceed.
STATEMENT OF JEFFREY SHUREN, M.D., J.D., DIRECTOR OF THE CENTER
FOR DEVICES AND RADIOLOGICAL HEALTH, FOOD AND DRUG
ADMINISTRATION, SILVER SPRING, MD
Dr. Shuren. Thank you, Mr. Chairman and members of the
committee.
As mentioned, I am Dr. Jeff Shuren, Director of the Center
for Devices and Radiological Health at the Food and Drug
Administration. Thank you for the opportunity to testify today.
In late 2009, and soon after I came to CDRH, The Center for
Devices and Radiological Health, we initiated a review of our
medical device premarket programs in response to concerns
expressed by industry and others. We conducted an honest and
frank self-assessment to these processes, including the 510(k)
program.
As part of that process, CDRH has begun to undertake a new,
more systematic approach to device regulation--one that
continues to focus on protecting public health by assuring that
devices are safe and effective, but also focuses on promoting
public health by facilitating device innovation. In fact, last
year, innovation became one of our top four strategic
priorities.
This new approach required that we move away from the
traditional misperception that safety, effectiveness and
innovation are incompatible. Rather than focus on more
regulation or less regulation, we began to focus on smart
regulation--how to most effectively achieve both aspects of our
mission as both a regulator and a facilitator.
We realized that FDA should help to create a regulatory
environment that allows innovation to thrive by eliminating
undue regulatory obstacles, while also ensuring consumer
confidence that our medical technologies are safe and
effective.
In 2010, we released two reports that concluded we at FDA
had not done as good a job managing our premarket programs as
we could have. The No. 1 problem we found was insufficient
predictability which can lead to inefficiencies, increased cost
for industry and the FDA, and sometimes in delays in bringing
safe and effective products to market.
These circumstances may also create challenges for smaller
startup companies in securing venture capital funding for new
early stage technologies.
We identified several root causes of these problems, and
they include very high reviewer and manager turnover at CDRH--
which is almost double that of our Center for Drugs and our
Center for Biologics, insufficient reviewer training, extremely
high ratios of front-line supervisors to reviewers,
insufficient oversight by managers, a rapidly growing workload
caused by increase in complexity of devices, and the rapidly
increasing overall number of submissions we receive, sometimes
unnecessary or inconsistent data requirements imposed on device
companies, insufficient guidance for industry and FDA staff,
and poor quality of submissions from industry.
We identified proposed solutions to these problems, and
after extensive public input last January, we announced a plan
of action detailing 25 specific actions that CDRH would take in
2011 to improve the predictability, consistency and
transparency of our premarket programs, and since then, we've
announced additional efforts.
The actions we are taking fall into three main areas of
emphasis: first, we must create a culture change toward greater
transparency, interaction, collaboration and the appropriate
balancing of benefits and risks; second, we need to focus on
ensuring predictable and consistent recommendations,
decisionmaking and application of the least burdensome
principal; and third, we need to take steps to implement more
efficient regulatory processes, and user resources.
Last month, we reviewed the progress we've made thus far,
and issued a 26-page report summarizing many of the concrete
actions that have already been implemented or will be
implemented, at least in part, in the first half of 2012.
We believe that these actions will have a visible positive
impact within the coming year by providing greater
predictability about data requirements through guidance,
reducing unnecessary or inconsistent data requests through
training, and policy and process changes, implementing policies
that lead to appropriately balanced benefit-risk determinations
using external experts more extensively and effectively,
creating incentives to create clinical studies first in the
United States, speeding up clinical trial approval decisions,
and implementing the innovation pathway.
We understand that in order to best serve patients, both
the medical device industry and FDA must have the flexibility
to be innovative, to be entrepreneurial, and ultimately
successful.
For this to happen, three things must occur. We must
continue to make the critical improvements to our device
program that we described in last month's report. Just as
important, CDRH and industry must work together to assure that
the Center receives high quality premarket submissions. And
finally, CDRH must have adequate and stable resources to get
the job done right, and as quickly as possible. This is the
subject of the upcoming user fee reauthorization.
We at CDRH believe that if these three things are
accomplished, we will provide the kind of value that patients
deserve and come to expect from FDA. Timely access to safe and
effective devices that address their healthcare needs, and the
medical device industry will have the kind of predictable,
consistent, transparent and efficient pathways to market that
spur continued innovation and success.
Mr. Chairman, I commend the committee's efforts, and I'm
pleased to answer any questions the committee may have.
[The prepared statement of Dr. Shuren follows:]
Prepared Statement of Jeffrey Shuren, M.D., J.D.
introduction
Mr. Chairman and members of the committee, I am Dr. Jeffrey Shuren,
director of the Center for Devices and Radiological Health (CDRH) at
the Food and Drug Administration (FDA or the Agency). I am pleased to
be here today to discuss CDRH's premarket review process and the
activities that we are undertaking to improve the predictability,
consistency, and transparency of our regulatory processes.
The Impact of Regulation on Device Innovation
FDA is charged with a significant task: to protect and promote the
health of the American public. To succeed in that mission, we must
ensure the safety and effectiveness of the medical products that
Americans rely on every day, and also facilitate the scientific
innovations that have the potential to save patients' lives. Our
ability to work with innovators to translate discoveries into safe and
effective products that can be cleared or approved in a timely way is
essential to public heath, as well as the growth of the medical
products industry and the jobs it creates. Importantly, FDA's premarket
review of medical devices gives manufacturers a worldwide base of
consumer confidence, both domestically and internationally.
U.S.-based companies dominate the roughly $350 billion global
medical device industry. The U.S.-medical device industry is one of the
few sectors, in these challenging economic times, with a positive trade
balance.\1\ In 2000, the U.S.-medical device industry ranked 13th in
venture capital investment--now, a decade later, it's our country's
fourth largest sector for venture capital investment.\2\ In fact, more
than 62 percent of the $631.4 million that venture capital invested in
the life sciences in the third quarter of 2011 went to medical device
companies.\3\ And, the medical device industry has produced a net gain
in jobs since 2005, even while overall manufacturing employment has
suffered.
---------------------------------------------------------------------------
\1\ PwC (formerly PriceWaterhouseCoopers), ``Medical Technology
Innovation Scorecard'' (January 2011) at page 8, available at http://
pwchealth.com/cgi-local/hregister.cgi?link=reg/innovation-
scorecard.pdf.
\2\ PriceWaterhouseCoopers/National Venture Capital Association,
MoneyTreeTM Report, Data: Thomson Reuters, Investments by Industry Ql
1995-Q4 2010, available at http://www.nvca.org.
\3\ ``Medical Device Developers Attract Cash: Venture Capital
Increases Its Funding of Medical Technology,'' The Burrill Report (Oct.
14, 2011), available at http://www.burrillreport.com/article-
medical_device_developers_attract_cash.html.
---------------------------------------------------------------------------
As noted in a January 2011 report on medical technology innovation
by PwC (formerly PriceWaterhouseCoopers), the U.S.-regulatory system
and U.S.-regulatory standard have served American industry and patients
well. As that report states,
``U.S. success in medical technology during recent decades
stems partially from global leadership of the U.S. Food and
Drug Administration. FDA's standards and guidelines to ensure
safety and efficacy have instilled confidence worldwide in the
industry's products. Other countries' regulators often wait to
see FDA's position before acting on medical technology
applications and often model their own regulatory approach on
FDA's.'' \4\
---------------------------------------------------------------------------
\4\ PwC (formerly PriceWaterhouseCoopers), ``Medical Technology
Innovation Scorecard'' (January 2011), available at http://
pwchealth.com/cgi-local/hregister.cgi?link=reg/innnovation-
scorecard.pdf.
In terms of time to market, an industry-sponsored analysis \5\
shows that low-risk 510(k) devices without clinical data (80 percent of
all devices reviewed each year) came on the market first in the United
States as often as, or more often than, in the European Union (EU). The
EU typically approves higher-risk devices faster than the United States
because in the EU, manufacturers must demonstrate safety and
performance, while in the United States the standard for approval is
safety and effectiveness.\6\
---------------------------------------------------------------------------
\5\ California Healthcare Institute and The Boston Consulting
Group, ''Competitiveness and Regulation: The FDA and the Future of
America's Biomedical Industry'' (Feb. 2011 ), available at http://
www.bdg.com/documents/file72060.pdf.
\6\ See ``Recast of the Medical Devices Directives: Public
Consultation,'' available at http://ec.europa.eu/consumers/sectors/
medical-devices/files/recast_docs_2008/public_consultation
_en.pdf; European Commission, ``Guidelines on Medical Devices: Clinical
Evaluation: A Guide for Manufacturers and Notified Bodies'' (Dec.
2009), at p. 4, available at http://ec.europa.eu/health/medical-
devices/files/meddev/2_7_Irev_3_en.pdf.
---------------------------------------------------------------------------
FDA has been meeting or exceeding goals agreed to by FDA and
industry under the Medical Device User Fee Amendments (MDUFA) for
approximately 95 percent of the submissions we review each year. FDA
completes at least 90 percent of 510(k) reviews within 90 days or less.
In the few areas where FDA is not yet meeting its MDUFA goals, the
Agency's performance has generally been improving--despite growing
device complexity and an increased workload--without a commensurate
increase in user fees.
However, average total days for the review of 510(k)s has been
increasing since 2005 (as described later in this testimony), and has
been increasing for Premarket Approval (PMA) applications since 2004,
with early indicators of longer review times, such as the average
number of cycles to review a 510(k), starting to increase since 2002.
FDA recognizes that, if the United States is to maintain its
leadership role in this area, we must continue to streamline and
modernize our processes and procedures to make device approval not just
scientifically rigorous, but clear, consistent, and predictable without
compromising safety.
Smart Regulation's Role in Facilitating Medical Device Innovation
Nearly 2 years ago, CDRH recognized that, given the growing
complexities of medical product development, we needed to re-evaluate
and modernize our regulatory review processes in order to ensure that
patients had timely access to safe and effective medical devices. At
that time, CDRH began to undertake a new systematic approach to device
regulation, moving away from the traditional misperception that safety
and effectiveness and innovation are incompatible. Rather than focus on
more regulation or less regulation, we began to focus on ``smart
regulation.''
Our goal has been to ensure that safety and effectiveness and
innovation are complementary, mutually supporting aspects of our
mission to promote the public health. As part of our process to improve
CDRH's internal systems, we first reached out to stakeholders to hear
their concerns and listen to their recommendations about our premarket
programs. This is what we heard: industry felt that inadequate
predictability, consistency, and transparency were stifling innovation
and driving jobs overseas; and consumer groups, third-party payers, and
some health care professionals believed that one of our premarket
pathways--the 510(k) program--did not provide adequate protection for
American patients and did not generate
sufficient information for practitioners and patients to make well-
informed
treatment and diagnostic decisions. In turn, CDRH employees expressed
concerns that the 510(k) program had not adapted to the increasing
complexity of devices, and that poor-quality 510(k) submissions, poor-
quality clinical studies conducted in support of PMA applications, and
an ever-growing workload were straining already overburdened premarket
programs.
We also began two assessments of our premarket programs to identify
issues, their root causes, and the appropriate solutions. One
assessment focuses on the 510(k) program. The other looks at how we use
science in regulatory decisionmaking, touching on aspects of several of
our premarket review pathways, such as our clinical trials program. In
addition, we contracted with the Institute of Medicine (IOM) to conduct
an independent evaluation of our 510(k) program.
In August 2010, following extensive public input, we released two
reports that identified issues regarding our premarket programs and
proposed potential actions for us to take to address the underlying
root causes. The No. 1 problem we found was insufficient predictability
in our premarket programs, which can create inefficiencies, increase
costs for industry and FDA, and delay bringing safe and effective
products to market. We identified several root causes of these issues.
They include very high reviewer and manager turnover at CDRH (almost
double that of FDA's drug and biologics centers); insufficient training
for staff and industry; extremely high ratios of employees to front-
line supervisors; insufficient oversight by managers; CDRH's rapidly
growing workload, caused by the increasing complexity of devices and
the number of overall submissions we review; unnecessary and/or
inconsistent data requirements imposed on device sponsors; insufficient
guidance for industry and FDA staff; and poor-quality submissions from
industry.
While it is true that providing more user-fee resources alone won't
solve the problems with our premarket programs, insufficient funding is
at the root of, or a contributing factor to several of these problems.
Adequate and stable funding is one key component to our and industry's
success in bringing safe and effective devices to market quickly and
efficiently.
After considering extensive and varied public input on our
recommendations, in January 2011, FDA announced a Plan of Action that
included 25 specific actions that we would take this year to improve
the predictability, consistency, and transparency of our premarket
programs. The following month, we announced our Innovation Initiative,
which included several proposals to help maintain the position of the
United States as the world's leader in medical device innovation,
including the creation of a new approach for important, new
technologies called the Innovation Pathway.
Since then, we have announced additional efforts to improve our
premarket programs, including actions to improve our program for
clinical trials and the Investigational Device Exemption (IDE) program.
The actions we are taking can be grouped into three main areas of
emphasis. Overall, our actions seek to:
Create a culture change toward greater transparency,
interaction, collaboration, and the appropriate balancing of benefits
and risks;
Ensure more predictable and consistent recommendations,
decisionmaking, and application of the least-burdensome principle; and
Implement more efficient processes and use of resources.
Specific steps that we are taking include:
Issuing guidance clarifying the criteria used to make
benefit-risk determinations a part of device premarket decisions. This
will provide greater predictability and consistency and apply a more
patient-centric approach by considering patients' tolerance for risk in
appropriate cases (draft guidance issued August 15, 2011);
Creating standard operating procedures for when a reviewer
can request additional information regarding a premarket submission and
identifying at what management level the decision must be made. These
steps are intended to provide greater predictability, consistency, and
the appropriate application of the least-burdensome principle by
reducing the number of inappropriate information requests (Standard
Operating Procedures issued November 10, 2011);
Developing a range of updated and new guidances to clarify
CDRH requirements for predictable, timely, and consistent product
review, including device-
specific guidance in several areas such as mobile applications (draft
guidance released July 19, 2011) and artificial pancreas systems (to be
completed by the end of 2011);
Revamping the guidance development process through a new
tracking system, streamlined processes, and, to the greatest extent
possible within available resources, core staff to oversee the timely
drafting and clearance of documents (to be completed by the end of
2011);
Improving communication between FDA and industry through
enhancements to interactive review (some of these enhancements will be
in place by the end of 2011);
Streamlining the clinical trial (IDE) processes by
providing industry with guidance to clarify the criteria for approving
clinical trials, and the criteria for when a first-in-human study can
be conducted earlier during device development. These actions aim to
create incentives to bring new technologies to the United States first
(guidances issued November 10, 2011) (IDEs are required before device
testing in humans that involve significant risks may begin, and they
ensure that the rights of human subjects are protected while gathering
data on the safety and efficacy of medical products);
Implementing internal business process improvements to
ensure that decisions are made by the appropriate level of management,
that decisions are made consistently and efficiently, and that we
appropriately apply the least-burdensome principle. For example, CDRH
created the internal Center Science Council to actively monitor the
quality and performance of the Center's scientific programs and ensure
consistency and predictability in CDRH scientific decisionmaking
(Center Science Council established March 31, 2011);
Creating a network of experts to help the Center resolve
complex scientific issues, which will ultimately result in more timely
reviews. This network will be especially helpful as FDA confronts new
technologies (Standard Operating Procedures issued September 30, 2011);
Instituting a mandatory Reviewer Certification Program for
new reviewers (program launched September 2011);
Instituting a pilot Experiential Learning Program to
provide review staff with real-world training experiences as they
participate in visits to manufacturers, research, and health care
facilities, and academia (to begin in early 2012);
Providing industry with specific guidance on how to ensure
the quality and performance of clinical trials while applying the
least-burdensome principle, so that industry conducts studies that are
more likely to support the approval of their products (guidance
released August 15, 2011); and
Streamlining the de novo review process, the pathway by
which novel, lower-risk devices without a predicate can come to market
(draft guidance released October 3, 2011).
To best serve patients, both the medical device industry and FDA
must have the flexibility to be innovative and entrepreneurial. First,
CDRH must continue making critical improvements to our device program.
Second, the medical device industry and CDRH must work together to
ensure that the Center receives high-quality submissions, which contain
the information we need to make well-informed and timely decisions.
Finally, CDRH must have adequate and stable resources to get the job
done right and quickly. The latter is the subject of medical device
user-fee legislation reauthorization and congressional appropriations.
We believe that the actions we are taking now will have a positive
impact within the coming year by providing greater predictability of
data requirements through guidance, reducing unnecessary data requests
through training and policy and process changes, implementing policies
to appropriately balance benefit-risk determinations, using external
experts more extensively, creating incentives to conduct clinical
studies first in the United States, speeding up IDE approval decisions,
implementing the Innovation Pathway 2.0 (a priority review program to
expedite development, assessment, and review of important
technologies), and instituting efficiencies in the premarket review
process.
Performance Issues in the Premarket Review Process
As noted above, FDA has been meeting or exceeding goals agreed to
by FDA and industry under MDUFA for approximately 95 percent of the
submissions we review each year. FDA completes at least 90 percent of
510(k) reviews within 90 days or less. In the few areas where FDA is
not yet meeting its MDUFA goals, the Agency's performance has generally
been improving--despite growing device complexity and an increased
workload--without a commensurate increase in user fees.
However, MDUFA metrics reflect FDA time only; they do not reflect
the time taken by device sponsors to respond to requests for additional
information. As the graphs below illustrate, while the time FDA spends
reviewing an application has improved (for both low- and high-risk
devices), overall time to decision--the time that FDA has the
application, plus the time the manufacturer spends answering any
questions FDA may have--has increased steadily since 2001.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
FDA bears some responsibility for the increase in total time to
decision, and we have been instituting management, policy, and process
changes to address this issue. As a result, in 2011, CDRH for the first
time began reducing what previously was an increasing backlog of
unresolved 510(k) submissions.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
There has also been a prolonged increase, since fiscal year 2002,
in the percentage of 510(k) submissions requiring an Additional
Information (AI) letter after the first review cycle. The increasing
number of AI letters has contributed to the increasing total time from
submission to decision.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Submission quality problems are a driving force in this increase
and we are pleased that, in response to FDA calls for improving the
quality of premarket submissions, the medical device industry trade
association, AdvaMed, is improving and making available more training
courses for its companies to help them develop 510(k) and PMA
submissions that meet FDA standards.
We believe that the actions we are taking now will have a positive
impact within the coming year by providing greater predictability of
data requirements through guidance, reducing unnecessary data requests
through training and policy and process changes, implementing policies
to appropriately balance benefit-risk determinations, using external
experts more extensively, creating incentives to conduct clinical
studies first in the United States, speeding up IDE approval decisions,
implementing the Innovation Pathway 2.0 (a priority review program to
expedite development, assessment, and review of important
technologies), and instituting efficiencies in the premarket review
process.
Moving Forward: Reauthorization of MDUFA
When MDUFA was last reauthorized in 2007, Congress directed FDA to
take additional steps to ensure that public stakeholders would have
adequate opportunity to provide input to any program enhancements. In
addition to FDA receiving input from stakeholders during an initial
public meeting in September 2010, Congress directed the Agency to meet
with public stakeholders every month while conducting negotiations with
regulated industry to hold discussions on their views about the
reauthorization and hear their suggestions for changes to the MDUFA
performance goals. We have been meeting with stakeholders, including
representatives of patient and consumer groups, since January 2011.
Since last January, we also have been holding discussions with
regulated industry in an effort to develop a package of proposed
recommendations for MDUFA reauthorization. Upon completion of these
negotiations and discussions, the public will have an opportunity to
comment on these proposals prior to our submission of final MDUFA
recommendations to Congress.
As the MDUFA reauthorization process moves forward, it is important
to understand and keep in mind the significant differences between
FDA's medical device premarket review programs--the 510(k) and PMA
programs--and the Agency's program for review of drugs under the
Prescription Drug User Fee Act (PDUFA). PDUFA fees account for about
two-thirds of the drug review program's budget--nearly $568 million in
fiscal year 2010--while user fees under MDUFA fund only about 20
percent of the device review program.
The structures of the user fee programs also differ in very
significant ways. The fee for fiscal year 2012 associated with review
of a New Drug Application (NDA) requiring clinical data is $1,841,500
\7\--much greater than the $220,500 fee \8\ charged for review in
fiscal year 2012 of a PMA for high-risk medical devices (a business
with gross receipts under $30 million qualifies for the ``small
business'' PMA fee of about $55,000--75 percent less than the full
fee). For lower-risk devices cleared under the 510(k) review program,
the fees are even lower: $4,049 per 510(k) application review ($2,024
for small businesses).
---------------------------------------------------------------------------
\7\ See U.S. FDA, ``Prescription Drug User Fee Rates for Fiscal
Year 2012,'' 76 Fed. Reg. 45,831-45,838 (Aug. 1, 2011), available at
http://www.gpo.gov/fdsys/pkg/FR-2011-08-01/pdf/2011-19332.pdf.
\8\ See U.S. FDA. ``Medical Device User Fee Rates for Fiscal Year
2012,'' 76 Fed. Reg. 45,826-45,831 (Aug. 11, 2011), available at http:/
/www.gpo.gov/fdsys/pkg/FR-2011-08-01/html/2011-19335.htm.
---------------------------------------------------------------------------
While we work with industry toward a reauthorization of medical
device user fees in order to provide adequate and stable funding for
the program, we also continue to move forward on CDRH program
improvements, with a focus on smart regulation. As these new policies
and processes continue to be implemented, we expect to see notable
improvements in the consistency, transparency, and predictability of
our premarket review programs.
Smart Regulation's Role in Assuring Patient Safety
As we continue to look for ways to improve our ability to
facilitate innovation and to speed safe and effective products to
patients, we must not lose sight of the benefits of smart regulation to
the medical device industry, to patients, and to society. Smart
regulation of medical devices results in better, safer, more effective
treatments as well as worldwide confidence in, and adoption of, the
devices that industry produces.
We at FDA see daily the kinds of problems that occur with medical
devices that are poorly designed or manufactured, difficult to use,
and/or insufficiently tested. We appreciate the concern that some
devices come on the market in the EU before they do in the United
States. While we want devices to be available to American patients as
soon as possible, we believe that, consistent with U.S. law, they need
to be both safe and effective. The U.S. system has served patients well
by preventing devices from entering the U.S. market that were later
shown to be unsafe or ineffective.\9\
---------------------------------------------------------------------------
\9\ See, e.g., D. Cohen and M. Billingsley, ``Europeans Are Left to
Their Own Devices,'' British Medical Journal, 342:d2748 (2011),
available at http://www.bmj.com/content/342/bmj.d2748.
---------------------------------------------------------------------------
Some have suggested that the United States adopt the medical device
regulatory system of the EU. Yet, outside the United States, pressure
is growing toward greater premarket scrutiny of medical devices. A
recent report concluded that ``[f]or innovative high-risk devices the
future EU Device Directive should move away from requiring clinical
safety and ``performance'' data only to also require pre-market data
that demonstrate ``clinical efficacy,'' and ``[t]he device industry
should be made aware of the growing importance of generating clinical
evidence and the specific expertise this requires. \10\
---------------------------------------------------------------------------
\10\ Belgian Health Care Knowledge Centre, ``The Pre-market
Clinical Evaluation of Innovative High-risk Medical Devices,'' KCE
Reports 158 (2011) at p. vii, available at http://www.kce.fgov.
be/index_en.aspx?SGREF=202677.
---------------------------------------------------------------------------
There are significant differences between the EU and U.S. medical
device review systems. In the EU, manufacturers must demonstrate safety
and performance, while in the United States the standard for approval
is safety and effectiveness.\11\ In the EU, more than 70 private,
nongovernmental entities called ``Notified Bodies'' review and approve
devices by giving them a ``CE mark.'' These decisions are kept
confidential and not released to the public or to EU regulatory bodies.
In fact, the EU does not have one centralized regulatory body. Instead,
each country can designate an entity as a ``Notified Body,'' yet the
decision of one Notified Body applies to all EU countries.
---------------------------------------------------------------------------
\11\ See ``Recast of the Medical Devices Directives: Public
Consultation,'' available at http://ec.europa.eu/consumers/sectors/
medical-devices/files/recast_docs_2008/public_consultation
_en.pdf; European Commission, ``Guidelines on Medical Devices:
Clinical Evaluation: A Guide for Manufacturers and Notified Bodies''
(Dec. 2009), at p. 4, available at http://ec.europa.eu/health/medical-
devices/files/meddev/2_7_Irev_3_en.pdf.
---------------------------------------------------------------------------
Because of these factors, it is impossible to track medical device
approvals, adverse events, or recalls in the EU, since there are few to
no publicly accessible, centralized systems for collecting and
monitoring information about medical device approvals or safety
problems. The use of Notified Bodies has been criticized as encouraging
``forum shopping'' by sponsors to identify those Notified Bodies with
the most lax operating standards, and the varying levels of expertise
among Notified Bodies has been critiqued.
In May 2011, the European Society of Cardiology (ESC) issued a
``case for reform'' of the European medical device regulatory system:
that body's recommendations included creating a unified regulatory
system, imposing stronger clinical data requirements, and requiring
more accountability for notified bodies.\12\ The ESC cited examples of
several different cardiovascular technologies that were implanted in
patients in the EU that were later proven to be unsafe and/or
ineffective through clinical trials required under the U.S. system, and
were subsequently removed from the European market.
---------------------------------------------------------------------------
\12\ See ``Clinical evaluation of cardiovascular devices:
principles, problems, and proposals for European regulatory reform''
Fraser, et al., European Heart Journal, May 2011.
---------------------------------------------------------------------------
Also in May 2011, a series of feature articles was published in the
British Medical Journal, criticizing the opacity of the European
medical device regulatory system, and raising concerns about the
regulation of high-risk devices and how well they are tested before
coming on to the European market.\13\ Several of the featured articles
cited the FDA system's transparency as helping physicians to make
informed decisions about which devices to use and providing patients
with access to information about the devices that will be used on them.
---------------------------------------------------------------------------
\13\ `` The Truth About Medical Devices,'' British Medical Journal,
vol. 342. at pp. 1115-30 (May 21, 2011), available at http://
www.bmj.com/content/342/7807/Feature.full.pdf (Deborah Cohen, ``Out of
Joint: The Story of the ASR,'' British Medical Journal 2011; 342:d2905;
Deborah Cohen and Matthew Billingsley, ``Medical Devices: European
Patients Are Left to Their Own Devices,'' British Medical Journal 2011;
342:d2748); see also Fiona Godlee, ``Editorial: The Trouble With
Medical Devices.'' British Medical Journal'' 2011; 342:d3123, available
at http://www.bmj.com/content/342/bmj.d3123.full; Carl Heneghan et al.,
``Medical-Device Recalls in the UK and the Device-Regulation Process:
Retrospective Review of Safety Notices and Alerts,'' BMJOpen (May
2011), available at http://bmjopen.bmj.com/content/early/2011/05/12/
bmjopen-2011-000155.full.pdf.
---------------------------------------------------------------------------
FDA continues exploring ways to get medical products to patients
with serious and life-threatening diseases or conditions faster, but
lowering U.S.-approval standards isn't in the best interest of American
patients, our health care system, or U.S. companies whose success
relies on the American public's confidence in their products. According
to the IOM, ``FDA should be clear that its role in facilitating
innovation in medical devices is to develop regulatory thresholds that
are rigorous enough to satisfy the agency's primary objective of
ensuring that marketed devices will be safe and effective throughout
their life cycles but realistic enough to permit timely entry of new
devices into the market.'' \14\
---------------------------------------------------------------------------
\14\ Institute of Medicine, ``Medical Devices and the Public's
Health: The FDA 510(k) Clearance Process at 35 Years'' (2011), at p.
197, available at http://books.nap.edu/openbook.php?record_
id=13150.
---------------------------------------------------------------------------
We are pleased that a U.S.-medical device industry trade
association, AdvaMed, has stated that it supports maintaining our
current rigorous standards of safety and effectiveness for marketing
medical devices: ``The medical technology industry has long recognized
that a strong and well-functioning FDA is vital to maintaining
America's pre-eminence in medical technology innovation, and we support
the current regulatory framework in the United States.'' \15\
---------------------------------------------------------------------------
\15\ Advanced Medical Technology Association (AdvaMed), ``AdvaMed
Statement on the House Energy and Commerce Subcommittee Hearing on FDA
Device Regulation'' (July 20, 2011).
---------------------------------------------------------------------------
conclusion
Over the course of the last 2 years, CDRH has been working, with
extensive stakeholder input, to take concrete actions toward creating a
culture change toward greater transparency, interaction, collaboration,
and the appropriate balancing of benefits and risks; ensuring
predictable and consistent recommendations, decisionmaking, and
application of the least-burdensome principle; and implementing
efficient processes and use of resources. These actions--geared toward
a system of smart regulation--have already started to have a
measurable, positive impact on our premarket programs, and we fully
expect that positive trend to continue as we proceed to implement the
improvements we have committed to make.
MDUFA II is scheduled to expire on September 30, 2012, and FDA is
ready to work with you to ensure timely reauthorization of this
critical program. If we are to sustain and build on our record of
accomplishment, it is critical that the MDUFA reauthorization occurs
seamlessly, without any gap between the expiration of the old law and
the enactment of MDUFA III.
Mr. Chairman and members of the committee, I share your goal of
smart, streamlined regulatory programs. Thank you for your commitment
to the mission of FDA, and the continued success of our medical device
program, which helps to ensure that patients and practitioners have
access to safe and effective innovative medical technologies on a daily
basis. I am happy to answer questions you may have.
The Chairman. Thank you very much, Dr. Shuren.
I had reserved some time for the Ranking Member to make an
opening statement, but, we'll hold that. I will begin a round
of 5 minute questions, in the order of appearance.
Dr. Shuren, I've heard a lot of comparisons between FDA's
review times for devices here to those in Europe. Some are
touting the EU's system as one we should emulate; others
express concern about the fact that in the EU, unlike in the
United States, manufacturers do not have to demonstrate that
their products are effective. Can you talk about the
implication of that difference in our approval standards?
Dr. Shuren. It's a very critical difference in our approval
standards. As you mentioned in Europe, unlike in the United
States, you don't have to show that your device is effective--
that, in fact, it really has benefit to patients.
And, on top of it, that it is a system that is not
transparent. The public does not know the basis for approvals,
they don't even know about adverse events, unlike in the United
States where we make that information available to the public,
and the decisions to approve are made by private companies that
are paid for by industry. In fact, there are over 70 of them,
and concerns have been raised about inconsistent oversight,
non-uniform expertise amongst those over 70 different
companies, and form shopping from the manufacturers to those
particular companies. In fact, the same kind of device could be
reviewed by two different notified bodies, and they could ask
for different kinds of information in both cases.
We think the implications are huge--that we should not be
exposing patients to devices that we don't know if they're
effective. We think the U.S. standard is the right standard,
but what we need to do is to assure that there is timely access
to devices that are, in fact, safe and effective, and the steps
we're taking are trying to get there.
In fact, there have been a number of devices that have come
on the market in Europe that later have been shown to be unsafe
or ineffective. Many times when they are doing studies in
support of approval in the United States, we're talking about
heart valves, drug alluding stents, breast implants, device to
detect breast cancer, device to monitor glucose, implanted
wireless heart monitor, elbow implants, abdominal aortic
stents.
In fact, if you have ballooning enlargement of the large
blood vessel in your abdomen, what you want to do is you put in
a stent, a tube, so that the blood goes through that tube
rather against the weakened walls of the artery which puts you
at risk for rupture.
In the EU, those devices have come on the market first,
since at least the late 1990s. And yet, nine of those devices
which didn't come to the United States, came on the market in
the EU, and subsequently came off the market because they were
found to be unsafe, and we don't think that's in the best
interest of patients. We don't think that's ultimately in the
best interest of industry.
The Chairman. What are the weaknesses in your current
postmarket authority? What, if any, additional authorities do
you need to better track the safety of marketed devices? I sort
of referred to that in my opening statement.
Dr. Shuren. This is a question that was also put to us by
the Institute of Medicine which recommended that we take a look
back at our postmarket authorities. We are due to put out a
response to their recommendations and that's currently going
through administration clearance. So, it is an answer that we
hope to get back to the committee very shortly.
One of the issues we, ourselves, put on the table is about
predicates, and are there circumstances where a device that's
currently on the market should no longer be available for use
as a predicate. That is a recommendation that we also plan to
respond to as part of our answer to the IOM report.
The Chairman. Actually, why--this is something I don't
think very many people know--why aren't all implantable
devices, things that go in your body, why aren't they
considered high-risk and regulated as class III devices for
which premarket approvals are required? How is it that implants
like the metal hip implants, and I have a whole number of
others over the last several years that have been done here--
bladder sling, surgical mesh, that type of thing--how is it
they get evaluated through the 510(k) process, I mean things
that are implanted in your body?
Dr. Shuren. In some cases, it does make sense for an
implantable device to be considered under 510(k), if it truly
is moderate risk. And sometimes, we know that upfront.
Sometimes we learn over time about that risk and we realize
that, in fact, what was previously a high-risk device, we make
a low-risk device, and we change that. Our law is based upon
taking risk into consideration in terms of applying what our
requirements are, and we take that very seriously.
But you raise a very good case with the metal-on-metal
hips. Those are devices, right now, that we have brought to an
advisory panel to look at whether or not we should actually
keep them in a class III, and we should put them on a premarket
approval.
In fact, in our efforts to not only apply risk, we also
look to apply the least burdensome principle. I will tell you,
we have done that inconsistently, and we're taking a number of
steps to address that.
But, in one of those cases, for the metal-on-metal hips,
and this was the Depuy device, we, under the least burdensome
principle, decided--you know what--we're not going to ask for
clinical data in this case. And it turns out these are always
tradeoffs and judgment calls as we try to figure out what the
least justified burden is to impose on the company--that's a
scientific decision--and sometimes we slide to a place, and it
turns out not to have been the right call, and in the case of
that particular hip, there were failures that we might have
detected if we had asked for other kinds of information.
The Chairman. I have a couple of followups. My time has run
out. I'll do that in the next round.
Senator Burr.
Statement of Senator Burr
Senator Burr. Thank you, Mr. Chairman. Dr. Shuren, thank
you for being here. I'm sorry I missed part of your testimony,
but I read the written part.
Let me ask you, why did it take 2\1/2\ years to put all
these new proposals on the table that you highlight will speed
up the process?
Dr. Shuren. Actually, it didn't take us 2\1/2\ years to put
them on the table. I will tell you, when I came in late 2009,
and having heard concerns about the program, the very first
thing I said, the best way to address this is let's assess it
because we had industry saying that the program was not
sufficiently predictable, consistent, transparent. It was
stifling innovation and driving jobs overseas. But we were
hearing from some of the consumer groups, third party payors,
some of the healthcare professionals, that, in particular, the
510(k) program was letting unsafe devices on the market and not
providing good enough information to make well informed
treatment and diagnostic decisions.
My own staff had concerns that the current instantiations
of those programs were not well-suited to some of the new
technologies.
Senator Burr. We're 2\1/2\ years down the road since then,
and now, I don't want to cut you short, now your testimony is--
we have all these things that we're just putting in place, and
they're going to solve a lot of the questions that have been
raised.
Let me read you your quote, tell me if it's accurate.
``Ninety-five percent of more than the 4,000 medical
device applications subject to user fees that FDA
reviews every year are reviewed within the goals that
were agreed to by the medical device industry under
medical device user fees amendment 2007.''
Is that an accurate statement?
Dr. Shuren. Yes, for the user fee goals. But I will tell
you, in spite of that, no one's happy with the program because
quite frankly, we're seeing that the total times are going up;
so, while our times have gotten better since the start of that
program, the total times aren't there, and so industry's
unhappy, we're unhappy.
Senator Burr. Let me tell you why some of the unhappiness
exists because that statement, one, is misleading. It's
misleading, first, because the FDA has not reported 510(k)
performance data beyond 2009, second, the device center has
failed to achieve 55 percent of its user fee performance goals,
and third, user fee performance goals are based on a metric
called FDA days. FDA days accrue when FDA, not the
manufacturers working on an application to reduce FDA days, FDA
can offload the work onto manufacturers.
Let me just ask you simply, would you be in favor of going
to calendar days? I think this would bring transparency and
clarity, and better understanding of communications.
Dr. Shuren. Sir, in terms of days, let me say for the FDA
days, that's what we agreed to with industry, and actually was
the deal then supported by Congress, and we've been following
that and reporting on what we agreed to. But as part of our
user fee reauthorization discussions now, let me premise by
saying we've got ground rules in place, so please understand
when there are limited things I can say about it, but I can
talk about what's out there publicly.
We've been talking about changing those goals to total
time. Now that means that there are things you're going to have
to expect from FDA--we're responsible for part of that time,
industry is responsible for the other part of the times. Part
of those discussions----
Senator Burr. Maybe I was unclear in my question. My
question was simply this, would you be in favor of switching to
calendar days from FDA days?
Dr. Shuren. If we're talking about total time? By calendar
days, you mean calendar days where it's just with FDA?
Senator Burr. Clock starts--doesn't get restarted, it
continues to tick.
Dr. Shuren. Oh, we're talking about the same thing, total
days.
Senator Burr. OK.
Dr. Shuren. So, that's what we're talking about as a part
of user fee reauthorization.
Senator Burr. How do you explain the fact that since 2007,
the original PMAs--the length of time that it takes to get
processed has gone up 75 percent, to 27.1 months? The average
number of months for clearance of a 510(k) since the 2003-7
period has gone up 43 percent, to 4\1/2\ months' clearance.
Tell me where the companies, based upon their user fees, were
benefited in this process?
Dr. Shuren. Yes, so first, in terms of--I'm not going to
quibble over times, we may have disagreements. But, first of
all, the problems----
Senator Burr. Tell me where I'm wrong.
Dr. Shuren. The problem started----
Senator Burr. No, tell me where I'm wrong, don't just say I
disagree, show me where my information is inaccurate.
Dr. Shuren. Oh, for the timeframes? Yes, some of the
timeframes are off in terms of the increases. But I would say,
so what? So what? We all agree that times are going up, that's
the important thing. And if you look at what's been going on,
it started in 2002.
Senator Burr. So, what's the answer to it, user fees?
Increased user fees would eliminate this?
Dr. Shuren. What's that? If you just throw money at it, you
won't solve it. You need to do three things. We need to make
improvements to the program--that's what we've been doing over
the course of 2011. We do need to work with industry, and we're
doing that right now with their representatives on assuring
we're getting quality submissions.
We're working on criteria for when we would not accept an
application. It's exactly what the drug program does right now,
and in fact, in a report we just put out on performance in the
drug program, where we're now getting new tech, new device, new
drugs, rather, on the market fairly quickly, one of the driving
forces was the fact that companies are submitting higher
quality submissions. We get it done faster.
And the third part to it is we need to have the adequate
resources to get the job done. If you just throw money at this,
you won't solve it. But, if we don't have the adequate
resources to get it done, we will fail, and ultimately,
industry will fail, and that's why we moved forward on these
assessments, which we wrapped up in 2010. It involved lots of
input from the outside. In fact, not only did industry ask, but
many Members of Congress came back to us and said, please don't
rush to judgment--make sure you get lots of input.
We tried to push forward quickly because we knew changes
needed to be made. But we got from the Hill and from industry,
don't do anything until everyone's had enough opportunity to
weigh in. Then, we start in 2011 actually implementing, and now
people are upset we're not moving fast enough.
And, as it is, as a Federal agency, quite frankly, we get
public comment on our policies. Now, that's good because we get
lots of input, but it takes time. I mean, we live in a
fishbowl. If a company were to try to make changes, and do it
in a public process, they'd never get anything done. This adds
time to what we're doing, and already, though, we're starting
to see some things change.
Let me show you something on 510(k), because you asked
about it, if I could actually get chart six. I'd mentioned the
problems here actually started in 2002. It's when we started
seeing the first indicators where there was going to be reduced
performance, and 510(k) started to actually increase in total
time around 2004.
This shows the number of 510(k)'s that are still
outstanding at the end of the year. You can think about it as a
backlog. We're looking from 2005--steady increase, year after
year after year. But, now, finally, in 2011, as we're starting
to make some changes, that backlog is coming down.
If I can take the slide seven, what you'll see here is the
percent of 510(k)'s that we make the decision to allow on the
market. We call that clearing the device, and you'll see,
again, back up, this is now with 2004, look at the top line,
the percents in clearances is 88, it's just steadily coming
down--year after year, 2011--the first time that number is
starting to go up.
Now, these are early indicators. We still have a long way
to go. But much of the actions we're taking are still out
there, draft policies for comment or draft process changes that
are out for comment--that, in the coming months will be
finalized and implemented, and yes, I do think, that we'll have
a positive impact.
Senator Burr. Chairman, you've been very accommodating, and
I'll look forward to the next round.
The Chairman. Thank you very much, Senator Burr.
Senator Franken.
Senator Franken. Thank you, Mr. Chairman.
The Chairman. I just want to say in order--Senator Franken,
Senator Casey, Senator Merkley, Senator Bennet.
Statement of Senator Franken
Senator Franken. Thank you, and thank you, Dr. Shuren, for
being here, and I appreciate the conversations that we've had,
and I appreciate your going out to Minnesota. I understand
you're going out again soon.
Dr. Shuren. Yes, that's correct.
Senator Franken. I think we'd all agree that patient safety
is a priority of Congress and the priority of the FDA, and it's
my job, and the job of the HELP Committee, to help you to
protect patient safety to the best of our ability. As the
Chairman said, there are different perspectives on all of this.
I've spent many hours talking with patients across
Minnesota, and they tell me that they want access to the newest
treatments for their conditions, but too often, these devices
haven't been approved. When I talk to medical device
manufacturers in Minnesota, they tell me they're frustrated
that they are developing innovative devices, but they can't get
them to patients because the FDA hasn't approved them.
That's why earlier today, I introduced the Patient Access
to Medical Innovation Act with my colleague, Senator Alexander.
This legislation will get devices to the market faster and more
safely, I believe, by allowing the FDA to consult with experts.
It will also reward companies for developing devices for
patients with rare conditions.
I believe this bill is a step toward making the process
more efficient, and part of this is talking about what some
consider--a lot of groups consider--the overly restrictive
conflict of interest rules that exist in terms of the experts
who have been in the industry--and you've talked in your
testimony about the attrition that you've had at the top of
experts, and so, that's what my legislation addresses, in part.
When Commissioner Hamburg testified here in July, she
expressed an openness to working with us on this issue in
committee, would you be willing to work with us as well?
Dr. Shuren. Yes, absolutely. We consider having access to
the right experts to be critical, and so, I'm very happy to
explore the issue with you. I'll add it's one of the reasons
why we're setting up a network of experts.
Just a few weeks ago, we put out standard operating
procedures for working with healthcare professional societies,
scientific organizations, to be able to tap into their networks
that when we are dealing with challenging scientific questions,
that we can rapidly identify who are the experts in the field,
kind of supplement who the industry may be bringing in with
experts and experts we may already have, so that we get a full
discussion, we get the right people at the table to help us
address those tough questions.
Senator Franken. I really appreciate your working with Dale
Wahlstrom in Minnesota, in talking about creating a center for
studying regulatory science because that's really what we're
talking about here.
Dr. Shuren. That is very fair to say.
Senator Franken. OK. I just want to get into the 510(k)
proposals that the IOM made, and I think you and I are in basic
agreement on things that we didn't like about it. Can you talk
a little bit, though, about the things you did, like including
postmarket surveillance and that kind of thing, and what you
didn't.
Dr. Shuren. I'm a little ham-strung at the moment.
Senator Franken. OK.
Dr. Shuren. As our response to the report is currently in
administration clearance. But what I can say is, the IOM
engaged in a thoughtful report. There's a lot of things in
there, and to their credit, they also tried to look at a number
of issues beyond just 510(k) at things that might affect the
program like postmarket. They looked at software, and I think
in the debate, while there was one key recommendation on 510(k)
program where we felt, no, we shouldn't get rid of the program
in its entirety. There are many parts to that report that
really merit a good and thorough conversation, and in our
response, we'll kind of come back with what our perspective is
on all of the recommendations that the Institute of Medicine
made.
Senator Franken. I look forward to seeing that, and, in
fact, you said, ``FDA believes that 510(k) process should not
be eliminated.'' That's pretty, pretty clear.
Dr. Shuren. And I stand behind that.
Senator Franken. Thank you, Mr. Chairman.
The Chairman. We'll have another round. Thank you very
much, Senator Franken.
Senator Casey.
Statement of Senator Casey
Senator Casey. Doctor, thanks so much for your testimony
and for your public service. These are hard issues.
I wanted to ask you about, a kind of basic workforce issue.
I'll start with a kind of a fundamental question about what
some of the data has shown. Isn't it true that the data shows
that the turnover rate is much higher for medical device
reviewers than drug reviewers--is that correct?
Dr. Shuren. That's correct. It's about double, and in fact,
our drug and biological centers had this same problem many
years ago, and same reasons--too much workload for the staff,
not enough management oversight. In fact, for our premarket
approval offices, the ratio of front-line managers to staff in
some cases, are as high as 1:27, which is untenable, and they
finally solved that through user fee discussions, and the drug
industry, who, 10 years ago was not happy with the program
rolling into PDUFA III, much like we're rolling into MDUFA III,
and they finally said, ``You know what, we're not happy with
performance, but we recognize if we're going to address a lot
of these problems, we have to make sure that you are
sufficiently funded to get it done,'' and with us, some of
these problems with high turnover, we're not going to solve if
we don't have enough people to do the work, and enough managers
to assure that we have good adult supervision.
Senator Casey. That leads me to a question--I would like to
know what direction you are heading in, and what FDA is doing
to make an investment in its reviewers? How do you deal with
this basic problem of high turnover?
Dr. Shuren. One step is to make sure we provide them with
good training opportunities, and to make the processes as
streamlined as possible.
Let me tell you about training. One of the things I
encountered when I came to CDRH is that we had no standardized
training program for our new reviewers. So we have now
instituted and just launched in September a reviewer
certification program where all new reviewers will go through
standardized courses, and they go through oversight of the
applications that they're reviewing. We're going to audit that,
see if it works, and from there, think about expanding it, if
appropriate, to some of the other reviewers as well.
We're also getting ready to launch what we call an
experiential learning program. We think one of the ways to help
keep us on top of new technologies is to get us out of the FDA
to actually get our reviewers, get our managers to go out to
manufacture facilities, see the new technologies that are being
developed, go out to healthcare facilities, see the
technologies in real world use.
Right now, we've tried it on more of a pilot basis and
we've gotten very good feedback. It is my hope as we mentioned
with resources, that we'll also have not only enough people to
get, do the work in a timely manner, but also that our people
can also take time off to go get this training. Because one of
the challenges my staff face everyday is a gallon workload and
they're getting judged on the workload and they're moving the
workload at the same time. We'd like them to leave the workload
for a little bit of time, and get out of the office, and get
that training, and that means we need enough people to both
move the freight, if you will, and to let people take the time
off to get such critical training.
Senator Casey. I have a related question. In some fields
such as healthcare, there is a lot of turnover and often it's a
burnout factor. Along those same lines you just described, what
are you doing about recruiting and staff retaining? Because
sometimes you're going to have a problem with both. Your
recruiting can go well, they get in the door, but you can't
retain them. In other cases, it's the opposite or sometimes a
combination of both. What are you doing about that? Have you
already answered that or is there something additional you'd
like to add to that?
Dr. Shuren. The critical issue here comes down to funding.
I will tell you right now for one of my offices, premarket
office, it takes about three people to bring on board to
actually get a net gain of one person because of the turnover
rate, and in some cases, I can't really pay for people, for
some of the top-notch people to come out from the private
sector, to come to the FDA and stay at the FDA. It's one of the
issues they also addressed in the drugs program.
Right now, it's on the table that we're talking about as a
part of user fee reauthorization. But, I really would like that
ability to be able to pay some of these very highly talented
people to stay in their jobs rather than leaving. In fact,
right now, almost half of my reviewers have 4 years or less of
experience, and from my front-line managers, more than half, a
little over half, have only 3 years or less experience, and
that turnover--oh, I was saying that almost half of my
reviewers have 4 years or less experience because they wind up
leaving. And from my front-line managers, little over half have
about 3 years of experience or less. And we know even from
industry that these disruptions and new people coming through
disrupts the review that's going on.
At the same token, I can do all the training in the world,
but if these people are leaving, I've lost that, if you will,
that investment. It's one of the challenges I have on least
burdensome principle, which, by the way, I'm very committed to.
I mean, to me, I consider that good government, that you're
looking at what is the least justified burden that you impose.
But, if I'm constantly dealing with new people, it's much
harder to get them up to speed on what the expectations are
when we apply that principle, and I think if we can break this
cycle, we'll win.
It's one of the reasons why, and I empathize with industry,
by the way--they are paying more money, they're not seeing the
kind of performance they want to see. But quite frankly, what
we never tackled is making sure we have enough resources, not
only to handle the workload, but to actually get over this hump
of too much workload for the individual people, and not having
enough managers. And if we can break that cycle, then we will
have a program where people stay, they get trained up, they
stay. We have enough people to do the work. We make the other
program improvements in policies and processes, then we have a
top-notch medical device program.
And last, and let me say, I am not trying to put down my
staff. They are amazing. I have very talented people, but I
don't think we've quite served them well in making sure they
have the tools available, the oversight available, and the
opportunities available where they can really thrive and have a
good work place.
Senator Casey. We've heard a lot about this in
Pennsylvania, so I appreciate your answers. Thank you.
The Chairman. Thank you, Senator Casey.
Senator Bennet.
Statement of Senator Bennet
Senator Bennet. Thank you. And just to pick up where
Senator Casey left off, we've heard a lot about this in
Colorado, too, and my understanding, Dr. Shuren, is that,
actually, you even lose people to other parts of the agency
where people are able to compensate better than the device
section.
Dr. Shuren. That's correct.
Senator Bennet. I want to thank you for all the time you've
spent with me, and in Colorado as well. I deeply appreciate it,
and as you know, Senator Burr and I on the committee introduced
medical device legislation, along with Senator Klobuchar. And
we hope to make headway here on the HELP Committee, Mr.
Chairman, on this legislation.
A lot of the bill stem from the idea that FDA should not
make approval longer or more costly than is necessary to meet
the statutory standard of reasonable assurance of safety and
effectiveness. I certainly think a pragmatic approach like this
makes sense. And it seems the FDA has acknowledged this need
for pragmatism as well. I'm referring to the least burdensome
guidance document from 1997, and you just raised that concept
in the context of training. So, I'd like to hear you a little
bit more on this least burdensome concept. How important is it
to the agency. Is it a priority for you? And in addition to the
training that you just talked about, because I don't want you
to have to repeat yourself, how can FDA better apply this
concept?
Dr. Shuren. So, it is important for me, and it's something
that I, personally, am committed to. In fact, over this summer,
I sent out a central-wide email reaffirming our commitment to
the least burdensome principle.
In fact, for folks who may be interested, that report I
mentioned from about a month ago--and there's no new activities
in here--it's encapsulating things that we talked about
beforehand, but what we put on our Web site is an attempt to
link all the different actions together that we're taking while
we're taking them. And you will see in there, in our top three
objectives, one of them actually pertains to least burdensome
principle. And along with it, even a chart that lays out
everything we're doing. If we put something out, we actually
provide the link to that information. If not, we give a
timeframe for when we think it's going to come out.
For least burdensome, one of the things we've been doing as
a change is, first of all, in our guidance documents--try to
more and more, in the guidance documents, themselves, apply the
least burdensome principle as opposed to just say something in
front as a boilerplate. It's really critical we put that in the
guidance themselves, and we've done several of those over the
past year, so it's very clear for our staff, and it's also very
clear for industry.
The other is assuring that when we're making decisions, and
we're going to, if you will, ask for something more, or
something different, that that decision is being made at the
right level of management. Too often, that decision was being
made at the reviewer level. So, we have now put out a change in
our standard operating procedures saying for these kinds of
decisions, they've got to be made at this level. And, again,
another check is to make sure that we are applying the least
burdensome principle.
In fact, a few months ago, we created a center science
council comprised of our senior leadership. It includes myself
and experienced scientists who now oversee those programs,
again, for science programs for ensuring consistency,
predictability and application of the least burdensome
principle. Some of the most important scientific questions are
coming up there. And I will tell you that in some cases, we've
actually come back to say, we're going to do something
different than what was recommended. In part, it's an
application of the least burdensome principle.
Senator Bennet. I want to thank Senator Burr for his help
on this, and also say how much I'm looking forward to working
with the agency to try to get this done.
In your testimony, you cited some statistics about venture
capital investing, and seemed to be suggesting U.S.-venture
capital investment medical devices is healthy. I want to call
your attention to a recent survey of life sciences, VC
investors by MVCA, which found that our venture capital
companies are deciding to use their dollars to invest overseas
in places like Europe and China rather than here because of
their lack of confidence in our regulatory system.
I want to mention the importance of first-time funding
statistics. It's the measure of how many new technologies are
being taken directly from the lab and to development, that
first-time funding is down. That means in the future, there
will be fewer technologies advancing toward the market at all
stages of the pipeline. According to the Pricewaterhouse
MoneyTree Report, first-time fundings of medical device
companies fell more than 60 percent from 2008 to 2010, and
total medical device VC investment fell by more than $1 billion
during that same time period. So, in light of that data, I'd be
interested to know whether you think we ought to be worried
about those trends.
I'll say that I'm worried about it for two reasons. One is
that, I actually don't think there's a place where it's about
balancing safety and innovation. Actually, we need the
innovation for the sake of our patients, which is my principal
concern. The other is in States all across the country,
Minnesota's one, Colorado's one; we really see much of our
economic future in the development of these innovations. And
so, that's why you're hearing from the two of us and others
about this. And I want to make sure that we've got an ecosystem
in this country where venture capital is being invested here,
and not going to Asia instead out of frustration. Do you want
to respond to any of that?
Dr. Shuren. Yes, I want to see the venture capital dollars
flowing here in the United States as well. I'll tell you, one
of the things I heard from the VC community, and quite frankly,
I've been spending a lot of time traveling around the country.
Rather than expecting people to come to Washington, DC, I've
been going out there, and I've been out to Colorado----
Senator Bennet. And I can vouch for that.
Dr. Shuren [continuing]. Minnesota. Many times, been to
North Carolina, and all of this is to really hear from people
directly. We've held town hall meetings where anyone from the
public can come.
One of the things the venture capitalists have said is,
where they'd like to have the ability to start clinical trials
in the United States earlier. And the reason is, a company will
bring their device to a group of doctors. They'll do some
testing, and when they come back to do more testing, they go
back to the same group of doctors because those are the doctors
who have experience with the device.
So, these early tests, they're called early feasibility
tests--are so important. That's why just last week, we put out
a new policy on these early feasibility, and in some cases,
first in-human studies, that would allow them to start in the
United States earlier in the device development pathway than
has occurred previously. And to allow companies to make changes
to that device without necessarily first having to come back to
FDA. Because as you know, with a device, the early prototype is
not the one that's going to eventually be sold on the market.
They test it, they learn from it, they make changes. We're
trying to create a more rapid innovation cycle here in the
United States. And we think if we do that, that becomes a
greater incentive for the VC community to invest here, and I
think that is good for small companies who are really putting
up and developing these new technologies.
Senator Bennet. Thank you, and thank you, Mr. Chairman, for
your indulgence.
The Chairman. Senator Mikulski.
Statement of Senator Mikulski
Senator Mikulski. Good afternoon, Dr. Shuren. I'm glad to
see you and to welcome you to the committee. I want you and the
people who work at the FDA--the several thousand, almost 7,000,
who work at FDA in food, prescription drugs, and medical
devices--to know how proud we are of the job they do every day.
The fact that we are as safe as we are in regard to our food,
drugs and medical devices says something must be working. But I
think we all agree it must work better.
I want to go to the personnel issues and rapid turnover
raised by Senator Casey. What is the reason for the rapid
turnover? Is it all money? Or is it also the fact that Federal
employees--in particular, FDA employees--have often been the
subject of a sustained, relentless, and persistent demonization
as ``pointed-headed PhDs'' ``Pointed-headed bureaucrats--
technocrats? '' Maybe I'm wrong.
People come to FDA because they want to make a difference.
Why then, don't those same people stay in their jobs when every
day FDA is actually making a difference?
Dr. Shuren. You're spot on. First off, we did an
organizational assessment of CDRH in 2010, and found that the
commitment to the mission is actually off the charts. Over 90
percent of my staff, committed to the mission, were willing to
put in extra time, which is beyond what you see in most
industries.
But at the same time, morale is hurting. And the No. 1
thing I hear--I just had one of my office directors back in my
office again a few days ago--was this issue of demonizing.
That, for so long, all they have heard from the device
industry, all they've heard, and I really do not mean this as a
negative comment, but from many colleagues here on Capitol
Hill, much has been focused on what's wrong. And my people
everyday are working hard, and if all they hear is, ``you're
doing a bad job, there's something wrong with you,'' that kills
morale.
In fact, it is really hurting my ability to try to drive
change at the FDA. It's one of the reasons, as Senator Franken
mentioned, Dale Wahlstrom at LifeScience Alley, he and I have
been talking about--how do we change that dialog? Because if we
don't do that, if we don't change the atmosphere, if our people
actually don't hear about when they're doing things right--if
the things we're doing I'm talking about today makes sense, my
folks need to hear that. Because, otherwise, we won't be able
to make those changes, we won't change the morale issue, and
we'll continue to lose people. I've got to fix the morale, and
we will have to have the people to reduce that workload, and
also make sure I've got the managers.
Senator Mikulski. Well, first of all, I am also deeply
troubled by the demonization. It has shifted focus away from
the regulatory process and the IOM report. In other words, the
focus is on demonization rather than acquisition of valid
information from either industry or learned societies through
debate, discussion, and ultimately problem-solving.
But let's also talk about the issue of money. Does the talk
about 2-year pay freezes and other monetary proposals being
directed at Federal employees also exacerbate the discontent?
Dr. Shuren. I think it's a contributor. I mean, on the flip
side, though, quite frankly, everyone knows the tough straits
we're in as a country. There are people without jobs today,
there are people who can't find work. And my people get that,
they really get that.
Senator Mikulski. We're facing a joint committee report. If
the joint committee fails to act, or we fail to adopt the joint
committee recommendation, we will go to a sequester. The
sequester, though not until next July, would mandate what I
believe is an 8.5 percent across the board cut.
What would the impact be on your operation in light of such
a cut--the literal functionality of it--regardless of how well,
able and bi-partisan our MDUFA reauthorization is? What would
the impact be on you? I'm not asking about which three jobs
you'd need to lay off. But, instead I'm asking what you think
the impact would be on your ability to recruit and retain
during this year, facing both a sequester and the current
demonizing?
Because first, there's the specter of demonization, which I
think is just wrong. I just think it's wrong. Quite frankly,
it's unbecoming for an American democracy to not recognize that
we need an independent civil service. One of the hallmarks of
great democracies is that they all have an independent civil
service. Regardless of who is in charge, independent civil
service serves the Nation.
Then, there's also the fact that there's the sequester.
What would the impact of both of these concerns be on morale
and recruitment and retention?
Dr. Shuren. It would have significant adverse impact. I
mean, our program, first off, unlike the drug program, is
predominantly dependent upon congressional appropriations, that
is where we get most of our money. And a big cut means that,
not only are we at risk of losing people, but people know that
the workload goes up, so the work environment deteriorates. We
won't be able to, then, bring in people to handle that
workload. I won't be able to keep my good people there, the
first people who leave, and, ultimately, the program spirals
downward.
Senator Mikulski. I know my time is up. Let me just say
this. First of all, I'm so glad to hear you want to go out and
meet with those who are working on innovations in technology
and manufacturing. The reason so many of us are here today is
that those types of industries create great jobs in our States.
People working in my medical device community are so excited
that they're actually building a product that saves or improves
lives.
And their work creates an export product, so they're
excited about that too. We've got to work together on this
issue. This is really, truly an opportunity for jobs,
innovation, and exports. I think what we need is to value our
civil servants, and let them know they can count on us. If we
do that, then we can count on them as we go through this
process.
On behalf of myself and Senator Cardin, I would like to
thank our employees at FDA. We're really proud of you. We've
got a lot of reform ahead and a lot of self-reflection to do.
But I think self-reflection starts up here in the Senate as
well.
The Chairman. Thank you, Senator Mikulski.
Senator Blumenthal.
Statement of Senator Blumenthal
Senator Blumenthal. Thank you, Mr. Chairman. I want to join
Senator Mikulski in two points that she made so well. The first
is that the folks at the FDA, the people who work so hard and
well are really doing an enormous public service for the
American people. And probably, their only fault, and it's not
their fault, is there should be more of them to do the work
that's required.
And second, that demonization never makes sense in a
democracy, by definition, it makes no sense because to demonize
is, essentially, to shift the blame and focus it unfairly.
But, I want to raise a more philosophical question which
comes to me from my work at the State level where we have
agencies that are funded in whole or in part by the industries
that they are supposed to regulate. And, in terms of public
credibility, and long-range outlook, I realize it may not be an
issue for this reauthorization. Aren't we doing a disservice to
the FDA by making it as dependent as it is on funding from the
industries, not just device industry but pharmaceutical and
others, that it is supposed to regulate?
Dr. Shuren. It's an important philosophical question. I
also look at it, though, as a pragmatist. I know the near
impossible task that faces this Congress, about figuring out
where the dollars go. You have to make tough decisions all the
time. We know there's less money available than before, so,
realistically, for us to look at the U.S. Congress as, we're
going to continue to see the kind of increases we may need to
support the program. I don't know if you're in a position to do
that.
So, on its alternative, there are user fee dollars to
support us and to make sure we can get the job done. What's
critical when we set up these programs is that those dollars
support the overall program. There's not like a cut check to
individual people, so we kind of keep it separate from, if you
will, the dollars coming in, and the decisions being made.
And the second thing is that, on the flip side for
industry, I know for them, though, it kind of changes the
expectations that they have on the program. What I would like
at the end of the day is to have an adequately funded program.
I'm almost, I will tell you, personally, agnostic on the
source. As long as we have the people we need----
Senator Blumenthal. Oh, you're agnostic on the source. I
don't mean to interrupt you unfairly, but, the answer you've
given is sort of the second part of an answer that you haven't
given, which is, in the best of all worlds, you'd be funded
independently, but given the fiscal constraints of the moment,
pragmatically and realistically, as you've put it, probably,
this source of funding is the one that we need to rely on.
But, in terms of public credibility, I just wonder whether
we aren't doing the agency a disservice by making it so
dependent on the industry's funding. Certainly, it is raised,
commonly, and the public is increasingly aware of it--that it
depends on the industry. And frankly, at the State level, we
have it happen, and the same reaction is prevalent--whether
it's the utility industry or the banking industry, or the
insurance industry--as happens at the State level.
Dr. Shuren. Yes. And I recognize, also, perception can just
as much be a reality as facts are. And I know that there are
many who feel that, because we receive user fees, that that may
taint the decisions that we get.
All I can tell you is, I don't believe it does taint our
decisions. And, however, that funding is provided, as long as
we are free to make our independent scientific determinations,
and we have the funding to get the job done, and done right and
quickly, then I think we're in a good place.
Senator Blumenthal. You may not have the time, I'm
guessing, almost certain that you don't have the time, but what
I would like you to give me, perhaps in writing, is your
analysis of the three instances of FDA failure. And what you
think we should learn from them going forward. In other words,
case studies--I'm not asking you for a full case study, but
what you would point to as what we should regard as failures--
and the three success stories. And if you want, you can make
them five, you can make them seven on each side.
You can, but what I would like is an analysis of where you
regard the failures as having occurred, and what we should
learn from them. Because I think that will be useful as an
exercise for us, for me, as new to the committee and new to
this area compared to other members of this committee.
Unfortunately, my time has expired, so it's a question
that, for better or worse, you won't be asked to answer. But if
you could provide it in writing or in some other form, I'd be
grateful for it.
Thank you, Mr. Chairman.
The Chairman. Thank you. We'll begin just a short second
round.
Dr. Shuren, FDA acted very quickly to reject the core
conclusions of the IOM report on the 510(k) process. I'm going
to ask why, but I'm going to preface that question by saying,
we put a lot of faith in the Institute of Medicine and the
National Academy of Sciences. And I can tell you there are
times in the past when they have come out with findings that I
probably didn't like for a certain, political reason, or that
might have gone against a, perhaps a, constituent of mine. But,
we recognize the soundness of the process of the National
Academy of Sciences and the Institute of Medicine.
So, their request to do this, they said here in their
conclusion, they said that the FDA asked the Institute of
Medicine to review the 510(k) clearance process, answer two
questions. Does the current 510(k) clearance process optimally
protect patients and promote innovation in support of public
health? Answer: No. Second, if not, what legislative,
regulatory or administrative changes are recommended to
optimally achieve the goals of the 510(k) clearance process.
Their answer was probably, ought to get rid of it and come up
with a new regime. Why did you reject the core conclusion of
the IOM?
Dr. Shuren. Because we think that the 510(k) program, in
most cases, has actually worked well. If we're going to be
thinking about changing it, putting something else in place,
I'd first ask, what are we putting in place instead of it,
which is a question the Institute of Medicine felt they were
not in a place to actually say?
And the second is to just, at the outset, get rid of the
program, which would be highly disruptive, and before anyone
would think to do it is, then what are we gaining in return? Do
we have an alternative program we're actually turning to that
can ensure safety and effectiveness? Is it going to do a better
job than what we have right now? And what about a transition to
a new program? What will the impact be?
So, for us, we thought that getting rid of the program in
its entirety was going too far. That said, much of what we are
doing in this past year is about improving the programs that we
have. And we still welcome additional thoughts, if there are
other actions that we need to take. And if so, we'd be very
happy to consider them.
But, it's more focused on making the programs we have work.
And some of those products have, kind of, fallen, I'll say a
little bit on the wayside.
The Chairman. I understand that. Thank you very much.
Senator Burr.
Senator Burr. Thank you, and Dr. Shuren, I, for one,
commend the FDA on their decision on the 510(k) because I think
that it does embrace what it set out to, and it establishes a
pathway for devices that have an equivalence that's
established, and I think it was the right thing.
The industry says delays in reviews are because FDA keeps
moving the goal post, causing regulatory uncertainty and lack
of predictability in the review and approval process. In
contrast, the FDA has repeatedly said that the problems at the
device centers are the industry's fault.
Now, you highlighted earlier that there are improvements in
approval times. And I think you said that companies are
improving in the quality of their applications. Did I hear that
accurately?
Dr. Shuren. No. First off, I would say we have not been out
there saying the problems are all due to industry. I've been
saying for a long time, now, that actually the problems are
multi-factorial, and a good part of it is due to the FDA. And
that's what I tried to make clear in my opening statement, and
I've said this to Congress on the House side several times
before, and in other forums.
One piece of it is we do get from a number of companies
submissions that are of poor quality. And unlike the drug
center, which will send them back, we'll let them in. We'll
work with companies, we'll take it and we'll go many rounds to
try to get it right. But, as a result, it's making us
increasingly more inefficient. That is just one piece.
Senator Burr. You would agree that quality applications are
an incentive for a company to move through the process as
quickly as possible?
Dr. Shuren. Here's been the challenge. When we take in a
submission that may not be of sufficient quality, and we're
going to work with them, we actually end up doing some of the
work for the company. We've had companies who come in; they
don't even identify the predicate device.
Senator Burr. Let me, I'm trying to make this easy for you.
You would agree that a quality application enhances the
timeline, and there's an incentive there for the manufacturer
to have a quality app, is that an accurate statement?
Dr. Shuren. Yes.
Senator Burr. OK, great, you made a statement that you
haven't been out there bashing the industry. In fact, in a
response to the perception that FDA's to blame for the increase
in 510(k)'s, you were quoted in February as saying, ``Those
510(k) times are all on the industry.''
Dr. Shuren. I would be interested to see the quote and it's
context. If we're talking about the times, the increase in
times or the increase in the industry times, that is correct.
But the factors that led to the increase in times are due to
things on our end, and some things that are on industry's end.
So, hopefully, I will go back and see that quote in context.
Senator Burr. This was reported in the Medical Device
Daily. It was in an ENC Subcommittee hearing in February, and
your quote was, ``Those 510(k) times are all on the industry.''
Dr. Shuren. So, if it's ENC Committee happy to go back, and
we'll pull the testimony to go ahead and look at it.
Senator Burr. I've learned that press is accurate, no
matter what they say.
Dr. Shuren. You must be dealing with a different press than
I'm used to. No offense to the press who are here.
Senator Burr. Bottom line, the average total times is
getting worse instead of better. While the agency may not be
held responsible for every single submission, doesn't the fact
that the average total time getting worse clearly shows the FDA
is contributing to the problem? I think you admitted to it
earlier.
Dr. Shuren. I agree, we're part of the problem. And I've
said we've not managed the programs as well as we should have,
and that's why you're seeing a number of the actions we're
taking are improvements on the FDA programs, the things we need
to do better.
Senator Burr. I looked at the official meetings that were
published by the FDA, on the medical device user fee agreement.
Those negotiations suggest that the agency is requesting more
than a 250 percent increase in user fees from current levels.
Let me ask you, would you agree to pay somebody any money
at all, much less a 250 percent increase above what they are
currently paying, when the terms of what you're paying for
aren't being met?
Dr. Shuren. Yes, in some cases, I actually would. I know
this for research and development. Quite frankly, if you put a
little bit in research and development, you may not get enough
out of it. You just have to put enough into it, you get enough
of a return.
To sing for our program, and it's why I empathize with
industry, I actually do because they haven't seen that return
on investment, at least in numbers that they would like to see,
and to make the argument that in the absence of the funding,
things would have been worse, that is true, but a hard case to
actually show somebody.
But at the same token, those fundamental problems that we
showed in the program, some of them are due to not having
enough people, having that high reviewer turnover, not having
enough managers, and I can't solve with some missive that I
send out to my people. I can't solve that with a pep talk.
Senator Burr. Do you acknowledge that an increase in the
user fee is eventually passed on to the consumer? In other
words, an increase in the user fee raises the cost of
healthcare in this country?
Do you agree?
Dr. Shuren. Actually I don't know. And if you're saying,
then, the companies are passing it on to consumers, then on the
flip side, you're saying that the companies never actually
absorb the increase. In which case, they're not really paying
any more money. And that's not what the companies are telling
me. The companies are telling me, if they have to pay more,
it's coming out of their hides. So, I'm assuming they're not
passing it on to the consumers, in which case----
Senator Burr. I'm not going to get into negotiations that
you're having with the companies. But I think, throughout
healthcare, any place that we exercise an increase in the cost,
that finds its way to the consumer. It finds its way to
healthcare at a time that we've got to take $4 trillion out of
healthcare. I think you've got to weigh in this equation. How
much are we increasing the cost of devices? It troubles me, and
I think, I would hope that it would trouble you, that most
companies across this country tell horrific stories about going
through the approval process at the FDA.
You talked about the venture capital earlier. I won't get
into it, but the majority of the companies that come in to
file, are publicly traded companies. They've got a market
capitalization that they've got to be worried about. The
uncertainty of the process will dictate whether, in fact, they
decide to go for FDA approval. But, I will assure you at the
end of the day, if it negatively impacts their market capital,
they'll make a decision not to get involved. Not unlike the VC
community's decision as to whether they're going to get in with
a company pre-going public.
So, capitalization, whether it's on the VC side or whether
it's in market capital, is affected greatly by the decisions
that these companies make. And I think, need to be considered,
from a standpoint of how it's affected by the process that we
set up. I hope that's something that's food for thought for
you.
Mr. Chairman, I apologize, you gave me a tremendous amount
of time, and I have cheated my good friend and colleague,
Senator Hatch.
The Chairman. Does anyone have any further questions for
Dr. Shuren before I dismiss him and bring up the next panel?
Senator Hatch. Could I ask a few, Mr. Chairman?
The Chairman. Senator Hatch.
Statement of Senator Hatch
Senator Hatch. Thank you so much. Dr. Shuren, I appreciate
the work you're doing, and I appreciate the FDA, in general.
Multiple reports that examined FDA's databases, found that FDA
has an extraordinary safety record over the past decade.
Furthermore, the IOM stated that there was no evidence to
suggest patients are being exposed to unsafe or ineffective
products. Based upon these findings, why has CDRH seemingly
diverted precious financial and strategic resources away from
the premarket review process, and focused on overhauling a
program that really seemed to be working, pretty doggone well,
and for which you were complimented?
Dr. Shuren. Actually, a lot of the actions we are taking
are to improve the predictability, consistency, and
transparency of that program, and industry has complained about
that program not being sufficiently predictable or consistent
and transparent.
In fact, many of the actions we're taking are based upon
feedback that we had received as we've gone across the country
to get input.
Senator Hatch. Some of the things I'm asking may have been
asked already. But, sorry, I missed the early part of this. I
was at the Judiciary Committee.
Dr. Shuren, FDA performance data shows that the 510(k)
review times have increased 43 percent in just 4 years, and PMA
review times have increased 75 percent. When American industry
is losing its competitive edge and when patients are waiting
longer and longer for new treatments and cures, that's a matter
of great concern to me.
Now, you've said that the agency has implemented reforms
and announced initiatives to reduce these approval times, but I
have yet to hear from you any specific actions you will take
that will require a guarantee that these approval times will be
reduced. And, let me just ask this question, what metrics are
you using to measure the success of these initiatives and
reforms?
Dr. Shuren. We're looking at some right now, indicators on
performance, and I apologize. I showed one a little bit earlier
regarding our backlog on 510(k)'s, which in 2011, for the first
time was coming down. We've looked at, kind of, the percent of
510(k)'s that we're now clearing, going on the market. We've
got to be a little bit careful on that number because we should
only clear a device that, in fact, is substantially equivalent.
But that number has actually been going up.
Some of the measures we're looking at are showing early
signs we're going to be looking at our performance in terms of
the times on the review clock for products coming to market.
But, I'll tell you for success, it's going to require that we
continue to make the improvements we'd laid out. If there are
things that we missed, we still want to hear about it because I
keep going around asking people, and people saying, ``well,
that sounds like the list.''
We talked a little bit about what we need to do with
companies, making sure we do get quality submissions. And also
said that we won't be successful if we don't have the adequate
and stable resources. And I'll say right now in terms of the
fees being paid, like I said, I empathize with industry, but in
the one case of 510(k) right now, the full fee is $4,000. And
if you're a small business, which is $100 million in annual
sales or receipts, it's $2,000 for a 510(k). And yet, we say,
for some more money, if we can reduce the times, and it's not
just the times on review, if you have greater predictability,
you actually reduce the time on assessing that device, which
we're not even measuring.
I mean, you're talking about--you could be looking at
important savings. And yet, you could say, ``Well, suppose I
increase that fee $3,000? '' And yet I save days and people are
telling me in the industry, every day lost is thousands of
dollars, that, to me says, if I save 1 day, you may have gotten
your return on the investment, in that 1 day.
And those are part of the user fee discussions right now.
But, so, for success, we need to make sure we've got the
resources, we need to work on the submissions coming in. But, I
have to say, again and again, FDA has to do a better job. And
those are the actions that we're taking now.
Senator Hatch. I appreciate that, and I appreciate your
attitude about it. And we've met about it, and we've chatted
about these matters, and it's a tough job you have. But, some
find it shocking that the average time to clear a 510(k)
application has increased 50 percent since 2002. The average to
approve a PMA application has doubled since 2000, and the total
review times for both 510(k) and PMAs are now, actually, longer
than they were before the user fee program was instituted.
Now, these statistics are from FDA's own data, as you know,
yet user fees for CDRH have increased by nearly 30 percent. I
know you agree that these statistics are somewhat troubling,
they are to me, and I'm very appreciative that you're on top of
these things, and willing to do something about it. And I hope
that you can. Because I think we're falling behind the rest of
the world in approvals and yet we have some of the most
imaginative and intelligent people working in this country,
compared to the rest of the world.
So, you have a very important job, in my opinion, and I'm
looking to you to improve the Agency's performance. And I'm
counting on you.
Thank you, Mr. Chairman.
The Chairman. Senator Hagan.
Statement of Senator Hagan
Senator Hagan. Thank you, Mr. Chairman.
Dr. Shuren, I appreciate you coming before the committee,
but I also appreciate your time in September when you came down
to North Carolina to meet with the hard-working innovative
medical device companies in my State. In North Carolina, we
employ over 8,000 people in the medical technology industry,
and it's an industry that really does create thousands of jobs.
And I know that it was interesting and invaluable to hear their
perspective as well as the FDA's, so, thank you.
I know that you've made efforts already to improve the
quality of work among reviewers, and I appreciate your efforts
to increase reviewer training, thank you on that. But, one
complaint that I continue to hear from constituent companies is
that the FDA's medical device review process is unpredictable
and inconsistent, and I know you can't change this overnight.
But companies have told me that when they request a pre-
submission meeting, it may take months before they're able to
meet with the FDA, and this may be a discussion point in some
of the MDUFA negotiations.
I'm wondering if and how FDA can speed this up now? In
other words, can FDA meet with these companies sooner rather
than later, to ensure they are complying with FDA's
requirements at the beginning of this process, rather than in
the middle?
Dr. Shuren. We've been looking at what we can do to improve
the pre-submission meetings. I will tell you that the requests
for these meetings have essentially doubled in the past 5
years. So, it's really outstripping our ability to meet the
demand, if you will. And yet, we realize how helpful those
meetings are. So, you're quite right that as a part of user fee
reauthorization, we've been talking about making sure we have
the ability to hold more timely meetings, to hold more
meetings.
But, the other piece of it is to make sure those meetings
provide the maximum value. So, we are finalizing on policy, on
expectations that companies can have of these meetings. And
what we'll say in that document is, ``here are things we'd like
to see from you,'' so we can have a well-informed meeting.
The second is what you can expect from us. That if we're
giving advice, then we should write it down, and we should
stand behind it. Unless circumstances really change that you
can't do so. You bring us a device, and you say it's for
indication X, and you change later on what you're going to use
the device for. Well, maybe our advice changes.
But, in many other cases, it wouldn't and it shouldn't, and
so, we're going to make those improvements. Namely, we're going
to make the improvements we can with the resources we have,
regardless of what happens. But, hopefully, we will work things
out in user fee reauthorization that will have the ability to
hold more and more timely pre-submission meetings. And I think
that becomes a win for industry, and it becomes a win for us.
Senator Hagan. When you stated that you wanted to be sure
those meetings are valuable, well, in order for them to be
valuable, they do have to take place. I certainly do agree, and
anything we can do to speed that up. And I do think that
industry recognizes that delays costs them money. So, I think
for user fees, that they are definitely willing to pay upfront
in order to have the availability of the FDA to meet them early
on.
I had a question about, and I know we've also spoken about
this, but I wanted to reiterate that I've heard concerns with
FDA's proposal to regulate the laboratory diagnostic test as
medical devices. I'm concerned that duplicative regulation
could slow innovation, impeding improvements to patient care as
well as the job growth that has come from this innovation
around this industry. And I understand that FDA may be close to
releasing guidance to regulate the laboratory-developed tests
as medical devices. Can you tell me what the status of this
guidance is?
Dr. Shuren. Certainly. That framework is currently under
review by the administration.
Senator Hagan. What's your timeframe?
Dr. Shuren. That's not going to be my call.
Senator Hagan. Can you share with us what that is?
Dr. Shuren. I don't know. Because, it's currently under
administration clearance. So, I don't know when they'll wind up
making a decision.
I will say, we have been consistent all along in saying
that laboratory-developed tests for medical devices, they are
tests just like what's made by a traditional manufacturer. And
if we're out with a framework, our intent is not to duplicate
existing requirements on laboratories.
Senator Hagan. I think that's critical.
Dr. Shuren. Yes.
Senator Hagan. Thank you, Mr. Chairman.
Senator Franken. Just one question. I've heard from large
and small companies in my State that your guidance relating to
the modifications of the 510(k) process for devices is of
concern. Specifically, they are concerned that the new draft
guidance that you release does not provide adequate clarity on
when a new 510(k) submission is necessary for a modified
device. The companies are telling me that the Center of Devices
and Radiological Health may have greatly underestimated the
increased workload from the proposed changes that could
seriously delay the review process.
Given that the stated goal of reforming the 510(k) process
was to bring more clarity and efficiency to the processes,
these reports are concerning. How would you respond?
Dr. Shuren. We've heard the same concerns, and our intent
in putting out this guidance was not about seeing some big
increase in the number of 510(k)'s being submitted to us. We
put out the guidance because a manufacturer can make some
changes to a device, and they don't have to come back to the
FDA. Now, where that line is crossed is not always clear.
We've run into cases where manufacturers make changes. They
should've come to us, they didn't. And those circumstances,
it's very disruptive for the company, it's not good for
patients. So, the more clarity we can provide, the better. And
we're seeing new emerging technologies, and we wanted to
provide clarity in that space.
We've heard from companies, though, the policy we put out
would wind up leading to many more 510(k)'s being submitted. So
what we asked industry to do is, and we've got a number of
groups working on this, is to go back with individual companies
and apply. You know, say, ``here's how I interpret it, here's
how we view the impact. Give that back to us so we can go
through it, because our intent is to get this policy right.''
It's one of the reasons, too, why we've extended the time for
getting feedback on this guidance, because we want to make sure
we work with industry, we get the right policy in place.
Senator Franken. Thank you.
The Chairman. Dr. Shuren, thank you very much. I'll just
close this by saying, the committee--again, the IOM report, in
which I place a lot of stock, I think we all have to because it
is unbiased, and they don't have to answer to anyone except
their own, their own guidance as professionals. They said here,
``The FDA has procedures for developing, adopting and
implementing guidance and standards. The agency, however, is
persistently hindered in fully developing these materials by a
lack of, or limitations on human, fiscal and technologic
resources and capabilities,'' which is getting to what Senator
Mikulski was saying.
More and more is required of FDA in this area as in other
areas, and yet we continually cut the budget, cut the resources
from the Federal Government to the FDA. At the same time, the
companies that come in, as you said, we've had this huge
increase in the applications for 510(k)'s in the last few
years, and yet, as we cut the things down, more and more
pressures are put on you since you can't do your job, well,
then, we must make it easier. Because we want innovation, we
want companies to succeed for all those business reasons. I
don't think that's adequate. I don't think that suffices.
I think that we need to understand that the first
obligation you have is for the safety of patients, not whether
or not a company gets the approval in a timely manner, or
whether it makes a profit or not--your first is safety, and
effectiveness, safety and effectiveness. We'll leave it up to
CMS to talk about whether it's cost effective or not, that's
not in your realm, that's in a different realm.
And, so, it seems to me that you see the difference here in
drugs, because we're also kind of playing the reauthorizing of
the Prescription Drug User Fee Act also.
Pharmaceutical companies now pay $1.8 million for a drug
application. Devices pay, what did you say, $4,000?
Dr. Shuren. Four thousand for a 510(k), the equivalent of
the new drug application. It's a premarket approval
application. That is $220,000, so about \1/9\th of what you
pay.
The Chairman. About \1/9\th of what that would be.
Dr. Shuren. Right.
The Chairman. So, it seems to me, I know that you're
undergoing some conversations now, with the industry, on this,
on user fees. But, I would just say this as the Chairman of
this committee, that if we want better results from the FDA,
more timely results and more transparency, and all the things
that the IOM suggested we do, and we want to continue the
510(k) process, which they had said we shouldn't, we should
come up with a new regime, and perhaps we'll be looking at
that--then it seems to me that the industry is going to have to
come up with just a little bit more resources to help the FDA
do its job.
Thank you very much, Dr. Shuren. We'll move to our second
panel.
Dr. Shuren. Thank you.
The Chairman. And we have on our second panel----
Senator Franken. Would you like me to----
The Chairman. Yes. I have the first one will be introduced
first by Senator Franken from Minnesota.
Senator Franken. As the witnesses take their places, I'm
very pleased to introduce one of them. And that would be
Professor Ralph Hall from my home State of Minnesota.
Professor Hall is a member of the faculty at the University
of Minnesota Law School, where he teaches courses on food and
drug law, as well as corporate compliance and negotiations. As
an expert in medical device regulation, Mr. Hall is also the
CEO of a small medical device company in Minnesota. And he also
serves as counsel to the law firm of Baker and Daniels. With
several decades of experience working with the FDA, Professor
Hall has an understanding of the strengths of FDA's regulatory
processes, as well as a clear view of the areas where it could
be improved.
And thank you very much for joining us, Professor Hall.
The Chairman. Thank you, Senator Franken. Our next witness
up is Dr. David Challoner who is vice president emeritus for
Health Affairs at the University of Florida. His clinical
specialty is in internal medicine with a sub-special interest
in endocrinology.
He has previously served as Dean and Professor of Medicine
at St. Luke's University School of Medicine. And from 1988 to
1990, following an appointment by President Reagan, he was
chair of the President's Committee on the National Medal of
Science.
Significantly, for today's discussion, Dr. Challoner
chaired the IOM Committee that evaluated FDA's 510(k) review
process and issued the report that I mentioned earlier, this
past July.
Our third panelist, Dr. Gregory F. Curfman, who is
executive editor of the New England Journal of Medicine and
assistant professor of medicine at Harvard Medical School. Dr.
Curfman also practices cardiology and internal medicine at
Massachusetts General Hospital. He graduated from Harvard
University Medical School in 1972. He's published numerous
articles on FDA issues; particularly in the area of medical
device regulation.
We thank you all for being here today. All of your
statements will be made a part of the record in their entirety.
I'm going to ask you to sum it up in about 5 minutes or so, and
we'll go from left to right, just as I introduced everyone.
So, Dr. Hall, Professor Hall, I guess I should say. Oh, you
have a J.D., that's a Doctor too, what the heck?
Mr. Hall. That's what they tell me.
The Chairman. You're first up. Please proceed.
STATEMENT OF RALPH F. HALL, B.A., J.D., PROFESSOR OF PRACTICE,
UNIVERSITY OF MINNESOTA LAW SCHOOL, MINNEAPOLIS, MN
Mr. Hall. Thank you, Senator Franken, for the kind
invitation. Chairman Harkin, members of the committee, I
appreciate the option to come and discuss with you the medical
device regulatory system. My disclosures are in the written
materials. I need to be clear that I'm speaking on my behalf,
and I don't speak for any other organization or individual, so
these are my personal views.
My comments are going to focus on how the medical device
system currently addresses safety and effectiveness. I want to
start with three broad points.
The system has changed substantially over the roughly 35
years of its existence, as Congress has continuously worked to
improve the system, as has FDA. That means that we need to look
at the system as it exists today, not how it existed, perhaps
20 or 30 years ago.
I think it's also important to recognize that drugs and
devices present very different regulatory challenges, and what
is appropriate for one realm is not necessarily the right
answer for others. Drugs, universally, have a potential
systemic effect. Most devices do not. Many devices are tools,
not the therapy itself, and tools present, then again,
different regulatory needs and regulatory challenges. Devices
operate and are improved in an iterative process with very
short product lifecycles.
Clinical trial issues between drugs and devices are also
very different. In the drug world, the gold standard are the
double-blind placebo-controlled studies. In many cases, in the
device world, those studies simply are not appropriate. How
does one do a double-blind placebo controlled study on a heart
valve? You can't really cut somebody open, pretend to put in a
heart valve, sew them up and see if it works?
So, we need different mechanisms here. In the device world,
there's a greater importance, generally speaking, of
engineering, design, material sciences--those types of hard
sciences.
And, finally, as everyone has commented, there is a need
here to balance safety, preventing risk with the benefits of
access, particularly to innovative new products.
Devices present a broad range of risk, from implantable
neuro-modulators to crutches, and there, we have three systems
to address those. But it's critical to recognize that each of
the classifications, are to provide a reasonable assurance of
safety and effectiveness. We differ in how that's achieved, not
in the fundamental objective.
Class I, simple devices--those are general controls,
quality systems; class II, the 510(k), I'll get into this in a
bit more detail use a variety of systems, including general
controls, special controls, and the 510(k) system; class III
products used a PMA. Within the 510(k) system, we actually have
three major pathways or approaches to provide the reasonable
assurance.
First of all, the 510(k) premarket system is not simply a
physical comparison of one device with another. It's much
broader than that. We start with a classification process by
which, before the first one is reviewed, FDA in conjunction
with experts, makes a risk-based safety and efficacy-based
decision as to the appropriate classification. Which class best
provides a reasonable assurance? That's subject to
reclassification, depending upon new information.
Then, there are general controls, I mentioned, and special
controls. Special controls are product-type specific. These can
include clinical data, registries, performance standards, data
testing requirements, design requirements, materials
requirements all designed to ensure that there's this
reasonable assurance of safe and effectiveness.
There's then the process of comparing the product, the new
product, with the predicate, the old product--and that looks
at, from other things, the track record of the product. Safety
effectiveness data is available to the agency on those
products.
If there's any new intended use, or a change in technology,
and technology is defined very broadly, new safety and efficacy
information has to be provided.
Second, there are quality systems, these include design
controls, postmarket surveillance, looking at product trends,
iterative designs, etc.
And finally, there are other regulatory systems to ensure
the products are safe. For example, here a product is
misbranded, if you can't label it for safe and effective use.
Congress did not create, and the agency is not implementing
a system by which the agency has no choice but to approve an
unsafe product. Data from multiple source indicate that most
devices, the vast majority, are safe, and the area of greatest
improvement is in the quality system--that's work I've done,
Dr. Maisel, IOM's conclusions, and also a recent FDA report.
In conclusion, the FDA uses the 510(k) system to provide a
reasonable assurance of safety and effectiveness. There are
multiple tools that exist within this system to ensure that
those standards are reached.
I'll be pleased to answer questions at the appropriate
time. Thank you.
[The prepared statement of Mr. Hall follows:]
Prepared Statement of Ralph F. Hall, B.A., J.D.
summary
A Critical Balance: FDA and the Reform of the Medical Device Approval
Process
There are several key points that one should keep in mind in
assessing the current medical device regulatory system.
First, drugs and devices present very different regulatory issues.
What works well in one system may not be appropriate or effective in
the other arena.
Second, the system created by Congress and implemented by FDA is
intended to provide a reasonable assurance of safety and effectiveness
for all products, regardless of classification. The means or method by
which this is accomplished varies between Class I, II and III devices
but the objective of safety and effectiveness does not. Class I uses
``general controls'' to provide this assurance, Class II uses special
controls, general controls and the 510(k) system to provide this
assurance of safety and effectiveness. Class III uses the PMA process.
These are different means to achieve the same safety objective.
Next, the current 510(k) system gives FDA substantial authority to
clear only products with a reasonable assurance of safety and
effectiveness. FDA has multiple tools within the Class II/510(k) system
beginning with initial product classification and extending through
special controls and data submission requirements to assess product
safety and effectiveness. FDA is not forced to clear a product just
because it is physically identical to an older ``predicate'' product.
Finally, and perhaps most importantly, available data demonstrates
that the system is working well from a safety perspective. Overall,
products approved or cleared by FDA, including 510(k) products, have
very good safety records. Of course, all stakeholders should always be
striving to improve on this already good record. Improvements in QSR
(quality systems) offer the greatest potential patient benefit.
______
Good afternoon, my name is Ralph F. Hall. I appreciate this
opportunity to speak to this committee on these important medical
device matters affecting patients, physicians, innovation and jobs. I
am here to provide an overview of the medical device regulatory system,
with particular focus on how the medical device regulatory system
assesses product safety and effectiveness. In addition, I will discuss
research I and others have done into the safety of 510(k) products.
I want to be clear that I am here speaking in my personal capacity
and not on behalf of the University of Minnesota or any other entity.
background and disclosures
To start, I serve as Professor of Practitioner at the University of
Minnesota Law School where I concentrate my teaching, research and
writing in the area of FDA law and compliance matters. In addition, I
am part time Counsel at the law firm of Baker & Daniels where I work
with clients on a variety of FDA matters and also provide counsel to a
national 510(k) coalition. Finally, I serve as CEO at MR3 Medical LLC--
a four person startup medical device company working on a new
technology for cardiac rhythm devices generally regulated under the PMA
process.
I. Medical Device Regulatory Overview
a. Medical Devices are Significantly Different Than Drugs
Many commentators simply compare drug regulation and device
regulation. When differences between these systems appear, as they do,
these commentators assume that there is some problem. It is absolutely
critical to understand that there are important differences between
drugs and devices that mandate some different regulatory approaches.
These differences include:
Drugs have a systematic effect on the body. A
cardiovascular drug, for example, will also circulate throughout the
body and potentially impact the liver, kidneys, muscles, lung, brain,
etc. The vast majority of all devices do not have any systemic effect.
Thus testing issues and needs are fundamentally different between drugs
and devices.
Medical device development is an iterative process with
substantially shorter life cycles. That is not the case with drugs.
Drug life cycles cover several decades while device life cycles are
often measured in months. Also, drugs do not have the iterative
development process found in devices. Any molecular change in a drug
creates a new molecule and a whole new set of issues and questions.
Most iterative changes to a device (making a catheter longer, for
example) do not create new therapeutic issues.
Essentially all drugs are the actual therapy. Many devices
are actually a tool by which a physician delivers therapy, not the
therapy itself. For example, a scalpel is a tool by which a medical
intervention is performed. In such cases, FDA primary focus should be
on whether the tool is performing as required, not whether the therapy
such as an appendectomy (a physician decision) is effective.
Engineering, design controls, human factors and material
sciences are much more important to devices than to drugs. As detailed
below, most postmarket safety issues with medical devices involve
engineering, design, materials and manufacturing issues, problems not
discoverable through clinical risks. Available data indicates that this
is a different pattern from drugs. This difference in risk should
impact premarket requirements.
Devices span a much greater risk profile than drugs. While
essentially all drugs pose some systemic questions that is not the case
with devices. There is a world of difference between the risk/benefit
of an implantable neuromodulator and that of a crutch. This huge risk
spectrum mandates phased regulation of medical devices.
There is incredible product differentiation within medical
devices. Medical devices include diagnostic tools that never touch a
patient, multimillion dollar pieces of capital equipment such as CT
scanners, simple tools like a scalpel or bandaid and complex
implantable devices such as an ICD.
Device regulation includes robust and broad quality system
requirements (often referred to as QSR requirements).
The overall impact of these differences is that device regulation
needs different premarket requirements, a risk-based approach to
regulation and an emphasis on quality systems.
b. Device Regulatory Overview
By statute and regulation, all medical devices, regardless of risk
classification, are to have a ``reasonable assurance of . . . safety
and effectiveness'' before they are marketed.\1\ What differs is the
method by which FDA and other stakeholders assess whether there is such
assurance of safety and effectiveness for different classes of
device.\2\ These different ways to provide this assurance of safety and
effectiveness and the complex language and statutory systems for
medical device regulation can lead to inadvertent confusion and
misunderstanding. However, all products of whatever risk classification
must provide this reasonable assurance of safety and effectiveness.\3\
---------------------------------------------------------------------------
\1\ 21 U.S.C. 393(b)(2)(C).
\2\ It goes without saying that all therapeutic products have some
risks. The objective is to ensure a positive risk/benefit for each
product.
\3\ Ensuring patient access to beneficial products is also
critical. As such, FDA is also charged with promoting product
innovation. While the focus of my comments is on the safety aspects of
the device regulatory system, one cannot forget the importance of
making innovative products available to physicians and patients.
---------------------------------------------------------------------------
Because medical devices differ so much--from a tongue depressor to
a multimillion dollar robotic surgical system--one regulatory approach
does not fit all. To address this, Congress created a three-tier
regulatory structure.
Class I devices are the simplest, lowest risk devices.
These include crutches, tongue depressors and scalpels. These products
usually do not go through a premarket review and are generally referred
to as Class I exempt.
Class II devices are medium risk products such as
angioplasty catheters. Class II devices generally go through the 510(k)
system.
Class III devices are the highest risk devices and include
heart values and pacemakers. These products reach market through the
PMA process.\4\
---------------------------------------------------------------------------
\4\ This is a general description. Some higher risk Class I devices
must go through the 510(k) system and some higher risk Class II
products require a PMA. There are also some other pathways to market
including the HDE process and the rarely used PDP system. For our
purposes these alternative pathways are not relevant.
Each device class, with some overlaps, uses a different method to
provide assurances of safety and effectiveness.
Class I products are those for which ``general controls'' are
``sufficient to provide reasonable assurance of the safety and
effectiveness of the device''.\5\ General controls can include, as
appropriate, manufacturing controls, labeling, quality systems, etc.
---------------------------------------------------------------------------
\5\ 21 U.S.C. 360c(a)(1)(A)(i).
---------------------------------------------------------------------------
Class II products are those for which general controls by
themselves are not sufficient but for which ``special controls'' do
provide a reasonable assurance of safety and effectiveness. These
products use a different, multipronged system to provide the reasonable
assurance of safety and effectiveness. Specifically, Congress provided
that a Class II device is:
A device which cannot be classified as a Class I device
because the general controls by themselves are insufficient to
provide reasonable assurance of the safety and effectiveness of
the device, and for which there is sufficient information to
establish special controls to provide such assurance [of safety
and effectiveness], including the promulgation of performance
standards, postmarket surveillance, patient registries,
development and dissemination of guidelines (including
guidelines for the submission of clinical data in premarket
notification submissions in accordance with section 510(k)),
recommendations, and other appropriate actions as the Secretary
deems necessary to provide such assurance.\6\
---------------------------------------------------------------------------
\6\ 21 U.S.C. 360c(a)(1)(B).
As mandated by Congress, Class II devices generally must receive
clearance under the 510(k) system described in more detail below and as
part of that process must satisfy both special controls and general
controls. 510(k) submissions can include clinical data, bench testing,
labeling, reports on prior investigations, etc.
The 510(k) system (described in more detail below) generally
requires a product to establish that it is ``substantially equivalent''
to a predicate 510(k) device. Substantial equivalence is more than a
physical comparison of one device to another. 510(k) products must also
meet all special controls, all applicable standards and QSR
requirements. FDA has the authority under the 510(k) system to request
a wide variety of data, including clinical data, bench testing,
proposed labeling, and material information, as part of its review of a
510(k) submission.\7\ This submission explicitly includes a variety of
safety and effectiveness information.\8\
---------------------------------------------------------------------------
\7\ 21 CFR 807.87, .90, .92 and .93 set forth more details about
the content and format of a 510(k) submission.
\8\ See, for example, 21 CFR 807.92(c)(3).
---------------------------------------------------------------------------
Class III products must go through the PMA process.\9\ This often
includes clinical testing and submissions include detailed
manufacturing information, labeling, bench test data, etc. FDA reviews
this data for safety and effectiveness.
---------------------------------------------------------------------------
\9\ 21 U.S.C. 360e.
---------------------------------------------------------------------------
It is important to understand that there are a number of other
systems that also impose safety and effectiveness controls on products
as part of an integrated system to provide a reasonable assurance of
safety and effectiveness. For example, the QSR system \10\ requires
design controls to help ensure a safe and effective design. There are
also product and adverse event trending requirements, reporting
requirements, etc. In addition, FDA has the authority to require
postmarket testing on higher risk devices (including specifically Class
II/510(k) products).\11\ There are also general labeling requirements
including 21 U.S.C. 352(f) which mandates that a product labeling
include ``adequate directions'' for safe use.
---------------------------------------------------------------------------
\10\ QSR requirements are generally found in 21 CFR 820.
\11\ 21 U.S.C. 360l. 21 U.S.C. 360l(a)(1)(A) explicitly
includes Class II devices within the group of products subject to so-
called ``522 orders.''
---------------------------------------------------------------------------
II. The 510(k) System Includes Safety and Effectiveness Considerations
The 510(k) system has been the focus of recent attention. The
510(k) system does consider safety and effectiveness. Stated
differently, current FDA authority gives the agency multiple pathways
to keep an unsafe 510(k) product off the market, require whatever
testing or data is needed to establish safety and effectiveness and
remove unsafe products from the market.
From the beginning, Congress intended for the 510(k) system (and
the substantial equivalence part of that process) to include safety and
effectiveness determinations. As FDA itself explained to the IOM
committee in March 2010, Congress intended the 510(k) substantial
equivalence standard to be flexible in order to assure safety and
effectiveness. The 510(k) legislative history states:
The [congressional] committee believes that the term
[substantial equivalence] should be construed narrowly where
necessary to assure the safety and effectiveness of a device
but not narrowly where differences between a new device and a
marketed device do not relate to safety and effectiveness.\12\
---------------------------------------------------------------------------
\12\ See FDA's presentation to IOM available at http://www.iom.edu/
Activities/PublicHealth/510KProcess/2010-MAR-01.aspx.
Note the specific linkage of the substantial equivalence
determination to safety and effectiveness.
In order to see how the Class II/510(k) system ensures an
assessment of safety and effectiveness, one must understand the process
from the start. The 510(k) process actually begins before the first
product is reviewed. By statute, FDA is obligated to classify each
product type into Class I, II, or III. This classification process
includes expert advisory panels, assessment of data and an opportunity
for stakeholder input.\13\ The purpose of the classification process is
to determine which oversight system is best positioned to provide
assurances of safety and effectiveness. The product classification is
based on safety and effectiveness considerations as confirmed by the
implementing regulation which states:
---------------------------------------------------------------------------
\13\ 21 U.S.C. 360c(c) and (d).
(b) In determining the safety and effectiveness of a device for
purposes of classification, establishment of performance
standards for Class II devices, and premarket approval of Class
III devices, the Commissioner and the classification panels
---------------------------------------------------------------------------
will consider the following, among other relevant factors:
(1) The persons for whose use the device is represented
or intended;
(2) The conditions of use for the device, including
conditions of use prescribed, recommended, or suggested
in the labeling or advertising of the device, and other
intended conditions of use;
(3) The probable benefit to health from the use of the
device weighed against any probable injury or illness
from such use; and
(4) The reliability of the device (emphasis added).\14\
---------------------------------------------------------------------------
\14\ 21 CFR 860.7(b).
Further, by statute, if a product is implantable or is used to
support or sustain human life, the default classification is Class III/
PMA unless the agency and classification panel specifically determine
that the Class II/510(k) process is sufficiently robust and that Class
III/PMA systems are not necessary to provide reasonable assurance of
safety.\15\ Thus, before any device is even eligible for 510(k) review,
FDA, in concert with expert classification panels, has made a
determination that the Class II/510(k) system provides an adequate
assurance of safety and effectiveness for that product type.\16\
Therefore, every 510(k) product type has been assessed and it has been
determined that the 510(k) system provides the adequate assurance of
safety and effectiveness.
---------------------------------------------------------------------------
\15\ 21 U.S.C. 360c(c)(2)(C).
\16\ As GAO and a number of commentators have noted, FDA is
delinquent in classifying 26 out of, I believe, approximately 1,800
product types. These products continue to be reviewed under the 510(k)
system. FDA is currently in process of rectifying this situation and
completing the classification of these remaining products.
---------------------------------------------------------------------------
Once classified, the 510(k) system uses the concept of
``substantial equivalence'' as a method to assess safety and
effectiveness.\17\ The policy behind the 510(k) system is that once it
has been determined that a product type is safe and effective for its
intended use, future products that are ``substantially equivalent'' to
the initial product and which meet all other regulatory requirements
are likewise safe and effective. Substantial equivalence is more than a
physical comparison of one product to another.
---------------------------------------------------------------------------
\17\ In reviewing 21 U.S.C. 360c--the key statutory section
relating to Class II/510(k) devices--one can see that Congress used the
term ``safety'' with regard to Class II/510(k) products more than 17
times by my count. One can only wonder why Congress would discuss
safety so many times unless Congress intended for the 510(k) to
consider safety and effectiveness.
---------------------------------------------------------------------------
The 510(k) submission provides the information to FDA by which it
can determine that the safety profile of the new product meets the
established safety profile of the prior (or predicate) device, all
special controls or similar requirements have been met and that the
product is otherwise safe and effective for its intended use. The
submission specifically includes safety information.
For example, 21 CFR 807.92(c)(3) states that a 510(k) summary
must include:
The conclusions drawn from the nonclinical and clinical tests
that demonstrate that the device is as safe, as effective, and
performs as well as or better than [the predicate device].
21 U.S.C. 360c(i)(3)(A) requires a 510(k) submission to include
adequate information respecting the safety and effectiveness of the
device and/or to make that information available. Under 21 U.S.C.
360c(i)(3)(B), this summary shall include detailed information
regarding adverse health effects relating to the product and this
information shall be made available to the public.\18\
---------------------------------------------------------------------------
\18\ One is hard pressed to argue that Congress intended FDA to
have this safety and effectiveness information and then mandated that
FDA ignore that data in making 510(k) clearance decisions.
---------------------------------------------------------------------------
In addition to data submission requirements, Congress has given FDA
another powerful tool to ensure product safety and effectiveness. The
510(k) system makes explicit use of ``special controls'' to ensure
safety and effectiveness and any 510(k) product must satisfy all
special control requirements. Under 21 U.S.C. 360c(a)(1)(B), special
controls are explicitly used to provide a reasonable assurance of
safety and effectiveness. Special controls can include clinical data
requirements, performance standards, patient registries, guidance
documents, etc. Any new product must comply with all applicable special
controls. These special controls are used in addition to physical
identicality to establish safety and effectiveness.
Products can and usually do evolve over time. The ``substantial
equivalence'' process is designed to subject any new product use or
technology to an explicit safety and effectiveness review. Congress
specifically required FDA to assess new intended uses and new
technologies for safety and effectiveness.\19\
---------------------------------------------------------------------------
\19\ 21 U.S.C. 360c(i)(1). Note that new technology is broadly
defined to ensure that product changes are reviewed for safety and
effectiveness. 21 U.S.C. 360c(i)(1)(B).
---------------------------------------------------------------------------
There is, of course, the concern that changing information or new
data may call into question prior classification decisions or special
controls. Congress anticipated this concern and explicitly established
reclassification processes under 21 U.S.C. 360c(e) that FDA can use
(and any stakeholder can request) in the event of new information. This
reclassification process can address any new information and either up
classify or down classify a device type as the data directs. Any down
classification from Class III to Class II requires a determination that
Class II special controls provide a reasonable assurance of safety and
effectiveness. Likewise, FDA can create new or enhanced special
controls under 21 U.S.C. 360c(a)(1(B) to address new safety or
effectiveness issues.
Congress has provided FDA with other tools to ensure that an unsafe
Class II product does not reach the market. For example, FDA has the
authority to ban unsafe devices,\20\ and ensure that the product
labeling permits safe use.\21\ Any product that has been removed from
the market ``at the imitative'' of FDA or has been found to be
misbranded or adulterated by a court cannot be used as a predicate to a
later product.\22\ This is one method by which a ``bad'' predicate
cannot be used for future products. Other tools include the ability to
develop new or enhanced special controls and to require additional data
to be submitted.
---------------------------------------------------------------------------
\20\ 21 U.S.C. 360f(a).
\21\ 21 U.S.C. 352(f). Even if a product is ``substantially
equivalent'' to another, if it cannot be labeled so that it can be used
safely, the product is misbranded and distribution of such a product
triggers civil and criminal liability.
\22\ 21 U.S.C. 360c(i)(2).
---------------------------------------------------------------------------
While the term ``substantial equivalence'' can sound like merely a
physical comparison of one product to another, an understanding of the
overall 510(k) system demonstrates that much more than physical
identity is needed to be cleared for marketing. Before a product can be
deemed to be ``substantially equivalent'' and the product legally
marketed the system requires, among other requirements:
Product classification into the 510(k) system based on
safety and effectiveness assessments.
Compliance with special controls explicitly intended to
provide assurances of safety and effectiveness.
Compliance with all applicable standards and guidances.
Assessment of any new intended uses or new technology for
safety and effectiveness.
Submission of safety and effectiveness data and adverse
health information.
Compliance with all applicable general controls.
Compliance with QSR requirements.
As such, FDA has multiple avenues to assess and address any safety
or effectiveness issues.
III. Key Examples
I will now apply the 510(k) system to the three key product
situations to demonstrate that, in each case, FDA has the authority to
assess safety and effectiveness.
a. The New Product
There are situations in which a product is developed for which
there is no predicate. Normally, these products are automatically, by
application of statute, classified as PMA products.\23\ Unless there is
an actual reclassification, these products go through the PMA process
and so there is no question about the robustness of the 510(k) process.
---------------------------------------------------------------------------
\23\ 21 U.S.C. 360c(f)(1).
---------------------------------------------------------------------------
However, such a product may well be a medium risk product and so
best regulated as a Class II/510(k) product. In these cases, the
product can be classified as a Class II/510(k) product pursuant to the
``de novo'' process under 21 U.S.C. 360c(f)(2). This classification
process explicitly considers whether the product can be safely
regulated under Class II systems including special controls.
As such, no ``new'' product can be regulated under the 510(k)
system unless FDA has made an explicit determination that the 510(k)
system provides adequate assurances of safety and effectiveness.
b. Changes to an ``Old'' Product
The next fact pattern involves an existing product, already in the
Class II/510(k) system, to which the company is making some change.
This can be a new intended use or some change in technology. In each
case, the change in the product must be explicitly assessed for safety
and effectiveness.\24\ The product cannot be cleared if the product
raises some new issue of safety or effectiveness.\25\
---------------------------------------------------------------------------
\24\ 21 U.S.C. 360c(i) and 21 CFR 807.
\25\ FDA's internal processes and flow charts reinforce the fact
that any change in intended use or technology is assessed for safety
and effectiveness. There is a ``not substantially equivalent''
(``NSE'') determination if there is some new question of safety or
effectiveness.
---------------------------------------------------------------------------
Remember that one of the core concepts of the 510(k) system is that
once safety and effectiveness has been determined, like products can
establish safety and effectiveness based on the prior assessment. Of
course product changes can challenge this concept and so Congress as
decreed and FDA has insisted that any change in the use of the product
or the technology (broadly defined) must be assessed to ensure safety
and effectiveness. Thus, Congress and FDA have assured that product
iterations or changes will be assessed for safety and effectiveness.
c. Continued Marketing of an ``Old'' Product
The final challenge is the one that seemed to bother the IOM
committee and others the most and this is the old product that hasn't
changed.\26\ Some seem to believe that these ``old'' products have
never been assessed for safety and effectiveness and that FDA is bound
to clear any such product without considerations of any safety or
effectiveness issues. This is simply not the case.
---------------------------------------------------------------------------
\26\ This includes situations in which the old, unchanged, feature
of the product presents some new safety or effectiveness issue.
---------------------------------------------------------------------------
FDA has multiple authorities to keep an unsafe 510(k) product--even
if literally identical to an old product--off the market.
To start, all products have been assessed for safety and
effectiveness issues through the classification process.\27\ Even if
the product existed before 1976, it has been specifically assessed and
a determination made that products of that type can be regulated under
the Class II/510(k) system for safety and effectiveness.\28\ Just
because a product was on the market before 1976 does not mean that it
is part of the 510(k) system.
---------------------------------------------------------------------------
\27\ I recognize that a few products (some number less than 26 out
of approximately 1,800 product codes) have not completed this process.
As many others have previously said, this process must be completed.
FDA is currently in the process of doing so.
\28\ See 21 U.S.C. 360c(c) and (d) and 21 CFR 807 and 860 for
more details.
---------------------------------------------------------------------------
The related question is what happens if new information is
developed on an ``old'' product subsequent to its classification.
First, FDA has access to such information through any number of
sources. Importantly, Congress has decreed that the company must
include adverse health information in its 510(k) submission.\29\
---------------------------------------------------------------------------
\29\ 21 U.S.C. 360c(i)(3).
---------------------------------------------------------------------------
Once such information comes to FDA's attention, FDA has any number
of approaches to prevent an unsafe product from being cleared via the
510(k) system. Examples of these tools include:
Creating new or enhanced special controls to mitigate or
eliminate the newly discovered risk.\30\
---------------------------------------------------------------------------
\30\ 21 U.S.C. 360c(a)(1)(B).
---------------------------------------------------------------------------
Reclassifying the device into Class III.\31\
---------------------------------------------------------------------------
\31\ 21 U.S.C. 360c(e).
---------------------------------------------------------------------------
Creating or adopting new guidances or standards.
Requiring new labeling to mitigate or eliminate the issue
(or concluding that such improved labeling would not be effective and
thus the product is misbranded).\32\
---------------------------------------------------------------------------
\32\ 21 U.S.C. 352(f).
---------------------------------------------------------------------------
Imposing postmarket obligations.\33\
---------------------------------------------------------------------------
\33\ 21 U.S.C. 360l.
---------------------------------------------------------------------------
Banning the device.\34\
---------------------------------------------------------------------------
\34\ 21 U.S.C. 360f.
---------------------------------------------------------------------------
Utilizing QSR requirements to address the issue.\35\
---------------------------------------------------------------------------
\35\ 21 CFR 820.
Thus, each product type is reviewed for safety and effectiveness
issues at the time of initial classification. Post classification, FDA
has multiple statutory and regulatory authorities available to prevent
an unsafe product from being cleared.
Congress did not create--and FDA is not implementing--a regulatory
system under which FDA has no choice but to clear an unsafe device.
IV. FDA in Fact Makes Safety and Effectiveness Determinations in
Product Clearances
How the 510(k) system actually works is best demonstrated by
looking at actual product clearances. In many cases, FDA specifically
indicates in the clearance documents that the product in question is
safe and effective for its intended uses.
For example, Via Biomedical, Inc.'s Stent Graft Balloon Catheter
has been determined substantially equivalent and cleared for market
distribution.\36\ Included in the 510(k) summary was the following:
---------------------------------------------------------------------------
\36\ 510K Summary from Via Biomedical, Inc., on the Stent Graft
Balloon Catheter (May 29, 2009), http://www.accessdata.fda.gov/
cdrh_docs/pdf9/K091624.pdf.
The Stent Graft Balloon Catheter underwent mechanical,
performance, and Biocompatibility testing to verify that the
device functions in a safe and effective manner. The results of
the tests provide reasonable assurance that the device has been
designed and tested to assure conformance to the requirements
for its indications for use.\37\ (emphasis added).
---------------------------------------------------------------------------
\37\ Id. (emphasis added).
Becton, Dickinson and Company's (Becton) BD Flu+ Syringe was
cleared for market on July 2, 2009. As part of its submission, Becton
expressly indicated that [d]esign [v]erification tests were performed
based on the risk analysis performed, and the results of these tests
demonstrate that the BD Flu+ Syringe performed in an equivalent manner
to the predicate device and is safe and effective when used as
intended. \38\
---------------------------------------------------------------------------
\38\ 510K Summary of Safety and Effectiveness from Becton,
Dickinson and Company on the BD Flu+ Syringe (Jul. 2, 2009), http://
www.accessdata.fda.gov/cdrh_docs/pdf9/K091377.pdf.
---------------------------------------------------------------------------
Likewise ArthoCare's Bone Cement Opacifier was cleared under 510(k)
after the FDA confirmed that the performance testing and device
comparison demonstrated that the subject device [was] substantially
equivalent to the predicate device, and is safe and effective for its
intended use. \39\ (emphasis added).
---------------------------------------------------------------------------
\39\ 510K Summary from Becton, Dickinson and Company on BD Flu+
Syringe, http://www.accessdata.fda.gov/cdrh_docs/pdf4/K042947.pdf. The
device was not found to be as safe as the predicate, but there was an
independent assessment. The device was both substantially equivalent to
the predicate as well as safe and effective.
---------------------------------------------------------------------------
There are numerous other examples of 510(k) submissions that have
been included in the safety and effectiveness data and have been
assessed by FDA for safety and effectiveness. A few examples include
the Master Healthcare's Easy Touch Insulin Syringe,\40\ ZOLL
Circulation's Central Venous Catheter and Thermal Regulating System
\41\ and Medtronic's Cardiopulmonary Centrifugal Blood Pump.\42\ All of
these submissions included performance data specifically relating to
the safety and effectiveness of the device as part of the 510(k)
clearance.
---------------------------------------------------------------------------
\40\ 510K Summary from Masters Healthcare on the Easy Touch Insulin
Syringe (May 14, 2009) http://www.accessdata.fda.gov/cdrh_docs/pdf9/
K091474.pdf.
\41\ 510K Summary from ZOLL Circulation for Venous Catheter and
Thermal Regulating System (Oct. 12, 2010), http://
www.accessdata.fda.gov/cdrh_docs/pdf10/K101987.pdf.
\42\ Summary of Safety and Effectiveness from Medtronic for the
Cardiopulmonary Centrifugal Blood Pump (Jun. 21, 2010), http://
www.accessdata.fda.gov/cdrh_docs/pdf10/K100631.pdf.
---------------------------------------------------------------------------
As these and other examples demonstrate, FDA in fact considers
safety and effectiveness in product decisions.
Furthermore, in numerous presentations, guidance documents and
public statements, FDA has said that the 510(k) system includes safety
and effectiveness protections.
In summary, it can be seen that products going through the 510(k)
system are assessed for safety and effectiveness beginning with the
initial classification process. FDA has a variety of tools including
special controls to ensure product safety. Congress did not create a
system by which literally thousands of devices have been cleared
without protecting patient safety.
V. Medical Device Safety Study Summary
The actual safety of medical devices is, of course, of prime
importance to patients, physicians and other stakeholders.
There have been several studies of medical device safety (or
reasons for medical device problems) over the past 2 years. These
include a study I have done (and presented to the IOM 510(k)
committee), a study by Dr. William Maisel (also presented to the IOM
510(k) committee) and a recent report by FDA itself.
In my view, these studies, individually and together, support two
key conclusions:
(1) There is no evidence of any overall systemic issue with the
safety of 510(k) products; and
(2) The primary cause of medical device safety recalls are quality
system issues, not a lack of premarket clinical testing. I will also
note that the IOM 510(k) committee itself also found no evidence of a
systemic issue with the safety of 510(k) products. The committee has
explicitly stated: ``The committee is not suggesting that all, many, or
even any medical devices cleared through the 510(k) process and
currently on the market are unsafe or ineffective.'' \43\
---------------------------------------------------------------------------
\43\ See IOM report brief available at http://www.iom.edu/Reports/
2011/Medical-Devices-and-the-Publics-Health-The-FDA-510k-Clearance-
Process-at-35-Years.aspx.
My comments will focus on the study I performed assessing the
overall safety profile of medical devices approved or cleared by FDA
from 2005-9 by using Class I safety recall data. This research was
funded by the Ewing Marion Kauffman Foundation, a private nonpartisan
foundation based in Kansas City, MO. Their generous support made this
research possible. The Kauffman Foundation has given me complete
academic freedom to pursue this research.\44\
---------------------------------------------------------------------------
\44\ I want to thank Amanda Maccoux, Mark Jones, Chris Walker and
Ron Song--the research assistants at the University of Minnesota Law
School--who spent long hours doing the detailed data collection and
coding required for this study. Their talents, hard work and dedication
are vital to this research and I appreciate all that they did.
---------------------------------------------------------------------------
This study \45\ evaluated Class I (or high risk) recalls of all
medical devices, regardless of whether they were approved through the
PMA system, cleared through the 510(k) process or were otherwise
exempt.
---------------------------------------------------------------------------
\45\ An earlier version of this research into the safety of medical
devices through an analysis of safety recalls was presented to the
Institute of Medicine committee reviewing the 510(k) system and
reviewed with FDA.
---------------------------------------------------------------------------
The key conclusions from my research are as follows:
(1) Overall, 510(k) regulated medical devices have an excellent
safety profile. Over 99.5 percent of 510(k) submissions assessed during
this study period did not result in a Class I safety recall. Over 99.7
percent of 510(k) submissions did not result in a Class I recall for
any reason relevant to the 510(k) premarket system.
(2) Products approved through the PMA system also have an excellent
safety record. Again, greater than 99.5 percent of PMA or sPMA
submissions do not result in a Class I safety recall during the study
period.
(3) Very few (less than 9 percent), Class I recalls during the
study period involve possible undiscovered clinical risks. As such,
increased preapproval clinical testing would not have any meaningful
impact on reducing the number of Class I recalls.
(4) The majority (approximately 55 percent) of all Class I recalls
involve problems or issues that arose after market release and could
not be affected by premarket approval systems or requirements. For
example, a manufacturing mistake made 3 years after FDA approval or
clearance may trigger a Class I recall. However, any premarket
requirements such as clinical testing are irrelevant to preventing such
a recall.
(5) A very significant majority (over 90 percent) of all Class I
recalls (including both premarket and postmarket issues) are directly
related to quality system issues (so-called QSR systems \46\). Improved
QSR systems will have the greatest effect in reducing the number of
Class I recalls.
---------------------------------------------------------------------------
\46\ QSR requirements are intended to provide ``cradle to grave''
product quality in a closed loop, learning system. QSRs include design
input and processes, design validation, product testing, manufacturing
controls, process controls, change controls, management review and
postmarket assessments. See, generally, 21 CFR 820.
---------------------------------------------------------------------------
(6) My study did identify a bolus of Class I recalls in two device
types--automatic external defibrillators (``AEDs'') and infusion pumps.
Any changes to the premarket review process should be targeted to
demonstrate problems rather than applied in some random, shotgun way.
(7) Finally, one should not confuse classification for premarket
review processes with recall classification. These are very different
things and serve very different purposes.
VI. Study Background
The need for the research that I will describe goes back several
years when a number of stakeholders started to question the robustness
of the 510(k) system. I was particularly struck by the fact that there
was no good, objective data to support or refute the assertion that the
510(k) system needed to be changed because of these presumed safety
issues.
Given my concerns over the lack of hard data, I commenced a study
(with the able assistance of four research assistants) assessing the
safety performance of FDA approval processes. To my knowledge, this was
the first study designed to systemically assess the safety performance
of the 510(k) system.
VII. Study Methodology
This study assessed the overall safety profile of medical devices
approved or cleared by FDA from 2005-9 by using Class I safety recall
data.
Class I safety recalls were chosen as the measure of safety as
these recalls involve any medical device problem posing any significant
risk of serious health consequences to patients and also correctly
exclude risks considered as part of the approval or review process.
Class II recalls involve generally remote risks to patients and Class
III recalls involve minimal or no risk to patients. FDA, not industry,
is responsible for assigning the recall classification.
Note that the Class of recall assigned by FDA is independent of the
product's device classification. For example, no one would argue that a
tongue depressor is a high-risk device or needs a clinical trial. For
premarket purposes it is classified as a low-risk, exempt device.
However, if the tongue depressor gets contaminated with deadly bacteria
because of product tampering or some manufacturing problem, there is a
significant risk to patients. This would be a high-risk or Class I
recall even though for premarket review purposes it is a low-risk
device.
Using FDA databases, we identified all Class I recalls posted by
FDA on public databases during 2005-9. We first combined all duplicate
recalls into one data set of unique or stand alone recalls. (FDA may
have several recall announcements and thus there may be multiple data
entries for the same issue because of different package configurations,
brand names or product sizes).
There were 118 unique recalls identified. We then coded each recall
for a number of factors including regulatory pathway, medical
specialty, whether implantable and three letter product code. We also
coded each recall with 1 of 13 reasons for recalls. Generally speaking,
these 13 recall reasons can be combined into three broad groupings of
premarket issues (i.e., something that could, at least theoretically,
have been discovered during a premarket review process), postmarket
issues and miscellaneous (counterfeit and ``quack'' products). We used
FDA Web sites and publicly available information for this coding.
This study must be assessed in light of the following factors and
limitations:
(1) First, we relied entirely upon publicly available data. We did
not identify any meaningful errors in this data but did not conduct any
structured assessment of the accuracy of FDA's data.
(2) Second, while companies are obligated to report recalls, there
may be situations in which the company failed to meet this obligation.
We believe that any such missing recalls would tend to be small and not
common because of the penalties for non-compliance and the variety of
information sources that would disclose any such recall. Importantly,
there is no reason to believe that the distribution of the causes of
such recalls would be different than the data we had.
(3) Third, we reviewed Class I recalls and not Class II recalls.
(FDA defines a Class II recall as a situation in which the problem
might cause a temporary health problem, or pose only a slight threat of
a serious nature.) We believe that Class I recalls represent all
recalls with any meaningful risk to patients and so that represents a
valid safety picture. Remember that Class II recalls are for remote
risks or low impact problems. Class I recalls represent the majority of
actual patient risk and tend to err in the direction of higher rather
than lower classification. Risks as low as 1/20,000 have been
classified as Class I recalls thus demonstrating the breadth of risks
captured by Class I recalls.
(4) Anecdotal review of some Class II recalls indicate (but does
not establish) the same general pattern of reasons for recalls between
Class I and Class II recalls.
(5) Finally we did not assess any effects of various regulatory
systems or actions on patient access to new products, innovation or the
economy in general.
We also determined the percentage of 510(k) submissions that
resulted in a subsequent Class I recall. The numerator for this
calculation is the number of recalls. The denominator is the number of
submissions. The denominator for this calculation is a close estimate
as there is no direct connection between the date of the submission and
the subsequent recall. For example, a recall for a design defect might
occur within a month after market release while a recall for a
manufacturing error or packaging mistake could occur literally years
after approval or clearance.
We determined an annualized number of submissions by taking the
average number of submissions for a 10-year period (2000-2009) and
annualizing that number. We used this number for all percentage
calculations. Those percentages, however, are approximations due to
this data challenge.
VIII. Study Results and Data
Initially, we looked at the reasons for recalls for these 118 Class
I recalls. We determined the reason for the recall by examining FDA's
public data bases and also reviewing publicly available information
including physician notification letters and SEC filings. I was
responsible for all decisions relating to the reason for recall. I
blindly recoded 10 percent of the recalls and had a complete match with
the initial determination of the reason for the recall.
As shown below, the majority of all recalls (approximately 55
percent) are for postmarket issues. For these recalls, no change in the
premarket 510(k) or PMA process would affect the recall occurrence or
frequency.
----------------------------------------------------------------------------------------------------------------
Recalled Percent of
Total Recalls for for Recalled recalls to
recalls premarket postmarket for other total
issues issues issues recalls
----------------------------------------------------------------------------------------------------------------
Class I or..................................... 7 1 6 0
u/k.......................................... (14.2%) (85.7%) (0%) 5.9
510(k)......................................... 95 43 46 6
(45.3%) (48.4%) (6.3%) 80.5
PMA............................................ 16 7 9 0
(43.8%) (56.3%) (0%) 13.56
----------------------------------------------------------------
Total...................................... 118 51 61 6 118
----------------------------------------------------------------------------------------------------------------
As seen below, a very small percentage of 510(k) submissions led to
a Class I recall during our study period. The first chart shows the
ratio of 510(k) submissions to all Class I recalls and the second chart
shows the ratio of 510(k) submissions to Class I recalls related to any
theoretical premarket issue.
Based on this data, approximately 99.55 percent of all 510(k)
submissions did not result in a Class I recall for any issue during the
study period. More importantly for assessing the 510(k) process,
approximately 99.78 percent of all 510(k) submissions did not result in
a Class I recall for any reason related to the premarket process.
Stated differently, the maximum theoretical impact of any change in the
510(k) system would be on 0.22 percent of all 510(k) submissions. This
data also demonstrates that additional premarket clinical testing would
be ineffective in reducing Class I safety recalls.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
------------------------------------------------------------------------
------------------------------------------------------------------------
Total 510(k) Submissions in 10 years....................... 39,747
Average Submissions in 5-year time period.................. 19,873
Total 510(k) Recalls for 2005-2009......................... 89
Total 510(k) Recalls for Premarket Issues for 2005-2009.... 43
The number of recalls related to premarket issues is most relevant
in assessing whether the 510(k) system is adequately addressing patient
safety during the review process. This data demonstrates that
postmarket issues, not premarket processes, should be the focus to
improve patient safety.
This conclusion is reinforced when we reviewed the role of quality
systems in recalls. As shown below, over 90 percent of all Class I
safety recalls are related to quality system issues and not to other
factors such as a lack of clinical trials.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Clearly, this data demonstrates that all stakeholders should
concentrate on QSR systems such as design control and bench testing--
not the 510(k) submission system--as the most effective way to provide
greater patient safety.
We also did sub-analysis by product type and medical specialty.
Such analysis can be used to identify concentrations of issues for
further investigation by FDA, industry and other stakeholders. As seen
below, Class I recalls are concentrated in several product types.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
medical specialty
Further analysis indicated that automatic external defibrillators
(AEDs) and infusion pumps accounted for 28 percent of all Class I
recalls and accounted for a substantial part of the bolus or recalls
seen in the cardiovascular and general hospital categories. Within the
past 9 months, FDA has triggered new regulatory initiatives for both
AEDs and infusion pumps.
Our confidence in our study design and results has been bolstered
by subsequent studies by others such as FDA itself, Dr. Maisel and
Battelle finding very similar numbers and reasons for Class I recalls.
IX. Study Conclusion
This study demonstrates that very few 510(k) medical device
submissions--less than 0.5 percent--become the subject of a Class I
safety recall. Even in this small number of Class I recalls, the
majority of Class I recalls involve postmarket issues such as
manufacturing mistakes, and are focused around two product categories
(cardiovascular and general hospital). These recalls involve quality
system issues, not premarket issues. Overall, in excess of 90 percent
of all recalls appear to involve quality system issues.
Our study shows that FDA has a very positive safety record in its
510(k) clearance decisions.
X. Conclusion
The current 510(k) system gives FDA substantial authority to clear
only products with a reasonable assurance of safety and effectiveness.
FDA has multiple tools beginning with initial product classification
and extending through special controls and data submission requirements
to assess product safety and effectiveness.
Overall, products approved or cleared by FDA have very good safety
records. Of course, all stakeholders should always be striving to
improve on this already good record. Improvements in QSR (quality
systems) offer the greatest potential patient benefit.
Again, I appreciate the opportunity to present to the committee and
would be happy to answer any questions.
STATEMENT OF DAVID R. CHALLONER, M.D., VICE PRESIDENT FOR
HEALTH AFFAIRS, EMERITUS, UNIVERSITY OF FLORIDA AND CHAIR,
COMMITTEE ON THE PUBLIC HEALTH EFFECTIVENESS OF THE FDA 510(K)
CLEARANCE PROCESS, INSTITUTE OF MEDICINE OF THE NATIONAL
ACADEMIES, GAINESVILLE, FL
Dr. Challoner. There we go, thank you.
I'm vice president for Health Affairs Emeritus at the
University of Florida, and I also served as chair at the
Institute of Medicine in this committee on the public health
effectiveness of the 510(k) clearance process.
The IOM is the health arm of the National Academy of
Sciences--an independent, non-profit organization that provides
unbiased and authoritative advice to decisionmakers and the
public.
I am accompanied this afternoon by several members of the
NRC and IOM staff, and by Mr. William Vodra, who was also a
member of my committee.
The IOM was asked by the FDA to review the 510(k) clearance
process, and to answer two questions, as has already been
pointed out. Does the current 510(k) process optimally protect
patients, and promote innovation and support of public health?
And if not, what legislative, regulatory or administrative
changes are recommended to optimally achieve the goals of the
510(k) process?
The IOM assembled an expert committee of which I was
chaired to address this task. The committee met in person 6
times over an 11-month period to gather evidence, deliberate on
its findings and recommendations, and write its report. That
report, then, underwent a rigorous, independent, external
review before being released in July of this year. More
detailed information on the committee's findings and
recommendations is included with my longer, written statement.
The committee's task was challenging for several reasons--
devices regulated within the 510(k) process encompass a broad
range of function, benefits, and risks. Also, the 510(k)
process is not a standalone process, but is part of a larger
regulatory system that is dependent upon all the components
functioning optimally to ensure the public health. And finally,
the 510(k) process was continuing to evolve as the committee
was conducting its work.
The committee evaluated legislative, regulatory and
administrative components of the 510(k) process, and other
related components of medical device regulation. We came to the
conclusion that the 510(k) process, as outlined in law,
generally does not evaluate the safety and effectiveness of a
device, only the new device's similarity to a predicate.
Furthermore, the 510(k) process cannot be transformed into
a premarket evaluation of safety and effectiveness, as long as
the standard for clearance is substantial equivalence to any
previously cleared device.
I want to emphasize that we believe the FDA has done an
admirable job of protecting the public, given the constraints
on its regulatory authorities, and limited resources, as we've
heard discussed earlier today. The committee recognizes that
replacing the 510(k) process will take some time, and that a
substantial amount of information will need to be obtained to
inform the design of a new framework.
However, there are steps that FDA can take now to improve
regulatory oversight of medical devices. The committee
believes, strongly, that it is important that regulatory
oversight of devices be conducted throughout their lifecycle.
In its report, the committee makes eight recommendations
geared toward using resources wisely to ensure both short-term
and long-term benefits. Among the actions recommended by the
committee are that the FDA strengthen its postmarketing
surveillance program for devices, identify limitations in the
use of its postmarket regulatory authorities, and to address
them, and implement a program of continuous quality
improvement.
We understand that the FDA, on its own initiative, is
already moving forward with making improvements to its
postmarketing surveillance, and quality assurance programs.
Ultimately, these changes will benefit patients, healthcare
providers, the industry, and the FDA.
Thank you very much for the opportunity to testify, and I
will be happy to answer any questions later.
[The prepared statement of Dr. Challoner follows:]
Prepared Statement of David R. Challoner, M.D.
summary
The Food and Drug Administration (FDA) asked the Institute of
Medicine (IOM) to review the 510(k) clearance process for medical
devices and to answer two questions:
1. Does the current 510(k) process optimally protect patients and
promote innovation in support of public health?
2. If not, what legislative, regulatory, or administrative changes
are recommended to optimally achieve the goals of the 510(k) process?
An expert committee assembled by the IOM addressed that task in its
report, Medical Devices and the Public's Health: The FDA 510(k)
Clearance Process at 35 Years. The report was released in July 2011.
On the basis of its review and evaluation of legislative,
regulatory, and administrative components of the 510(k) process and
other related components of medical-device regulation, the committee
concluded that the 510(k) process in general is not intended to
evaluate the safety and effectiveness of medical devices. Furthermore,
the 510(k) process cannot be transformed into a premarket evaluation of
safety and effectiveness as long as the standard for clearance is
substantial equivalence to any previously cleared device.
The committee recognizes that replacing the 510(k) process will
take some time and that a substantial amount of information will need
to be obtained to inform the design of a new framework. However, there
are actions that the FDA can take now to improve regulatory oversight
of Class II medical devices. In its report, the committee makes eight
recommendations geared toward using resources wisely to ensure both
short-term and long-term benefits. Among the recommended actions are
that the FDA should strengthen its postmarketing surveillance program
for devices, identify limitations in the use of its postmarket
regulatory authorities and address them, and develop and implement a
program of continuous quality improvement.
The committee believes that there should be an integrated premarket
and postmarket regulatory framework that provides a reasonable
assurance of device safety and effectiveness throughout the device
lifecycle. Implementation of the committee's recommendations will
improve regulation of Class II devices in the short term and inform the
design of the integrated regulatory framework in the long term.
Ultimately, these changes to the way Class II devices are regulated
will benefit patients, healthcare providers, the industry, and the FDA.
______
Mr. Chairman and members of the committee, I am David Challoner,
vice president of health affairs, emeritus, at the University of
Florida. I also served as chair of the Institute of Medicine's
Committee on the Public Health Effectiveness of the FDA 510(k)
Clearance Process. The Institute of Medicine, or IOM, is the health arm
of the National Academy of Sciences, an independent, nonprofit
organization that provides unbiased and authoritative advice to
decisionmakers and the public. Thank you for the opportunity to submit
testimony for the record based on the IOM's report, Medical Devices and
the Public's Health: The FDA 510(k) Clearance Process at 35 Years.
background
The Federal Food, Drug, and Cosmetic Act (FFDCA) requires a
``reasonable assurance of safety and effectiveness'' before a device
can be marketed. The U.S. Food and Drug Administration (FDA) is
responsible for enforcing this requirement. Devices that are deemed to
have a moderate risk to patients generally cannot go on the market
until they are cleared through the 510(k) process, named for Section
510(k) of the FFDCA.
The 510(k) process has become a major component of medical-device
regulation in the United States. Thousands of devices are cleared via
the 510(k) process each year--about one-third of devices entering the
market. The remaining devices are exempt from any premarket review (67
percent) or enter the market by the premarket approval (PMA) pathway (1
percent) or by other means such as the humanitarian-device exemption (1
percent).
In recent years, individuals and organizations have expressed
concern that the 510(k) process is neither making safe and effective
devices available to patients nor promoting innovation in the medical-
device industry. Several high-profile mass-media reports and consumer-
protection groups have profiled recognized or potential problems with
medical devices cleared through the 510(k) process. The medical-
device industry and some patients have asserted that the process has
become too burdensome and is delaying or stalling the entry of
important new medical devices to the market.
the charge to the iom committee
The FDA asked the IOM to review the 510(k) process for medical
devices and to answer two questions:
1. Does the current 510(k) process optimally protect patients and
promote innovation in support of public health?
2. If not, what legislative, regulatory, or administrative changes
are recommended to optimally achieve the goals of the 510(k) process?
the iom committee's conclusion on safety and effectiveness
On the basis of its review and evaluation of legislative,
regulatory, and administrative components of the 510(k) process and
other related components of medical-device regulation, the committee
came to the conclusion that the 510(k) process is not intended to
evaluate the safety and effectiveness of medical devices with some
exceptions. Furthermore, the 510(k) process cannot be transformed into
a premarket evaluation of safety and effectiveness as long as the
standard for clearance is substantial equivalence to any previously
cleared device.
the iom committee's recommendations
The committee believes that the FDA should obtain adequate
information to inform the design of a new medical-device regulatory
framework for Class II devices so that the current 510(k) process, in
which the standard for clearance is substantial equivalence to
previously cleared devices, can be replaced with an integrated
premarket and postmarket regulatory framework that effectively provides
a reasonable assurance of safety and effectiveness throughout the
device life cycle. Once adequate information is available to design an
appropriate medical-device regulatory framework, Congress should enact
legislation to do so. The committee believes that a move away from the
510(k) process should occur as soon as reasonably possible but
recognizes that it will take time to obtain the information needed to
design the new framework.
In its report, the committee outlines several actions that the FDA
should take in the short term to improve regulatory oversight of
medical devices. These actions also will serve to generate the
necessary information to inform the design of the new framework for
Class II devices.
The committee believes strongly that it is important that
regulatory oversight of devices be conducted throughout their
lifecycle. Premarket review, including the 510(k) process, and
postmarket oversight--from product labeling regulations to the
reporting of adverse events associated with use of a device--make up a
comprehensive medical device regulatory system. All the components of
the system need to be functioning well in order to provide a reasonable
assurance of the safety and effectiveness of medical devices.
No premarket regulatory system for medical devices can guarantee
that all new medical devices will be completely safe and effective when
they reach the market. Robust postmarketing surveillance is essential.
The FDA should give priority to postmarketing surveillance as an
invaluable investment in short-term and long-term oversight of medical-
device safety and assessment of device effectiveness. The committee
identified substantial problems in the current postmarketing
surveillance of devices, and recommends that the FDA develop and
implement a comprehensive strategy to collect, analyze, and act on
medical device aftermarket performance information. Congress should
support the capacity of the FDA's postmarketing surveillance programs
by providing stable and adequate funding.
The appropriate use of postmarket regulatory authorities, such as
seizing or banning a device, is an essential component of a successful
medical-device regulatory program. The FDA has stated that there are
limitations to the use of these authorities but has not identified the
limitations. The committee recommends that the agency review its
postmarket regulatory authorities to identify these limitations and
address them. If it is required, Congress should pass legislation to
remove unnecessary barriers to the FDA 's use of postmarket regulatory
authorities.
It is the committee's assessment that the FDA lacks a continuous
quality-assurance process for regulation of medical devices. As a
result, the FDA cannot effectively address new issues as they arise.
The committee recommends that the FDA develop and implement a program
of continuous quality improvement to increase predictability,
transparency, and consistency in all regulatory decisions for devices
and to address emerging issues that affect decisionmaking.
summary
The IOM committee believes that there should be an integrated
premarket and postmarket regulatory framework that provides a
reasonable assurance of device safety and effectiveness throughout the
device lifecycle. In its report, the committee outlines several actions
that should be taken by the FDA that will ensure both short-term and
long-term benefits. Among the actions recommended by the committee are
that the agency strengthen its postmarketing surveillance program for
devices, identify limitations in the use of its postmarket regulatory
authorities and mitigate them, and develop and implement a program of
continuous quality improvement.
Thank you, again. I would be happy to answer any questions the
committee might have.
The Chairman. Thank you very much, Dr. Challoner.
Dr. Curfman.
STATEMENT OF GREGORY D. CURFMAN, M.D., EXECUTIVE EDITOR, NEW
ENGLAND JOURNAL OF MEDICINE, BOSTON, MA
Dr. Curfman. Chairman Harkin, and other distinguished
members of the committee. My name is Gregory Curfman. I'm a
cardiologist and the executive editor of the New England
Journal of Medicine.
For 200 years, the New England Journal of Medicine has
published research on novel medical therapies. We're strongly
committed to innovation in medical treatments, but experience
has taught us that innovation does not come easily.
Now, Senator Blumenthal asked for a case study. And I'm
going to give you a case study. Exactly 2 months ago, on
September 15, we published an important clinical trial on a
novel medical device called Wingspan--a highly innovative blood
vessel stent. Wingspan was designed to reopen narrowed blood
vessels in the brain, and thus prevent strokes.
According to the American Heart Association, there are
795,000 new strokes each year in the United States, And a novel
therapy effective in stroke prevention would be a major
advance.
Wingspan looked very promising. In fact, video images of
the Wingspan in operation are nothing short of dramatic. Within
minutes of insertion, the Wingspan stent system can reopen a
blocked artery in the brain.
In 2004, a preliminary study of 45 patients showed that
Wingspan successfully reopened blocked arteries in a high
percentage. But, the study was very small, it included no
randomized controls, and so it was impossible to conclude that
Wingspan actually prevented strokes. Nevertheless, in 2005, on
the basis of this small study, Wingspan was approved for
unrestricted clinical use in Europe. It was also approved by
the FDA, under a humanitarian device exemption, which
authorizes use in up to 4,000 patients a year, but does not
require that the device be shown to be clinically effective.
It was not until November 2008, over 3 years after Wingspan
was approved for use, that a pivotal, randomized trial of its
clinical effectiveness, was begun. The study was funded, not by
the manufacturer of Wingspan, but instead by the taxpayers
through the National Institutes of Health.
In the trial, patients at high-risk of stroke were
randomized to receive Wingspan's stenting or no stenting. It
was expected that Wingspan would reduce the risk of stroke or
death. But it did not. In fact, there was a 2\1/2\-fold greater
risk of stroke or death in the Wingspan group than in the
unstented group--2\1/2\-fold greater risk.
Contrary to expectations, this innovative medical device
actually caused the very clinical problem that it was designed
to prevent. It was a disturbing result.
What are the lessons of Wingspan, which, by the way, is
still on the market with FDA approval. First, implantable
medical devices such as Wingspan are complex, and regardless
how innovative they may first appear, their ability to improve
human health cannot be known until they have been rigorously
tested in controlled clinical trials.
Second, benefit for a surrogate endpoint may not
necessarily translate into benefit for human health. In the
case of Wingspan, the surrogate endpoint was its success in
reopening the narrowed blood vessels in the brain. But this did
not prevent strokes, and in some patients, actually caused
them. The use of surrogate endpoints for FDA device approval is
advocated in Senate bill S. 1700.
Third, the European approach to medical device approval is
not an acceptable alternative for the United States. In Europe,
Wingspan received unrestricted approval, even though clinical
effectiveness had not been established.
Wingspan is but the latest of examples of a failed medical
device. Others include, as Senator Harkin already mentioned,
metal-on-metal artificial hip implants, and the Sprint Fidelis
implantable defibrillator lead, both of which were also high-
risk devices that received FDA approval without clinical data,
and harmed many patients.
Mr. Chairman, legislation recently introduced in both
chambers that would weaken the FDA approval process for complex
implantable medical devices, should be viewed skeptically. And
may not be in the best interest of the health of the American
people.
Thank you, Mr. Chairman.
[The prepared statement of Dr. Curfman follows:]
Prepared Statement of Gregory D. Curfman, M.D.
summary
Many Americans benefit from the implantation of medical devices;
tragically, many also suffer or even die from complications related to
medical devices that were never studied in clinical trials before being
implanted in a large population of patients. The current device
approval process, an outdated system created 35 years ago in an era of
much simpler and few devices, has become less capable of assuring
safety and effectiveness.
Two recent, but not rare, examples provide a cautionary tale about
the challenges of ensuring that complex medical devices are both
effective and safe. In 2005, a new metal-on-metal artificial hip
implant, the DePuy (Johnson & Johnson) ASR XL Acetabular System, was
cleared by the FDA through the 510(k) process by showing that the new
device was ``substantially equivalent'' to an already-marketed hip
resurfacing system. Because this was approved through the 510(k)
process, clearance was based not on clinical trials or other clinical
data but on bench testing in a laboratory. Once approved, it soon
became clear that the metal-on-metal hip implant failed at the
astonishing rate, requiring thousands of additional surgeries to
replace the defective, painful implants.
The second example, the Wingspan endovascular stenting system, was
approved through another less stringent humanitarian device exemption,
which primarily relied upon data from a small, non-controlled phase I
trial. A Phase III clinical trial showed a lack of both safety and
efficacy, 6 years later, just 2 months ago, . The disturbing experience
with the Wingspan stent system, which harmed many patients, serves as a
stark reminder that innovative medical devices, regardless of how
promising they may first appear on the basis of preliminary studies, do
not always prove to be successful when subjected to rigorous controlled
clinical trials.
Additionally, I believe that the July 2011 IOM report, which
concluded that it was impossible for 510(k) clearance to assure safety
and effectiveness, is insightful, judicious, sensible, and long
overdue. I support the IOM committee's recommendation that the 510(k)
process be replaced with an evaluation of safety and effectiveness.
As the best long-term improvements are contemplated, there are
important steps that the FDA can take now. First, the use of 510(k)
clearance for class III devices should stop, as Congress made clear 20
years ago. Second, the tenuous practice of allowing use of multiple
predicates in 510(k) clearance should be eliminated. Third, as was
recommended by the IOM committee, a formal system of postmarketing
surveillance for medical devices should be put in place. Fourth, I
strongly endorse the FDA's Sentinel Initiative and the associated Mini-
Sentinel pilot program. Fifth, I believe that the European medical
device regulatory system is not a suitable model for the United States
and would not be in the best interest of the American people.
I strongly believe that, in the interest of advancing human health,
patients must have easy access to innovative medical devices and that
the approval process needs to be sensible and efficient. But no one's
interest is served by putting defective medical devices onto the market
where they cause harm to patients, waste health care dollars, and may
kill jobs when they are withdrawn.
______
Many Americans benefit from the implantation of medical devices,
such as artificial joints and lifesaving defibrillators. Tragically,
many also suffer or even die from complications related to the same
types of medical devices, some of which have never been studied in
clinical trials before being implanted in a large population of
patients. As devices have evolved and become more complex, our device-
approval system has become less capable of assuring safety and
effectiveness. The system we use today was created 35 years ago in an
era of much simpler and fewer devices, and it is now inadequate.
A recent, but not rare, example provides a cautionary tale about
the challenges of ensuring that complex medical devices are both
effective and safe. Osteoarthritis of the hip joint is a common and
debilitating disorder. Each year, more than a quarter of a million
patients with advanced painful arthritis receive a total hip
replacement in the hope that it will restore mobility and improve their
quality of life.\1\ Conventional artificial hip implants consist of a
metal ball inserted into a plastic cup. In 2005, a new metal-on-metal
design was introduced in which both components were made from a metal
alloy. The design was touted as a major innovation that would improve
durability and reduce the risk of hip dislocation--advantages that were
especially appealing to younger patients. However, these design
innovations were never tested.
One metal-on-metal design is the DePuy (Johnson & Johnson) ASR XL
Acetabular System, which was introduced into the U.S. market in 2005.
The ASR was cleared by a Food and Drug Administration (FDA) process
known as 510(k), which refers to the section of the 1976 Medical Device
Amendments to the Federal Food, Drug, and Cosmetic Act that created it.
Under that section, the criterion for clearance of a new medical device
is that it be ``substantially equivalent'' to an already-marketed
device (a ``predicate''); clinical data are not required nor are data
on safety and effectiveness.
The ASR was constructed by borrowing a metal alloy cup from a
different hip device known as the ASR Hip Resurfacing System and
retrofitting it onto a standard hip implant. The manufacturer
successfully made the case that the re-engineered implant was
``substantially equivalent'' to a predicate device. Its marketing
clearance was therefore based not on clinical trials or other clinical
data but on bench testing in a laboratory, which was inadequate to
simulate the stresses that would be placed on it in patients' bodies.
It soon became clear that the device failed at the astonishing rate
of at least one in eight. According to a recent report presented at the
British Hip Society Annual Conference, 21 percent of these hips have
had to be replaced (revised) by 4 years after implantation, and the
revision rate rises to 49 percent at 6 years, as compared with 12 to 15
percent at 5 years for other devices.\2\ Failure appears to be due to
erosion of the metal in the articular surfaces and migration of
metallic particles into the surrounding tissues and the bloodstream.
Thus, the innovation led to tragedy for many patients.\3\ Before it was
recalled in 2010, the ASR had been implanted in nearly 100,000
patients, and the result was a public health nightmare.
The ASR is a class III device--the FDA's highest risk
classification. As a high-risk device, it should not be cleared
(without clinical data) via the 510(k) process, especially as its
design is novel and thus there is no predicate for a 510(k) clearance.
Congress envisioned that class III devices would be approved through
the more stringent premarket approval (PMA) process, which does require
clinical testing, and the Safe Medical Devices Act of 1990 requires
that the FDA either use the PMA process for class III devices or
reclassify them in a lower-risk category. Despite the clear intent of
Congress, a recent GAO report noted that most high-risk devices
continue to slip by this requirement. In fact, a recently published
study found that among high-risk device recalls from 2005 to 2009,
nearly three-quarters had been cleared through the 510(k) process.\4\
The Wingspan endovascular stenting system provides yet another
cautionary tale about the potential risks to human health of innovative
medical devices. The Wingspan stent was designed to be placed into
small blood vessels in the brain in patients at high risk of a stroke,
in order to re-open narrowed vessels to prevent a subsequent stroke
from occurring. The Wingspan system was approved for use in both Europe
and the United States in 2005. While in Europe the device received
standard approval by a notified body, in the United States the FDA
approved the device with a humanitarian device exemption (HDE), which
requires a less stringent approval process than standard pre-market
approval (PMA) and limits use to no more than 4,000 patients per year.
One phase I trial in 45 patients but no controls, which demonstrated
angiographic benefit, served as the basis for HDE approval. On the
basis of this phase I trial, the company optimistically referred to the
device as a ``groundbreaking system.''
Just 2 months ago, and 6 years after the Wingspan was approved by
the FDA, a phase III clinical trial (SAMMPRIS) comparing the device
with intensive stroke-prevention medical therapy was published in the
New England Journal of Medicine.\5\ The study was investigator-
initiated and funded by the National Institute of Neurological
Disorders and Stroke (the commercial sponsor, Stryker Neurovascular
[formerly Boston Scientific Neurovascular], donated the devices), and
thus was paid for principally by taxpayer dollars. The hypothesis
tested in the study was that the stenting system would improve patient
outcomes, as measured by the primary endpoint of stroke or death within
30 days of enrollment. After just 451 patients had been enrolled, the
study was terminated prematurely because of a serious adverse safety
signal in the stent-treated group. The incidence of the primary
endpoint (stroke or death) in the stent-treated group was 2\1/2\ times
greater than in the medically treated group (14.7 percent versus 5.8
percent), a worrisome result that was unanticipated by the
investigators. The comparable figures at 1 year were 20 percent and
12.2 percent. Despite these worrisome outcomes, the device remains
available in the United States.
The disturbing experience with the Wingspan stent system, which
harmed many patients, serves as a stark reminder that innovative
medical devices, regardless of how promising they may first appear on
the basis of preliminary studies, do not always prove to be successful
when subjected to rigorous controlled clinical trials. Implantable
medical devices are complex pieces of engineering, and bypassing
clinical testing to rigorously evaluate their function inside the human
body can result in substantial harm to patients.
On July 20, 2011, the U.S. House Energy and Commerce Subcommittee
on Oversight and Investigations held a hearing entitled ``Medical
Device Regulation: Impact on American Patients, Innovation, and Jobs.''
The subcommittee's chairman, Congressman Cliff Stearns (R-FL), argued
that FDA regulation of medical devices is too burdensome, stifles
innovation, and drives device manufacturers overseas. Since then a
number of bills have been introduced in Congress that would diminish
FDA's ability to assure safety and effectiveness of medical devices.
But the disastrous outcomes of the use of DePuy ASRs and the Wingspan
endovascular stenting system show that rushing untested and potentially
dangerous medical devices into the marketplace carries serious risks;
our regulators should not be in the business of creating jobs in the
manufacture of dangerous devices.
On July 29, 2011, the Institute of Medicine (IOM) released an FDA-
commissioned report on the 510(k) clearance process.\6\ \7\ The report
concluded that it was impossible for 510(k) clearance to assure safety
and effectiveness, because it assesses neither, instead establishing
only ``substantial equivalence'' to an existing device. The report
therefore recommended that 510(k) clearance be eliminated. In addition,
it recommended monitoring medical devices throughout their life cycle,
especially during the postmarketing period. Despite its reasonable (and
relatively modest) recommendations, the report has been aggressively
attacked by the device industry and by politicians from States where
device companies are located. In fact, the attacks began even before
the report was released, which is highly unusual for an IOM report.
I believe that the IOM report is insightful, judicious, sensible,
and long overdue. The 510(k) clearance process was established 35 years
ago, and although it may have been a reasonable approach then, it
surely is not today. The 510(k) process was never intended for use for
clearing Class III medical devices, defined by the Code of Federal
Regulations as products used for life-supporting or life-sustaining
indications, for preventing impairment of human health, or presenting a
potentially unreasonable risk of illness or injury. I support the IOM
committee's recommendation that the 510(k) process be replaced with an
evaluation of safety and effectiveness. It is important to maintain and
encourage innovation in medical devices. But true innovation requires
that safety and effectiveness be proven by scientific study in clinical
trials.
Under intense pressure from the device industry, the FDA leadership
has already indicated that it does not intend to implement this key
recommendation of the report, although it may be open to other changes.
As the best long-term improvements are contemplated, there are
important steps that the agency can take now.
First, the use of 510(k) clearance for class III devices should
stop, as Congress made clear 20 years ago. A substantial equivalence
standard for clearance of such complex devices is untenable. This
recommendation was made previously in a report from the Government
Accountability Office (GAO),\8\ but it has not been fully implemented
by the FDA.
Second, the use of multiple predicates in 510(k) clearance should
be eliminated. Now a device may be cleared if it is found to be
substantially equivalent to an existing device that was cleared, in
turn, by being found substantially equivalent to another device, and so
on. A device can receive 510(k) clearance by being substantially
equivalent to a device that is no longer on the market. This tenuous
process should be discontinued.
Third, if a substantial equivalence standard is to be retained for
certain devices deemed not of high risk, there must be a clear
definition of substantial equivalence including the authority of FDA to
require the submission of clinical data to assess whether the new
device meets the substantial equivalence definition.
Fourth, as was recommended by the IOM committee, a formal system of
postmarketing surveillance for medical devices should be put in place.
Strong, mandatory, and transparent postmarketing data, in registries,
allow rapid identification of serious problems that may emerge after
approval. Careful tracking of every patient with a high-risk device is
a crucial step for ensuring patient safety and avoiding nightmare
scenarios. To this end, I hope that the FDA will soon finalize its rule
about a system of Unique Device Identification (UDI), and then that the
Centers for Medicare & Medicaid Services will require the UDI to be
submitted with claims. That would allow safety surveillance for medical
devices to be much more tractable.
Fifth, I strongly endorse the FDA's Sentinel Initiative and the
associated Mini-Sentinel pilot program.\9\ Through the Mini-Sentinel
pilot program, capabilities are being developed for actively monitoring
the safety of approved medical products using the electronic health
information in claims systems, inpatient and outpatient medical
records, and patient registries. Such a system will be an important
step forward.
Sixth, I believe that the European medical device regulatory
system, in which 82 privately run notified bodies rather than a
government agency make decisions on market authorization for medical
devices, is not a suitable model for the United States and would not be
in the best interest of the American people. Notified bodies do not
adhere to uniform standards, and device manufacturers can select the
notified body that will put their device through the least stringent
assessment of safety and performance. Most surprising, manufacturers do
not have to demonstrate a beneficial effect on clinical outcomes.
I strongly believe that, in the interest of advancing human health,
patients must have easy access to innovative medical devices and that
the approval process needs to be sensible and efficient. But no one's
interest is served by putting defective or untested medical devices
onto the market where they cause harm to patients, waste health care
dollars, and may kill jobs when they are withdrawn. It is essential
that the FDA be adequately funded to carry out its mission to ensure
the safety and effectiveness of medical devices. The IOM report charts
a path that is right for the future, and despite well-financed outside
pressures, I urge the FDA to initiate an action plan with congressional
support to adopt these important recommendations.
References
1. National Hospital Discharge Survey: survey results and products.
Atlanta: Centers for Disease Control and Prevention, 2009. (http://
www.cdc.gov/nchs/nhds/nhds_products.htm.)
2. Updated guidance on large diameter metal on metal bearing total
hip replacements. London: British Hip Society and British Orthopaedic
Association, March 2011. (http://www.britishhipsociety.com/pdfs/
BHS_MOM_THR.pdf.)
3. Testimony of Katherine Korgaokar before the Senate Committee on
Aging, Washington, DC, April 13, 2011. (http://aging.senate.gov/events/
hr233kk.pdf.)
4. Zuckerman DM, Nissen SE. Medical device recalls and the FDA
approval process. Arch Intern Med 2011; 171(11): 1006-11.
5. Chimowitz MI, Lynn MJ, Derdeyn MD, et al. Stenting versus
aggressive medical therapy for intracranial arterial stenosis. N Engl J
Med 2011;365:993-1003.
6. Institute of Medicine. Medical devices and the public's health:
the FDA 510(k) clearance process at 35 years. Washington, DC: National
Academies Press, 2011.
7. Challoner DR, Vodra WW. Medical devices and health--creating a
new regulatory framework for moderate-risk devices. N Engl J Med
2011;365:977-79.
8. U.S. Government Accountability Office. Medical devices: FDA
should take steps to ensure that high risk device types are approved
through the most stringent premarket review process. GAO-09-190. 2009.
9. Behrman RE, Benner JS, Brown JS, McClellan M, Woodcock J, Platt
R. Developing the sentinel system--a national resource for evidence
development. N Engl J Med 2011;364:498-99.
The Chairman. Thank you very much, Dr. Curfman. And thank
you for that exposition of that one case, and I have a whole
list of others to go with it.
Professor Hall, you assert in your testimony that 510(k)
system generally works pretty well. But how do you reconcile a
system based on substantial equivalence with the phenomena of,
what they call, predicate creep? You know what that means--
where device after device is compared with a slightly newer or
a different version of an original device, such that the newest
models bear little resemblance to the original? You have the
original device, you have another iteration of it that's
substantially equivalent, then there's another device that is
substantially equivalent to that, and there's another device,
and on and on to the nth degree. Doesn't this system really
allow firms to compare apples to oranges, once you get out two
or three or four or five iterations?
Mr. Hall. Properly implemented, I do not think the system
does do that. Let me give you a couple of reasons why.
In each of these iterative steps, if there is either a new
indication or a new technology being employed, and again, new
technology is very, very broadly defined, the sponsor of that
510(k) submission has to provide data of whatever is the
appropriate type--clinical data, bench testing, whatever to
establish safety and effectiveness.
The agency reviews that and makes a decision. So, what you
see is a constantly increasing bar. That's how the system is
designed. So, then my latest generation with another tweak or
whatever, is not compared simply to the original one from,
let's say, 20 years ago, but rather, then, we have this
increasing knowledge.
Second, the agency has a number of tools to take into
account new information. We can revise special controls, put in
place performance standards, etc., all to address newly
discovered information or experience in the clinical setting.
The Chairman. Dr. Challoner, there was a lengthy discussion
in the IOM report about the need to enhance postmarket review
of devices. What sort of precise fixes, if any, would you
suggest? What are the gaps in the postmarket authorities that
you might respond to here?
Dr. Challoner. This is a major and important tool as you
look forward to where we're going to be in the next decade or
so in this industry.
At the moment, usually, except for a few cases in pediatric
realm or where you have an interested professional society,
there are some device-specific areas in which prospective data
on complications is gathered.
But, generally speaking, it's an ad hoc system with
multiple responsibilities for reporting up the chain, if you
will, to finally come to either public or FDA attention, and
there's no consistency or regularization of it.
Now, the opportunity is if you're going to speed up the
device process for public health safety and efficacy, to have
the opportunity to be not intrusive on the front end, except
for manufacturing standards and design issues, and to
streamline in those processes, as long as you could rely on a
very consistent population-wide reporting system for
complications of low incidence and high substance, you would
begin to have safety and efficacy over the lifetime of a drug.
New information systems----
The Chairman. Or device.
Dr. Challoner [continuing]. Or device. I'm sorry.
New information systems, new healthcare organizations,
organizations like Kaiser, and in particularly the VA, give us
an opportunity to get an early warning system for devices
through their electronic record systems, that will improve the
process over time. And I think that needs to be explored by all
the interested parties as we go forward.
The Chairman. My 5 minutes are up. I will turn now to
Senator Franken.
Senator Franken. Thank you, Mr. Chairman.
Professor Hall, thank you for coming from Minnesota. You
have decades of experience working with the FDA, and as CEO of
a medical device company, and also as a professor of food and
drug law at the University of Minnesota Law School.
As you know, I introduced a bill earlier today that will
get devices to the market faster and more safely, and I know
you know about the bill because you endorsed it, and thank you
again for your support. Would you explain why it is important
to expand the FDA's ability to consult with outside experts
while maintaining the requirement that the FDA disclose any
conflicts of interest that advisory panel members may have?
Mr. Hall. As we see burgeoning technology and whole new
areas of science, the agency cannot and should not be expected
to be an expert in anything and everything. And the agency, in
my opinion, needs access to experts to provide the scientific,
engineering, material science, whatever expertise, to advise
the agency so they, then, can make the appropriate risk-benefit
decision in a safety efficacy scenario.
I happen to have a personal interest in nanotechnology with
my academic hat on. It's incredibly complex, and if we don't
give the agency access to expertise, they're going to be making
decisions without the benefit of the best science. And we need
to make sure they have that, in my opinion.
Senator Franken. We heard from Dr. Shuren that they are
losing senior people, and there is attrition, and it would be
nice to work out a system where the higher bar of conflict of
interest--that exists in, just in this, in the FDA--in this
field is looked at again.
Can I, Professor Hall, can I get your take on the IOM's
report on the 510(k) process?
Mr. Hall. Let me make a couple of comments--there are
aspects of the report with which I do agree. For example,
postmarket surveillance, the importance of a quality system
within FDA, the need to finalize the classification of post-
amendment classification, and several others.
I must say, I do disagree with the fundamental conclusion
that the 510(k) system is not capable of providing a reasonable
assurance of safety and effectiveness. I think the statute in
Congress have provided that. I think FDA has implemented that,
and I think that even as the IOM said, the actual experience
with the system has been remarkably positive.
Senator Franken. Thank you very much. Thank you, Mr.
Chairman.
The Chairman. Senator Mikulski.
Senator Mikulski. Thank you. This is a great panel.
Dr. Challoner, first of all, we so value the Institute of
Medicine as Senator Harkin said. But this issue of totally
replacing the 510(k) gives us pause, and I think one of the
reasons it gives us pause is the looming question of who would
design the new system? Would we have to approve the new system
given both the political climate and the clock ticking--when we
also have to reauthorize MDUFA as we know it?
What prompted a learned society like IOM to say, ``Throw
out the whole thing,'' when you know how Congress is working--
or not working. If you're going to criticize FDA for approval,
criticize us for bill approval.
So do you, speaking for yourself, believe that there are
intermediate steps that we could take that would address the
most cogent and compelling of the issues raised in your report?
Dr. Challoner. Senator Mikulski, thank you, and the answer
is that the more we looked at it, and we had a very diverse
committee--I was on the board of directors of a device
corporation, a large one, Cortis Corporation, for 6 years. We
had two attorneys who spent most of their professional life in
the device industry. We had people, an individual, who had
taken a device through the process.
We were surprised, ourselves, as a group, that we all came
unanimously to the conclusion that the logic of this transition
system that was put in place with the 75 amendments, simply to
get from nothing to something, had now had 35 years of a life-
span, and we felt it was not going to be capable of dealing
with the technology and science of devices for the next two
decades.
Therefore, our expectation is not that the FDA would take
our advice, and dump it next week, and put something in place.
But we would begin a conversation such as we're having here,
and at other venues, just as it took 5 years to get the 75
amendments in place. It may take 5 years of conversation,
probably managed by the FDA, and the Congress, to be able to
put in place, that makes use of modern information technology
over the life-span of a device.
So, we expect a 5-year transition. In the meantime, there
are things that can be done with postmarket authorities that
the FDA has, and with improving postmarket surveillance that
will immediately improve public health and safety.
Senator Mikulski. When I read the report, I thought the
message was essentially--these are Barb Mikulski's words--dump
it and deal with it.
Dr. Challoner. No.
Senator Mikulski. And quite frankly, that message gave
every one of us a great deal of pause. But what you're saying
is--well, let's take a look at what you're saying--Mr. Hall,
Mr. Curfman, and Dr. Shuren have talked reforms; we're all
talking about reforms and what you are saying in your
recommendations is that the postmarketing suggestions would be
the most potent reforms right now to include in any
reauthorization bid.
Dr. Challoner. Yes. Would have immediate effects.
Senator Mikulski. I want to thank the IOM for what they
did. They certainly started the conversation.
Dr. Challoner. Good.
Senator Mikulski. And thank you very much. Thank you, Mr.
Chairman.
The Chairman. Senator Blumenthal.
Senator Blumenthal. Thank you, Mr. Chairman.
I think what the panel was hearing and what I've heard over
the last months, maybe longer, is a developing consensus that
the present system simply isn't working, that it's broken, and,
in fact, fatally defective, not just in implementation, but
also concept.
And I think Senator Mikulski raised the key question, which
is, what do we do to make it better while we try to improve the
law? And perhaps I can ask this question of all the members of
the panel. Isn't one of the things we can do is to improve
postmarket surveillance? Because that kind of greatly increased
scrutiny and oversight, at least, can stop harm that may be
developing, and save patients the hip on hip implants, or the
Wingspan stent problems, and try to deal with those problems at
an earlier stage, and thereby, save people from a lot of
suffering, and even death?
Dr. Curfman, you've raised those two case studies, and I
suspect that they may be included in Dr. Shuren's response to
my question on case studies. But, would increase postmarket
surveillance be advisable?
Dr. Challoner. Thank you, Senator Blumenthal. I think
that's critical, and I know that there are initiatives ongoing
in the FDA that are quite interesting. And I think that, as
they're implemented, they are going to really strengthen
postmarketing surveillance. Dr. Shuren can certainly speak to
that in more detail than I.
But, there is a program that is coming to fruition called
the Uniform Device Identification Program, which is a way of
giving a unique identifying number to each implanted medical
device, allowing that device to be tracked throughout its
lifetime, and hooked up with an individual patient.
These numerical codes can, then, be linked to outcome data,
claims data, so that we can, then, find out exactly what's
happening to the patients who have them.
Now, this program is very interesting. It's being headed by
a very high-quality individual within the agency, and I think
it's going to be a very important step forward when it's
actually implemented, and I understand that that will be fairly
soon.
That's associated with the larger sentinel initiative
that's ongoing within the FDA. That's associated with the
larger sentinel initiative that's ongoing within the FDA, and
the mini-sentinel pilot program, which is a smaller research
program going on to develop new ways of doing postmarketing
surveillance. And, here again, there are very high-quality
people working on these programs within the agency.
So, I expect that we will be seeing advances in
postmarketing surveillance, and I agree with you, Senator
Blumenthal, that this is quite critical as we await larger-
scale changes in the 510(k) program.
Senator Blumenthal. But, postmarket is really no substitute
for fundamental reform of the entire 510(k) system.
Dr. Challoner. It's part of a process, but I don't think
that it's really going to be enough, in the long run.
Senator Blumenthal. Professor Hall, I wonder if you would
comment on those two case studies, for lack of better word, and
whether you think they don't indicate faults in the present
process.
Mr. Hall. I'll be glad to.
First of all, I only know what's public about those
situations, and so, I'm not in a position or qualified to make
any conclusions. I have not reviewed the regulatory submissions
or anything like that.
So, what I will try to comment on, given on what I do know,
publicly, is whether the agency has authority to have addressed
those issues. I believe the agency does. In the case, again, of
the metal-on-metal, the agency, submission to the agency is to
include safety and effectiveness data. The agency can decide
whether it needs clinical data or not to make that assessment,
so they have that authority.
Dr. Shuren talked about that briefly in his comments, as to
whether they should have. And again, I don't know the details
on that. But they have the authority to get clinical data, if
that is what is appropriate in that situation. I do agree with
the other panelists of the importance of postmarket
surveillance, and the need to improve postmarket systems so
that they are more effective and more efficient.
Within any product issue, the agency has a number of
authorities to impose warnings, recalls, and postmarket studies
under section 522--I have to mention at least statutory
section, otherwise, I guess I'm not really a lawyer in this.
Senator Blumenthal. So, your point is that, even under
existing authority, the FDA has the power to be more vigilant
and vigorous in protecting the public, and, I think many of us
would agree.
Mr. Hall. Correct.
Senator Blumenthal. Thank you. Thank you, Mr. Chairman.
Senator Hagan. Thank you, Mr. Chairman. And, I too, echo
Senator Mikulski's comments about this panel. Thank you for
being here, and it's an excellent group.
Dr. Challoner, you state that the Institute of Medicine
Committee believes that there should be an integrated premarket
and postmarket regulatory framework that provides a reasonable
assurance of device safety and effectiveness throughout the
device lifecycle. Can you elaborate on this, and what you think
we need to see?
Dr. Challoner. We just don't have enough data at the
moment, and, our charge, if you will, in making our
recommendation to the FDA is that it needs to spend some time
with its varied constituencies over the course of the next
several years, finding out exactly what kind of data they need
about their processes, that will make them more transparent and
predictable.
Dr. Shuren has already undertaken, parallel with our
report, many of these items. But there are some things in which
we just don't have data.
There's a difference between postmarket surveillance and
the identifier issue that Dr. Curfman just raised, and
postmarket authorities already in place, for instance, that may
or may not be used adequately to survey and monitor the
postmarket arena. And it may be because the data just isn't
coming forth from the clinical environment to the leadership of
the FDA.
For instance, there's a seizure authority, and the agency
has brought about 13 seizure actions from fiscal year 2001 to
2008. Now, is that adequate? I don't know. The absence of
evidence is not evidence of absence; banning, used once since
1976; recall orders, the agency has not formally tracked recall
orders, but believes the authority has been used at least three
times in the last 20 years.
So, we were unable to get enough data to really understand
how to put all of this in place, which is why we suggested that
things need to change, and that a process needs to be put in
place to study it, gather the data, and understand it.
Senator Hagan. Thank you.
Professor Hall, we've heard from the other two panelists
that the 510(k) process or program isn't working well;
primarily from a safety perspective. However, you've got a
different view of this process.
Can you elaborate on the safety findings that you mention
in your testimony, and do you think the 510(k) program is,
generally, working well. Or should it be replaced, and what do
you envision the impact to industry and access to medical
devices are if the 510(k) program, was, in fact, scrapped?
Mr. Hall. There are a number of studies--I have done one
that has looked at the safety profile of medical devices. I
mention mine, Dr. Maisel, now with FDA, has done a study, FDA
recently did a study. IOM, itself, said that they find no
evidence of a systemic problem with medical devices.
What my study, in particular, did, is that we looked at all
class I recalls for a 5-year timeframe, looked at the reason
for the recall, etc. What that data indicates is that quality
systems are, by far, the most potent tool to use to increase
product safety.
This links in directly to your earlier question about
integrated postmarket and premarket. Outside of the formal
510(k) system, you have quality systems. So, for example, a
company is to have a corrective and preventive action program
by which they track information, product trends of problems,
then decide using that data--does a correction need to be made?
That data is also to be fed into the front end of the design
process for the next iteration of devices, so we get into this
continuous improvement loop, using data that exist.
Can that be made stronger? Do we need, I think we're all in
agreement we can do a better job on postmarket. But, the recent
study by FDA also indicated that the primary cause of recalls
are quality system challenges, not issues with, for example,
lack of clinical trials or other issues.
Senator Hagan. Thank you, Mr. Chairman.
The Chairman. Thank you, Senator Hagan.
I want to get on this issue of clinical data. I heard you
say, Mr. Hall, that FDA can--of course, we request clinical
data, to see if something is equivalent to the predicate.
If I reference the IOM report again here, in the United
States, use of clinical data in the regulatory review process
is defined by the enabling legislation, regulations and FDA's
implementation of the legislation and regulations. The clinical
data can be requested by the FDA only if necessary to determine
that the new devices are as safe and as effective as a
predicate device.
Moreover, the agency may not ask for scientific evidence,
greater than the ``least burdensome,'' to answer the question.
In practice, clinical data play a very small role in the 510(k)
process. The GAO found that from fiscal year 2005 to 2007,
about 15 percent of class II and class III 510(k) submissions
had new technologic characteristics.
The FDA found that only 8 percent of 510(k) submissions
include certain elements, I don't get into all of that. Less
than 1 percent of non-in-vitro diagnostic 510(k) submissions
reference a clinical trial. So, again, there's not very much
clinical data for the vast majority of the devices that are
asked for clearance under the 510(k) process, is that not true?
Mr. Hall. I think you have to look at the reasons the
510(k) was submitted. There are many 510(k)'s that are
submitted for changes in the product, for which clinical data
is simply irrelevant.
For example, one has to submit a 510(k) if one is adding a
warning to a product. Now, you don't need clinical data for
that.
Senator Franken asked about the new modifications guidance.
If you change suppliers under that, that requires a 510(k).
Again, clinical data would seem to be not important to deciding
whether it's appropriate to change from supplier A to supplier
B.
So, I think you need to look at the subset of submissions,
and then ask the question, does the agency have the authority
or is that information being submitted.
Senator Harkin. It would seem to me, Professor Hall--I ask
Dr. Curfman for his thoughts on this--that, if you're talking
about an implantable device, and a company is going through the
510(k) process, based on a predicate, obviously, that is where
postmarket surveillance would be most important.
What happened from the original through the first
iteration, the second iteration, the third iteration, the
fourth iteration? What happened to people out there? We don't
have that. The FDA is not really doing that, now.
Why aren't they doing it? Well, let's see. Let's go back
here to the report. The FDA's device postmarketing surveillance
programs have been adversely affected by the instability of the
agency's congressional financing. Interesting.
Moreover, user fees can be used only for premarket
activities. The inadequate postmarketing surveillance systems,
both those in the FDA and those which are privately funded, and
the resulting lack of useful, consistent, and reliable data
make it impossible to draw confident conclusions about the
performance of medical devices now in the market.
So, now I can get to the nub of it. If the clinical data is
not there because we don't have enough postmarket surveillance,
we don't have enough postmarket surveillance because Congress
doesn't fund it well enough, and user fees cannot be used for
postmarket surveillance, we have a conundrum. We have a
conundrum.
Mr. Hall. Let me try to be clear with my comments. There
are many circumstances in which clinical data is important. I
think you're absolutely--implantable, right.
The Chairman [continuing]. Implantable, right. I preface it
by saying implantable devices.
Mr. Hall. For many of those, it's important. I agree with
you completely, about the importance of postmarket
surveillance. I think you and I are in absolute agreement on
the importance of that, and the need to make sure that the
system is effective and efficient.
The Chairman. Let me ask you this question.
Mr. Hall. Yes.
The Chairman. You have a device company, or you're CEO of
one, or something--let me ask you this--should user fees also
be used for postmarket surveillance?
Mr. Hall. I believe in an adequately funded agency.
The Chairman. Hey, I'm asking you if----
Mr. Hall. I understand. And the reason I hesitate is--I'm
on record. I'm Don Quixote occasionally, OK? I'm on record as
saying that user fees are bad public policy. That doesn't mean
the agency shouldn't be funded. Alright? I just think the
source of funding--there are better sources of funding than
user fees. I recognize, completely, the current financial
situation, which is why I may be Don Quixote in my views on
that. But, I do think we need adequate funding.
One of the challenges of using user fees for postmarket
surveillance is that we are actually taxing innovation. Because
the people paying user fees are the ones with the new ideas,
the postmarket surveillance is for old products. They are
already out there.
And, so, that's one of the reasons, I think, we need to
come up with a better funding mechanism.
The Chairman. Yes.
Mr. Hall. But, that's me, and I didn't say----
The Chairman. But, the innovation, the new products is
based upon all those predicates that came before it.
Mr. Hall. Most of your user fees, as Dr. Shuren pointed
out, come from PMAs, not from 510(k)'s.
The Chairman. But, a lot still comes through the 510(k)
process.
Mr. Hall. Right. I'm in agreement. We need postmarket
surveillance, and it needs to be funded. I absolutely agree
with that.
The Chairman. Any observations from you, Dr. Challoner or
Dr. Curfman, about this? Dr. Curfman, I will single you out
first because you had, sort of, spoken about postmarket
surveillance.
Dr. Curfman. Mr. Chairman, I think you've focused, very
appropriately, on the implantable high-risk class III devices--
that's really where a lot of our concern is, and where we most
need the clinical information that you've referred to.
The use of the 510(k) clearance for class III devices
really needs to stop, and Congress called for that 20 years
ago. It still hasn't stopped. That needs to stop.
You've also referred to the use of multiple predicates--
this daisy chain of predicates to approve high-risk devices.
It's clearly inappropriate--I disagree with Mr. Hall about
this. It just doesn't make sense. And that, also, needs to be
eliminated.
If 510(k) is going to be retained for some devices of lower
risk, then we need to have a very clear definition of what
we're talking about for substantial equivalence, and FDA needs
to be able to call for clinical data to substantiate that
definition of substantial equivalence for a particular device.
So, I would agree with you, Mr. Chairman, we need more clinical
data in all of these realms.
The Chairman. Dr. Challoner, any observations on this? I
keep quoting from your report.
Dr. Challoner. Right. Thanks, you've already spoken for me,
Mr. Chairman.
The Chairman. The more I get into this--as a Chairman, I've
peripherally been involved in the past, the more I'm just
wondering if--perhaps a simple reauthorization is really not
what's needed. Perhaps, we have reached a point in time after
35, 36 years, that we need to take a more intense look at this
whole realm of the approval process, postmarket surveillance,
especially for certain higher risk devices. Obviously, some
were so low-risk, no big deal. And, perhaps, we need to step
back a little bit, and take a look at this.
I agree with the Professor, I fought all my adult life
against having user fees when I put on my agriculture hat,
against user fees for things like meat inspection. I mean,
you're going to have the companies that slaughter the meat do
the inspection? It's for the public good that we have the meat
inspectors.
I've often said all along that our FDA had to be fully
publicly funded. But, in this present climate, that is not
going to happen. And, so, the user fee system has been in place
for a long time, and it looks like something we're just going
to have to live with.
But, nonetheless, we still want to make this system one
that, first and foremost focuses on safety and that we have a
definable pathway that's transparent, that industry can rely
upon, and know, I mean, this is where I agree with the
industry. A lot of times, it's sort of opaque, on how the
process is going to go. And I think the IOM report pointed that
out also. So, we need some more clearly defined pathways.
I also say, just sort of off the top of my head after
reading all this, and going through the IOM report, I wonder if
there shouldn't be a limit on how many predicates there can be?
Maybe there should be the initial device. Maybe it can do one,
and maybe, to the second degree after that, they have to go
back and start the process over again, so you don't get devices
to the nth degree out there, that bear very little resemblance
to what was initially approved.
But, thinking about that, you limit the 510(k) process to a
certain limited number of predicates. After that, you'd have to
go back to the entire process again. And, then, looking at the
class III, the devices that are, well, I just call them
implantable devices, that really cause a lot of risk, and need
to have a lot of postmarket surveillance as we go along. How we
fund that, I just don't know. But, I'm thinking that, maybe,
user fees also need to be used for postmarket surveillance.
I know what you're saying, Professor Hall, that user fees
are to be approved. But a lot of times, the approval is based
upon what happened before, so I don't know where the two
separate, sometimes.
Those are just my thoughts. It's not an easy subject, and
it's not one that lends itself to any real easy answers, I
know. But, I think through this process, we might come up with
some better suggestions.
I'm concerned that we might be reauthorizing this, and not
doing a more adequate job of refining, and redefining the
process along some of the lines of the IOM, maybe not all of
it, but along some of the lines. And at least putting a
spotlight and some focus on, and some support for postmarket
surveillance.
Thank you all very much. This has been very healthy, a very
good interchange, and I learned a lot here, today.
I request that the record be kept open for 10 days so that
statements and questions can be submitted for the record. Did
anyone else have anything else they wanted to add?
Thank you all very much. I appreciate it, very much, for
being here. Meeting is adjourned.
[Additional material follows.]
ADDITIONAL MATERIAL
Response to Question of Senator Bennet by David R. Challoner, M.D.
Question. Professor Hall, you cited that over 90 percent of all
Class I recalls are directly related to quality system issues and Dr.
Curfman and Dr. Challoner you also cite the need for increased quality
assurance from companies as well. What can Congress do to directly
address this quality system and quality assurance problem?
Answer. The IOM Committee on the Public Health Effectiveness of the
FDA 510(k) Clearance Process focused on how the lack of an effective
continuous quality assurance program within the FDA's Center for
Devices and Radiological Health (CDRH) has hindered the agency's
ability to assess the effectiveness of the 510(k) program. As detailed
in the committee's report, without adequate management and information
technology infrastructure, the FDA cannot address new problems as they
arise or develop a long-term vision of CDRH and its mission. For
example, the FDA does not currently have the ability to trace the
history of the 120,000 or so 510(k) decisions made during the last 35
years and, therefore, the potential exists for problematic devices to
continue to be used as predicates because there is no systematic way to
identify them. The committee recommends that the FDA develop and
implement a program of continuous quality improvement to track
regulatory decisions on devices, identify potential process
improvements in the device regulatory framework, and address emerging
issues that affect decisionmaking. The committee did not recommend
specific actions by Congress with regard to establishing a program of
continuous quality assurance for devices within the FDA or the medical
device industry. It should be noted, however, that the committee did
determine that the FDA does not currently have adequate resources for
its multiple responsibilities and implementation of a quality assurance
program would require such from Congress.
[Whereupon, at 4:53 p.m., the hearing was adjourned.]