Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 149
6
External Factors That Affect the
Medical-Device Regulatory System
C
hapters 3, 4, and 5 review and analyze how the Food and Drug
Administration (FDA) regulates medical devices. However, that
regulatory process does not exist in isolation. It is a part of a broad
landscape consisting of such additional factors as the increasing complexity
of medical devices, the process of innovation, the business environment in
which medical devices are developed, and the international medical-device
regulatory arena. This chapter explores those factors and how they have
the potential to affect the regulation of medical devices in the United States.
THE GROWING NUMBER AND COMPLEXITY
OF MEDICAL DEVICES
The number of 510(k) submissions to the FDA varies from year to
year (see Figure 6-1). In 1976, fewer than 1,000 510(k) submissions were
received by the FDA’s Center for Devices and Radiological Health (CDRH).
The number of submissions reached about 7,000 in 1989 (in part because
of a change in the status of examination gloves from 510(k)-exempt to non-
exempt). In 2009, about 4,000 submissions were received (FDA, 2010a).
The number of 510(k) submissions declined most dramatically from the
early 1990s to the middle 2000s. At least three changes occurred during this
time, which might have impacted 510(k) submission numbers.
The first change was publication of the Temple report in 1993, a review
of the quality of clinical science submitted to CDRH in support of 510(k)
submissions and PMA applications (FDA, 1993). The report was critical,
observing (albeit on the basis of a small sample of PMA applications and
149
OCR for page 150
150 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
FIGURE 6-1 Original 510(k) submissions to CDRH, FY 1976–2009.
Figure 6-1.eps
SOURCE: FDA, 2010a.
bitmap
510(k) submissions containing clinical data) that studies submitted in sup-
port often failed to meet fundamental scientific standards. In response, the
FDA initiated recruitment of additional scientists, including physicians
and scientists, to perform premarket review of devices. While the review
standard for 510(k) submissions was unchanged (that is, substantial equiva-
lence), the review determinations began to shift from a descriptive to a
data-driven base. The second change was issuance of FDA’s final guidance,
in January 1997, advising industry as to when modifications to a cleared
device could be made without submitting a new 510(k) notification (FDA,
1997).1 This guidance gave manufacturers autonomy in making decisions
about devices with changes that were sufficiently minor as to preclude the
need for premarket review. Finally, passage of the FDA Modernization Act
in 1997 resulted in exemption of most Class I and many Class II devices
from premarket review.2 This Act resulted in decreased numbers of low-risk
and, in some cases, moderate-risk devices subject to premarket review.
Those changes in administration, regulation, and statue might have had
an effect on the volume of 510(k) submissions. The committee was unable to
draw conclusions, however, on the basis of the available data as to whether
innovation was influenced in any way by these changes.
1Updated guidance was issued by the FDA on July 27, 2011, after the committee had
completed its work.
2Food and Drug Administration Modernization Act of 1997, Pub. L. No. 105-115, 111
Stat. 2296 (1997).
OCR for page 151
151
EXTERNAL FACTORS
FIGURE 6-2 Number of 510(k) product codes, 1990–2009.
SOURCE: FDA, 2010a.
Figure 6-2.eps
bitmap
The number of types of medical devices also has grown. The FDA uses
product codes to identify generic categories of devices. The product codes
are organized by 16 medical specialties (for example, cardiovascular, general
and plastic surgery, and orthopedic), and the medical specialties are listed
in the Code of Federal Regulations.3 From 1990 to 2009, more than 1,000
product codes were added (see Figure 6-2).
In addition to the increase in 510(k) submissions to CDRH and the
greater variety of types of products that CDRH must review, submissions
have become longer and more detailed. As shown in Figure 6-3, the average
number of pages per 510(k) submission in 2008 was more than seven times
the number in 1983. In 2008, CDRH staff reviewed nearly 1.4 million pages
of 510(k) submissions. Reasons for the increase in the length of 510(k) sub-
missions are not readily apparent. The committee was not able to determine
how often some types of data (for example, clinical data, bench-testing data,
software-validation data, and labeling-comprehension studies) are included
in 510(k) submissions, nor was it able to review a representative sample of
complete (that is, not redacted) 510(k) submissions.
The technologic complexity of medical devices has increased sub -
stantially over the past 35 years as well. A 2010 FDA report states that
“devices are unique among medical products in that they are defined by
innovation, either through incremental evolution or disruptive revolution”
321 CFR §§ 862–892.
OCR for page 152
152 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
FIGURE 6-3 Pages per 510(k) submission and total page volume received, 1983–2008.
SOURCE: FDA, 2010a.
(FDA, 2010b). Examples of new technologies in medical devices are soft-
Figure 6-3
ware (incorporated in medical devices and as stand-alone medical devices),
nanotechnology, and medical robotics. The evolution (and revolution) of
science and technology creates many challenges related to the regulation of
medical devices.
As medical devices become more complex, so do 510(k) submissions.
As a result of the complexity and the increasing number and size of 510(k)
submissions received each year, the burden on CDRH review staff is increas-
ing (see Chapter 3). The growing complexity of medical devices also affects
how the medical-device industry approaches the 510(k) clearance process.
For example, industry representatives have indicated that they view the
increased length of 510(k) submissions as a response to the lack of predict-
ability in decision-making by CDRH. To avoid delays in clearance of their
products, industry includes more information in the submissions than it
might have in the past.4
Multiple Predicates and Split Predicates
As detailed in Chapters 3 and 4, predicates are used in the 510(k)
clearance process as the basis for demonstrating substantial equivalence.
The increasing complexity of medical devices is reflected in how predicates
4FDA Docket Number FDA–2010–N–0054.
OCR for page 153
153
EXTERNAL FACTORS
are selected and used. Applicants may submit more than one predicate for
several reasons, including these:
• he applicant is not sure which predicate is the most appropriate.
T
• he submission is part of a bundled submission.
T
• he new device combines functions of more than one predicate
T
device, so the applicant submits more than one predicate (termed
multiple predicates) to demonstrate substantial equivalence of the
new device.
• he new device has the same intended use as one predicate and the
T
same technologic characteristics as another predicate (termed split
predicates).5
Multiple predicates and, in some cases, split predicates are cited in
more than half the 510(k) submissions (FDA, 2010a). In general, multiple
and split predicates are used in 510(k) submissions for new devices that are
more complex than the predicates. Using more predicates leads to longer
CDRH review times (FDA, 2010a). It is not clear whether the increase in
review time of devices that have multiple predicates is related to the number
of predicates or to the increased complexity of the devices. It may be ap-
propriate for multiple and split predicates to play a role in premarket review
of Class II devices if the CDRH review team has the necessary expertise to
ensure a high-quality review and if appropriate postmarket activities (for
example, postmarketing surveillance) are used.
Combination Products
Combination products are therapeutic and diagnostic products that
combine drugs, devices, and biologic products (FDA, 2008). As new tech-
nologies emerge and older technologies evolve, combination products are
increasingly complex. Over the past decade, it has been increasingly com-
mon to enhance the performance of medical products by using multiple
products together. For example, a genetic test that is a 510(k)-cleared or
a premarket-approved (PMA) device may be used with a drug or biologic
with the aim of identifying patients whose genetic characteristics place
them at heightened risk for drug-related injuries, or a stent that is a medical
device may be coated with a drug with the aim of reducing complications
associated with the stent or altering the underlying pathologic condition
for which the stent was placed. Combined uses of medical products present
complex regulatory issues because the resulting treatment or diagnostic test
5In January 2011, CDRH announced that it no longer intends to use the term split predicate.
It plans to issue guidance to clarify the circumstances under which it is appropriate to use
multiple predicates to demonstrate substantial equivalence (FDA, 2011a).
OCR for page 154
154 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
incorporates multiple components that cut across the traditional categories
of FDA regulation—drug, device, and biologic.
In response to a growing trend toward combined uses of medical prod-
ucts, Congress called for the FDA to establish an Office of Combination
Products (OCP) in the Medical Device User Fee and Modernization Act of
2002. Combination product is a term of art that has a specific meaning.6
To qualify as a combination product, the two (or more) constituent prod-
ucts need to be integrally combined or mixed with one another, packaged
together in a single package as a unit, or, if packaged separately and in the
particular circumstance of being used in combination, cross-labeled in a
way that makes it clear that one is specifically intended for use with another
product that is “individually specified” in the labeling. By that definition,
the mere fact that products happen to be used together in clinical practice
does not necessarily make them combination products for purposes of the
FDA’s regulations. If products are simply used together in practice without
constituting a combination product, each product retains its separate iden-
tify and is regulated separately. For example, a drug would be regulated as
a drug and a device as a 510(k)-cleared or PMA-approved device. However,
even in that circumstance, OCP may play a role in the clearance or approval
process for the device, as is discussed in more detail later in this section.
When a pair or set of products does meet the definition of a combina-
tion product, special regulatory provisions apply. The aim is to reconcile
conflicts that would otherwise make it difficult to comply with drug, device,
and biologic regulations. On the basis of the combination’s primary mode
of activity—that is, whether the combination achieves its medical purpose
primarily through the action of the drug, the biologic, or the device—pri-
mary responsibility for premarket review is assigned to one of the FDA’s cen-
ters—the Center for Drug Evaluation and Research, the Center for Biologics
Evaluation and Research, or CDRH. The responsible center, while having
primary jurisdiction over the combination, will work closely with the other
centers to ensure appropriate oversight of issues related to the combinations
of other components. Because postmarket regulatory requirements also dif-
fer for drugs, biologics, and devices, OCP in 2008 prepared a proposed rule
on postmarketing safety reporting for combination products (FDA, 2008).7
A final rule has not been published as of May 2011.
Information for FY 2008 indicate that a large percentage of combination
products involve 510(k)-cleared devices. There were 330 original applica-
tions related to combination products. Of those, 120 were original 510(k)
submissions, 2 were original PMA applications, 14 were original new drug
applications, and 4 were original biologic license applications. The remainder
621 CFR § 3.2(e).
774 Fed. Reg., 50,744 (October 1, 2009).
OCR for page 155
155
EXTERNAL FACTORS
consisted of 158 applications for investigational new drug status and 32 for
investigational device exemptions (FDA, 2008). Thus, many combination
products enter the market on the basis of having the primary mode of activity
of the incorporated 510(k)-cleared device.
OCP also has a role in situations in which such products as medical
devices and drugs are used in concert but not as true combination prod-
ucts. OCP may act as a convener and facilitator in developing guidance
documents. Guidance documents can influence the regulatory review of
medical devices. For example, imaging tests use devices, such as magnetic
resonance imaging (MRI) and computed axial tomography machines; they
may use contrast agents, such as gadolinium and iodinated contrast agents
that are regulated as drugs; and they may use software, which is regulated
as a device, for the reconstruction and processing of the resulting data.
Guidance coordinated through OCP has influenced the review of software
devices on the basis of perceived limitations imposed by the specific lan-
guage that describes the indications for use of the contrast agents (FDA,
2009a). For example, the use of MRI contrast agents that are approved
for use in imaging of the abdomen, but not specifically imaging of the liver
or kidneys, might limit the ability of CDRH to review a software tool that
analyzes data produced by approved MRI machines after administration of
the MRI contrast agents if the data analyzed pertain to the liver or kidneys.
Guidance documents may stymie innovation in software tools that would
be used merely to analyze data that are being acquired daily in a clinical
setting. Because the contrast agent is readily available for use in clinical
practice, there is no practical mechanism whereby the software company
can influence the supplier of the contrast agent to apply for expanded or
specific new label indications.
Finally, the acceleration of device evolution is leading to increasingly
novel devices that could not have been conceived of when the 1976 Medical
Device Amendments and later amendments were written and enacted. For
example, nanodevices include particles that are activated by biologic and
pathologic processes, and some of them depend on metabolic pathways for
the mechanism of action. Atomic and molecular computational platforms
are under development. Complete “laboratories on a chip” are being deliv-
ered and used in vivo. Even previously contemplated combination products,
such as drug-eluting stents, have evolved in ways that stretch the capabilities
of the current system.
Some novel device–drug–biologic products—not truly combination
products—are not well managed on the basis of the current combination-
product concept. For example, nanoparticle drug-delivery systems (combi-
nation products of a drug or biologic encapsulated in a device) could be
activated by external ultrasonographic energy (delivered by a device) and
monitored by software analysis of resulting real-time images (a device).
OCR for page 156
156 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
Other than arbitrary assignment, there is no obvious lead agency for the
review and approval of such a multipart product; in fact, the review might
best be accomplished by multiple centers and branches.
Finding 6-1 Medical-device technologies have evolved rapidly and
devices have become increasingly complex since the 1976 Medical
Device Amendments.
Software
Software is used in medical devices, as medical devices (for example,
medical-device data systems), and as a tool in producing medical devices.
Manufacturers are increasingly using software in their medical devices. An
analysis by Fu showed that a milestone was reached in 2006: since then,
more than half the medical devices on the market have relied on software
in some way (IOM, 2011). Software offers many benefits over hardware
in some situations, for example, flexibility, ease of change, and usability in
other applications. To reduce costs, many software vendors produce “soft-
ware product lines,” that is, software designed to allow ease of variation to
accommodate similar, and perhaps tailored, product lines. In some cases,
general-purpose, off-the-shelf (OTS) software is incorporated directly into
medical devices. The FDA’s software validation guidance addresses that situ-
ation, noting that “the use of off-the-shelf software in automated medical
devices and in automated manufacturing and quality system operations is
increasing” (FDA, 2002).
Software Is Responsible for an Increasing Number of Recalls
As software becomes more common in medical devices, it also is in-
creasingly responsible for device failures and recalls. For example, CDRH
(FDA, 2002) notes that
the FDA’s analysis of 3140 medical device recalls conducted between 1992
and 1998 reveals that 242 of them (7.7%) are attributable to software
failures. Of those software related recalls, 192 (or 79%) were caused by
software defects that were introduced when changes were made to the
software after its initial production and distribution.
Fu noted that from 1999 to 2005, 11.3% of recalls were attributed to
software, and 49% of recalled devices relied on software in some way (IOM,
2011). In the period 2002–2010, at least 537 recalls of software-based
devices affected at least 1,527,311 devices. However, because of the limita-
tions of recall data discussed in Chapter 5, there is insufficient information
to determine the underlying issues in the increased rates. Because there is no
OCR for page 157
157
EXTERNAL FACTORS
FIGURE 6-4 Example of a medical-device software malfunction.
Figure 6-4.eps
SOURCE: Reprinted with permission from Halperin et al., 2008. © 2008 IEEE.
bitmap
way to know the number of devices on the market or how many of them
use software, it is not possible to know whether the change is related to
the increasing proportion of software in medical devices or whether it is a
signal of new or different types of problems and vulnerabilities in medical-
device software.
A drawback of software is that it can be difficult to recognize problems,
find their source, and fix them without adverse consequences. When soft-
ware is used in medical devices, at least three outcomes are to be avoided:
• Unreliable and unavailable devices.8
• Insecure devices.
• Unpredictable behavior.9
Figure 6-4 is an example of how a software failure can be manifested.
Here, the arrhythmia log is read from the bottom up. The episodes are in
8Reliability means that a system works as expected almost all the time; it rarely fails.
Availability means that the system works whenever it is needed to work. There is no delay,
and it is usable on demand.
9A device has unpredictable behavior if, for the same inputs, it gives different answers at
different times.
OCR for page 158
158 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
BOX 6-1
Example Recall Announcement
URGENT: Medtronic Announces Nationwide, Voluntary Recall of Model
8870 Software Application Card
MINNEAPOLIS, Sept. 22, 2004 - Medtronic, Inc. (NYSE: MDT) today an-
nounced a voluntary recall that involves all Version AAA 02 Model 8870
software application cards in the U.S. that are used in conjunction with
all Model 8840 N’Vision™ Clinician Programmers. This action has been
classified by the Food and Drug Administration (FDA) as a Class I Recall.
. . . a reasonable probability that the use of or exposure to the product will
cause serious adverse health consequences or death.
Medtronic became aware in August 2003 that some users had mistakenly
entered a periodic bolus interval into the minutes field, rather than the
hours field, potentially resulting in drug overdoses. Data entry errors
have been related to seven serious injuries and two deaths. The previous
model 8870 software application card did not provide a label for the hours/
minutes/seconds fields; the new software has this labeling.
SOURCE: FDA, 2009b.
chronologic order, and they inexplicably and suddenly move from 2007
back to 2005. Such an error can occur only when software is faulty.
Similar software failures are noted frequently in recall announcements.
For example, Medtronic issued a press release related to the Class I medical-
device recall of the Medtronic 8870 Software Application Card Version
AAA 02 (see Box 6-1).
The Guidance for Premarket Submissions for Software Contained in
Medical Devices for industry and FDA staff issued on May 11, 2005, poses
questions about problems with OTS medical-device software (FDA, 2005a):
What is it about “network-connected medical devices” that causes so much
concern? . . . Vulnerabilities in cybersecurity may represent a risk to the safe
and effective operation of networked medical devices using OTS software.
Failure to properly address these vulnerabilities could result in an adverse
effect on public health. A major concern with OTS software is the need
for timely software patches to correct newly discovered vulnerabilities in
the software.
The “vulnerabilities” are problems with the software that can be ex-
ploited by a malicious agent or can simply malfunction if the software is
subjected to particular conditions. For instance, poorly designed software
OCR for page 159
159
EXTERNAL FACTORS
may malfunction if it receives a type of input that it does not expect or
if it conflicts with other devices that are communicating along the same
channels. A patch is the application of a software correction; the resulting
modified software should work as desired. However, the FDA’s request for
“timely patches” suggests that patching is straightforward. In the commit-
tee’s opinion, it is not.
Beattie et al. (2002) analyzed the timing of patch availability and ap-
plication. The researchers concluded, “Patch too soon, and you may suffer
from instability induced by bugs in the patches. Patch too late, and you
get hacked by attackers exploiting the vulnerability.” And patches are not
always available as soon as a problem is discovered. The second Tuesday
of each month is known to Microsoft Corporation’s customers as “patch
Tuesday”: the day on which patches are made available for application to
existing systems. Only when a problem is considered an emergency is a
patch offered off-cycle to customers.
Microsoft is not alone. For example, Oracle (2010) notes that “Oracle
Sun releases over 4,500 patches every year, for Solaris 8, 9, and 10, SPARC
and x86, Solaris Cluster, Middleware, Developer, Storage, and other prod-
ucts. Just 17 have been withdrawn after release in the last year due to seri-
ous issues.”
Software Is Different from Hardware
Pfleeger et al. (2002) have written extensively on how software is dif-
ferent from hardware. The authors identify three key characteristics of
software that make it different from hardware:
• oftware developers are overoptimistic. Careful empirical studies
S
have shown that software developers, particularly testers, often as-
sume that they have found the last problem in the software under
scrutiny. That is, they commonly stop looking once they find a prob-
lem, not recognizing that other problems remain. Their optimism
results in overconfidence in both testing techniques and the degree
to which the tests exercise all the functions implemented by the
software. Such optimism is shared by hardware testers but seems to
be extreme in software testers. Indeed, Beck (2004) noted that “op-
timism is an occupational hazard of programming, feedback is the
treatment”; and Jorgensen (2010) has shown that, counterintuitively,
identification of more risks can lead to developers’ overconfidence
and overoptimism.
• oftware is discrete, not continuous. Unlike hardware, software is
S
extremely sensitive to small errors. Off-by-one errors, negligible in
hardware, can result in huge changes in software. Just as an off-by-
OCR for page 178
178 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
with the essential requirement itself. Harmonized standards are developed
and issued by the European Committee for Standardization (CEN) and in
some cases the European Committee for Electrotechnical Standardization.
Those two committees are responsible for writing the standards that deal
with both entire classes of products and specific products. In some cases,
the standards are adopted wholesale from the International Organization
for Standardization (ISO). Manufacturers are free to use whatever standards
they like, but if their devices meet the CEN or, where applicable, ISO stan-
dards, they are judged to have met the essential requirements—often the
easiest path to approval.
Risk Classification14
A device is assigned to one of four groups, or classes, by using a set
of rules that take into account the potential of the device to cause harm to
a patient or user. This system is referred to as a “risk-based classification
scheme” although it does not take into account the probability that harm
will occur by modifying evidence requirements at the conformity-assessment
stage. Risk is determined on the basis of the risk associated with the use
of the device, whether the device is invasive, and the length of time it is
in contact with the body (Table 6-1) (Chai, 2000). It is incumbent on the
manufacturer to determine the level of risk associated with a device on the
basis of the specifications laid out in Annex IX of the MDD (Grubb et al.,
2011). As in the FDA classification system, there are three risk classes, and
subdivisions of classes to delineate the magnitude of risk further.
Class I devices are noninvasive, with some exceptions, and are judged
to be of low risk. There is no explicit subdivision of products within Class
I, but devices that are sterile or have a measuring function require greater
oversight. Manufacturers are able to market Class I products that are non-
sterile and do not have a measuring function without oversight from a noti-
fied body (NB, defined below) but must keep documentation and technical
specifications available for audit on request. Class I devices that are sterile
or have a measuring function must be certified by an NB (Chai, 2000).
Class II devices are divided into Class IIa and Class IIb. Class IIa devices
require the involvement of an NB at the production stage but not the design
phase. Class IIb devices are judged to have the potential to be of high risk
and require NB approval at the design and manufacturing phases.
Class III devices are deemed to be of high risk and require “explicit prior
authorization” from an NB for all phases of development and production
(European Commission, 1998). Manufacturers can also turn to guidance
14The essential requirements, classification rules, and conformity-assessment procedures are
substantially different for in vitro diagnostic medical devices, which are regulated under their
own directive. Some of the information in the following sections does not apply to these devices.
OCR for page 179
179
EXTERNAL FACTORS
TABLE 6-1 European Commission Risk Classification Scheme
Class I Noninvasive devices except devices that are intended to store
(low risk) or channel blood, body liquids, or tissues to be introduced to
the body at a later time
Class IIa Invasive devices intended for transient (<60 min) or short-
(medium risk) term (<30 days) use except devices used to examine ear,
nose, mouth, and throat, which are in Class I; all surgically
invasive devices intended for transient use unless they emit
radiation or have a biologic effect, in which case they are in
Class IIb
Class IIb Surgically invasive devices for short-term or transient use
(medium to high risk) if they have a biologic effect, emit radiation, or administer
medicines except devices that have direct contact with
the heart or central circulatory system or central nervous
system, in which case they are in Class III; all implantable or
surgically invasive long-term (>30 days) devices unless they
are placed in the teeth (in which case they are in Class IIa)
or have contact with the heart, central circulatory system,
or central nervous system (in which case they are in Class
III); all devices intended to prevent conception or sexually
transmitted disease unless they are implantable or long-term
invasive devices (in which case they are in Class III)
Class III All invasive surgical devices, whether for short-term or
(high risk) long-term use, that come into contact with the heart, central
circulatory system, or central nervous system; all implantable
or long-term invasive devices that have a biologic effect,
are absorbed, or undergo chemical change in the body; all
implantable or long-term invasive devices intended to prevent
conception or sexually transmitted disease; all devices that
use animal tissues or derivatives except where they come into
contact only with skin
documents released by the overseeing body in the European Commission,
the Directorate-General Health and Consumer that provide detailed instruc-
tions on classifying devices, including diagrams and examples. The guidance
is not legally binding but is provided for clarification purposes (European
Commission, 2010a).
Notified Bodies
The directives contain provisions for the establishment of NBs, the
backbone of the European regulatory structure for medical devices. An NB
is a third-party private organization that is responsible for ensuring that
a device meets the essential requirements laid out in the legislation. Each
OCR for page 180
180 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
member state designates a competent authority (CA), an official government
regulator, that is required to oversee the accreditation and designation of
third-party organizations that will perform conformity assessments and
various other tasks. The CA then notifies the European Commission and
the other member states as to which private organizations have been judged
to be competent according to the standards laid out in the directives and
for which tasks. The commission maintains an updated list of the NBs and
makes this information available to the public through the New Approach
Notified and Designated Organisations Web site (NANDO, 2011). NBs are
accredited under individual directives, and their number varies by directive.
For example, although more than 70 NBs are accredited to examine prod-
ucts under the MDD, only 19 are accredited for devices covered by the more
specialized AIMDD (NANDO, 2011). The NBs compete with one another
in that any manufacturer can go to any accredited NB in any member state
and receive a CE marking.
NBs must adhere to standards regarding conflict of interest and finan-
cial incentives. NBs must be independent of manufacturers. Directors and
assessment and verification staff must not be involved in the development
of the device in question, nor can they represent any party involved in its
development. Their compensation structure must not depend on the results
of their evaluations or on the number of verifications that they provide. They
must have the necessary staff and facilities to assess properly the devices for
which they are designated. Any subcontractors retained by an NB must also
meet all these requirements.
Although an NB is responsible for ensuring a device’s conformity with
the essential requirements, it also has a contractual relationship with the
manufacturer, which pays for the work. That affords some protection to the
manufacturer because, in theory, an NB that is inefficient or difficult to deal
with will soon find itself without any customers. The accreditation proce-
dures and standards of the European Commission ensure that the standards
of safety are met, and the private status of the NB ensures efficiency and
consistency for the manufacturer.
As noted above, the level of involvement of an NB is determined by the
classification of the product (Table 6-2).
The NB has the responsibility to inform the CA if it withdraws a design
examination certificate from the manufacturer. The CA then ensures that
the device is withdrawn from the market. However, the manufacturer has
the right to respond to a decision to restrict or withdraw a device before it
goes into effect unless there is an urgent need to take such a step.
OCR for page 181
181
EXTERNAL FACTORS
TABLE 6-2 Conformity Assessment Procedure by Class
Class I Manufacturer certifies that product meets essential
requirements without involvement of an NB unless device has
measuring function or is sterile; documentation of product
specification must be kept for auditing purposes; Class I
devices must be registered with a member state
Class IIa Manufacturer prepares technical documentation to support a
“declaration of conformity” and makes it available to the NB
for inspection and adopts a quality-management system that
applies to manufacturing and is subject to audit by the NB,
or manufacturer adopts a full quality-management system
that applies to both manufacturing and design control and is
subject to audit by the NB
Class IIb Manufacturer submits to type examination by an NB in
which representative sample of product along with relevant
technical documentation is examined to ensure that it fulfills
essential requirements and adopts a quality-management
system that applies to manufacturing and is subject to
audit by the NB, or manufacturer adopts a full quality-
management system that applies to both manufacturing and
design control and is subject to audit by the NB
Class III and Active Manufacturer submits to type examination by an NB
Implantable Devices in which representative sample of product with relevant
technical documentation is examined to ensure that it fulfills
essential requirements and adopts a quality-management
system that applies to manufacturing and is subject to audit
by the NB, or manufacturer adopts a full quality-management
system that applies to both manufacturing and design control
and is subject to audit by the NB; NB also examines design
dossier and issues a certificate to confirm that the device
complies with all relevant essential requirements
Postmarket Vigilance
In addition to the postmarketing surveillance functions of the NBs, the
MDD mandated the establishment of a European Databank on Medical
Devices, a centralized database containing information on manufacturers,
devices, conformity assessments and certificates, and clinical investigations.
Its use is currently voluntary but will be required of member states in 2011
(European Commission, 2010c). The committee is not aware of any com-
prehensive evaluations of the European system based on postmarketing
surveillance or adverse-event reports.
OCR for page 182
182 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
The Global Harmonization Task Force
The multinational nature of medical-device manufacturing and distribu-
tion has led governments, consumers, and industry to recognize the need
for international cooperation regarding medical devices. In 1992, regulatory
bodies and industry of the leading medical-device producer and consumer
nations—Canada, the European Union/European Free Trade Association,
Japan, Australia, and the United States—formed the GHTF. A collaborative
body whose purpose is to improve public health and safety, promote inter-
national trade, and provide guidance to countries with developing medical-
device regulatory systems, the GHTF works to form consensus among the
partners on regulatory and technical standards and postmarketing surveil-
lance efforts. Many international agreements are in place regarding medical
devices and their regulatory requirements for importation. The GHTF is an
attempt to bring nations that tightly regulate medical devices into line with
one another to ease the flow of devices between countries, manufacturers,
and consumers while ensuring safety. In addition to participants from the
founding members, the task force includes participants from non–founding
members and liaison bodies, which include public-health organizations and
international standard-setting organizations.15
The GHTF is structured around five study groups that are responsible
for examining different issues and different paths and barriers to harmoni-
zation within them. The study groups, made up of geographically diverse
members from government and industry, are (GHTF, 2008)
• tudy Group 1—medical-device regulatory systems, paths to harmo-
S
nization, and standards for premarket submissions and labeling.
• tudy Group 2—adverse-incident reporting, postmarketing surveil-
S
lance, and harmonizing data-collection and reporting systems.
• tudy Group 3—existing quality-system requirements.
S
• tudy Group 4—quality-system auditing practices.
S
• tudy Group 5—clinical safety and performance.
S
To date, the study groups have released 31 final documents providing
technical guidance and standardized definitions and outlining generally
agreed-on principles for use by regulators and manufacturers worldwide
(GHTF, 2009). The GHTF has met with mixed success in its efforts. For
example, the Summary Technical Document (STED) guidance is intended to
15In March 2011, the Steering Committee Chair of the GHTF announced that the organization
would be developing and transitioning to a “regulator-led harmonisation and collaboration
group.” According to the statement, the new entity would focus primarily on regulators of
medical devices, including seeking new members in order to more accurately reflect the current
medical-device market, while continuing to value input from industry and other stakeholders
as appropriate (GHTF, 2011).
OCR for page 183
183
EXTERNAL FACTORS
standardize the format whereby a manufacturer submits technical informa-
tion to a regulatory authority or conformity-assessment body (such as an
NB). STED would allow manufacturers to save time and effort in preparing
a product for market and allow better comparison and information-sharing
among regulatory jurisdictions. Although the STED guidance has been ad-
opted or implemented as a pilot program in all five GHTF founding-member
jurisdictions, it has not been widely used (IOM, 2010). Nations that have
highly developed regulatory frameworks agree in principle that harmoniza-
tion is desirable, but differences in legal systems and regulatory cultures
make implementation challenging.
The Summary Technical Document Program
In 2002, GHTF Study Group 1 proposed a pilot program for device
premarket review known as the Summary Technical Document (STED),
program (GHTF, 2008). The FDA then issued draft guidance to assist the
medical-device industry and FDA staff in implementing a voluntary pilot
program (FDA, 2005b). Not all devices are eligible to participate in the
pilot. Eligible devices are limited to those which are seen to be the subjects
of common and substantial interest by the member countries of the GHTF.16
The program, begun in June 2003, is intended to evaluate STED as an
alternative premarket submissions process. Its objective is to decrease the
regulatory burden and increase the efficiency of the approval process for
manufacturers that operate within the international market (FDA, 2005b;
GHTF, 2008). Industry groups have stated their support for the STED
program, but the FDA has identified only 27 STED submissions17 (IOM,
2010). Of the submissions, 24 were 510(k) submissions, and 3 were PMA
applications. The PMA applications were all approved. Of the 510(k) sub-
missions, 20 resulted in a determination of substantial equivalence, 2 were
not substantially equivalent, and 2 were withdrawn. Given the lack of data
needed to evaluate the program and the small number of devices eligible, it
is not possible to determine what factors contribute to the lack of industry
participation in the program.
Finding 6-8 Other countries that tightly regulate medical devices do
not rely solely on substantial equivalence to a predicate for premar-
16A list of eligible devices and information on the alternative format (developed by the Global
Harmonization Task Force) can be found in Appendix C of the Guidance on Traditional and
Abbreviated 510(k)s.
17The FDA’s Document Mail Center identifies all submissions on receipt. STED submissions
are flagged in the system. However, the FDA has stated that there are limitations in the system,
and that the data on STED submissions may not be accurate (e-mail from Philip R. Desjardins,
FDA, January 7, 2011).
OCR for page 184
184 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
ket review of medium-risk devices. The Global Harmonization Task
Force also does not offer as part of its guidance a predicate-based
system for premarket review of medical devices.
SUMMARY OF FINDINGS
The Growing Number and Complexity of Medical Devices
• inding 6-1 Medical-device technologies have evolved rapidly, and
F
devices have become increasingly complex since the 1976 Medical
Device Amendments.
• inding 6-2 Manufacturers are using increasing amounts of software
F
in devices and as devices; the increase is expected to continue. Soft-
ware offers many benefits over hardware, including flexibility, ease
of change, and the possibility of use in other devices.
• inding 6-3 Software is responsible for an increasing number of
F
recalls. There are insufficient data, however, to determine whether
the increase reflects the increasing proportion of software in medical
devices or a new and different set of problems and vulnerabilities.
• inding 6-4 Software is different from hardware and therefore re-
F
quires a different kind of evaluation.
The Medical-Device Ecosystem
• inding 6-5 Industry-funded assessments of the effects of the 510(k)
F
clearance process report that implementation of the process leads to a
lack of predictability and transparency, which in turn has an adverse
effect on venture-capital investment.
• inding 6-6 The committee did not find assessments of how much
F
and in which way innovation is influenced by the 510(k) clearance
process.
• inding 6-7 There is little collaboration in collection of postmarket-
F
ing surveillance data among the FDA, healthcare facilities, healthcare
providers, the medical-device industry, professional societies, payers,
and patient-advocacy groups.
Globalization of the Medical-Device Industry
• inding 6-8 Other countries that tightly regulate medical devices do
F
not rely solely on substantial equivalence to a predicate for premar-
ket review of medium-risk devices. The Global Harmonization Task
Force also does not offer as part of its guidance a predicate-based
system for premarket review of medical devices.
OCR for page 185
185
EXTERNAL FACTORS
REFERENCES
Ahn, H., M. Bhandari, and E. H. Schemitsch. 2009. An evidence-based approach to the adop-
adop-
tion of new technology. Journal of Bone and Joint Surgery American Volume 91(Suppl
3):95-98.
Barkun, J. S., J. K. Aronson, L. S. Feldman, G. J. Maddern, S. M. Strasberg, D. G. Altman,
J. M. Blazeby, I. C. Boutron, W. B. Campbell, P. A. Clavien, J. A. Cook, P. L. Ergina,
D. R. Flum, P. Glasziou, J. C. Marshall, P. McCulloch, J. Nicholl, B. C. Reeves, C. M.
Seiler, J. L. Meakins, D. Ashby, N. Black, J. Bunker, M. Burton, M. Campbell, K.
Chalkidou, I. Chalmers, M. de Leval, J. Deeks, A. Grant, M. Gray, R. Greenhalgh, M.
Jenicek, S. Kehoe, R. Lilford, P. Littlejohns, Y. Loke, R. Madhock, K. McPherson, P.
Rothwell, B. Summerskill, D. Taggart, P. Tekkis, M. Thompson, T. Treasure, U. Trohler,
and J. Vandenbroucke. 2009. Evaluation and stages of surgical innovations. Lancet
374(9695):1089-1096.
Beattie, S., S. Arnold, C. Cowan, P. Wagle, and C. Wright. 2002. Timing the application of
security patches for optimal. Presented at LISA 02: Sixteenth Systems Administration
Conference, Berkeley, CA.
Beck, K. 2004. Extreme programming explained: Embrace change. Boston, MA: Addison-Wesley.
Berkun, S. 2010. The myths of innovation. Sebastopol, CA: O’Reilly Media, Inc.
Bloomfield, R., and B. Littlewood. 2003. Multi-legged arguments: The impact of diversity upon
confidence in dependability arguments. Presented at 2003 International Conference on
Dependable Systems and Networks (DSN 03).
Chai, J. Y. 2000. Medical device regulation in the United States and the European Union: A
comparative study. Food and Drug Law Journal 55(1):57-80.
Davis, S., E. Gilbertson, and S. Goodall. 2011. EU medical device approval safety assessment:
A comparative analysis of medical device recall 2005-2009. Boston Consulting Group.
Defend, B., M. Salajegheh, K. Fu, and S. Inoue. 2008. Protecting global medical telemetry
infrastructure. Amherst, MA: I3P Technical Report.
Ergina, P. L., J. A. Cook, J. M. Blazeby, I. Boutron, P. A. Clavien, B. C. Reeves, C. M.
Seiler, D. G. Altman, J. K. Aronson, J. S. Barkun, W. B. Campbell, L. S. Feldman,
D. R. Flum, P. Glasziou, G. J. Maddern, J. C. Marshall, P. McCulloch, J. Nicholl, S. M.
Strasberg, J. L. Meakins, D. Ashby, N. Black, J. Bunker, M. Burton, M. Campbell, K.
Chalkidou, I. Chalmers, M. de Leval, J. Deeks, A. Grant, M. Gray, R. Greenhalgh, M.
Jenicek, S. Kehoe, R. Lilford, P. Littlejohns, Y. Loke, R. Madhock, K. McPherson, P.
Rothwell, B. Summerskill, D. Taggart, P. Tekkis, M. Thompson, T. Treasure, U. Trohler,
and J. Vandenbroucke. 2009. Challenges in evaluating surgical innovation. Lancet
374(9695):1097-1104.
Ernst and Young. 2009. Pulse of the industry: Medical technology report.
European Commission. 1998. Commission communication on the application of transitional
provision of directive 93/42/EEC relating to medical devices (98/c 242/05).
———. 2001. Interface with other directives: Guidelines relating to the application of: The
council directive 90/385/EEC on active implantable medical devices the council directive
93/42/EEC on medical devices.
———. 2010a. Classification of medical devices. http://ec.europa.eu/consumers/sectors/
medical-devices/files/meddev/2_4_1_rev_9_classification_en.pdf (accessed 02/07/2010).
———. 2010b. Consultation on the general product safety legislative initiative. http://
ec.europa.eu/consumers/safety/prod_legis/GPSD_consultation/index_en.htm (accessed
02/07/2010).
———. 2010c. Exploratory process on the future of the medical devices: Potential themes
for further reflection at the European level and issues identified by the members. http://
ec.europa.eu/consumers/sectors/medical-devices/files/exploratory_process/final_report_
en.pdf. (accessed 02/07/2010).
OCR for page 186
186 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
———. 2010d. Roadmap: Revision of the general product safety directive (2001/95/ec).
h ttp://ec.europa.eu/consumers/safety/prod_legis/GPSD_consultation/docs/GPSD_
roadmap_en.pdf (accessed 02/07/2011).
FDA (Food and Drug Administration). 1993. The Temple report: Final report of the commit-
tee for clinical review. Based on a review of selected medical device applications. Silver
Spring, MD: Food and Drug Administration.
———. 1997. Deciding when to submit a 510(k) for a change to an existing device. http://
www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/
ucm080235.htm (accessed 08/20/2010).
———. 2002. General principles of software validation: Final guidance for industry and FDA
staff. http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/
GuidanceDocuments/ucm085371.pdf (accessed 05/20/2011).
———. 2005a. Guidance for the content of premarket submissions for software contained
in medical devices. http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/
GuidanceDocuments/ucm089543.htm (accessed 05/20/2011).
———. 2005b. A pilot program to evaluate a proposed globally harmonized alternative
for premarket procedures. http://www.fda.gov/MedicalDevices/DeviceRegulationand
Guidance/GuidanceDocuments/ucm071520.htm (accessed 06/2/2011).
———. 2008. FY 2008 performance report to Congress for the Office of Combination Prod-
ucts. Silver Spring, MD: Food and Drug Administration.
———. 2009a. Guidance for industry: New contrast imaging indication considerations for
devices and approved drug and biological products. http://www.fda.gov/downloads/
RegulatoryInformation/Guidances/UCM126051.pdf (accessed 02/07/2011).
———. 2009b. Medtronic announces nationwide, voluntary recall of model 8870 soft-
ware application card. http://www.fda.gov/MedicalDevices/Safety/RecallsCorrections
Removals/ListofRecalls/ucm133126.htm (accessed 01/03/2011).
———. 2009c. Registered foreign establishments as of June 15, 2008. http://www.fda.gov/
MedicalDevices/DeviceRegulationandGuidance/InternationalInformation/ucm134816.
htm (accessed 02/07/2011).
———. 2010a. CDRH preliminary internal evaluations—Volume I: 510(k) working group pre-
liminary report and recommendations. Silver Spring, MD: Food and Drug Administration.
———. 2010b. CDRH preliminary internal evaluations—Volume II: Task force on the utiliza-
tion of science in regulatory decision making preliminary report and recommendations.
Silver Spring, MD: Food and Drug Administration.
———. 2010c. Innovation or stagnation: Challenge and opportunity on the critical path to
new medical products. http://www.fda.gov/ScienceResearch/SpecialTopics/CriticalPath
Initiative/CriticalPathOpportunitiesReports/ucm077262.htm (accessed 02/28/2011).
———. 2010d. What we do. http://www.fda.gov/aboutfda/whatwedo/default.htm (accessed
05/26/2011).
———. 2011a. 510(k) and science report recommendations: Summary and overview of com-
ments and next steps. Silver Spring, MD: Food and Drug Administration.
———. 2011b. CDRH innovation initiative. http://www.fda.gov/AboutFDA/CentersOffices/
CDRH/CDRHInnovation/default.htm (accessed 02/28/2011).
Foote, S. B. 2005. Frontiers of medical technology: Reflections on the intersection of innova-
tion and the health care system. Minnesota Journal of Law, Science, and Technology
7(1):79-88.
GHTF (Global Harmonization Task Force). 2008. Summary technical documentation for
demonstrating conformity to the essential principles of safety and performance of medical
devices (STED) (SG1-N11).
———. 2009. GHTF documents. http://www.ghtf.org/documents/ (accessed 02/07/2011).
———. 2011. Statement from GHTF chair: Update on future directions of GHTF. http://www.
ghtf.org/documents/statement-future-2011.pdf (accessed 04/25/2011).
OCR for page 187
187
EXTERNAL FACTORS
Grubb, A., J. Laing, J. McHale, and I. Kennedy, eds. 2011. Principles of Medical Law. Third
ed. Oxford, UK: Oxford University Press.
Halperin, D., T. Heydt-Benjamin, K. Fu, T. Kohno, and W. H. Maisel. 2008. Security and
privacy for implantable medical devices. IEEE Pervasive Computing 7(1):30-39.
IOM (Institute of Medicine). 2010. Public health effectiveness of the FDA 510(k) clearance
process: Balancing patient safety and innovation. Washington, DC: The National Acad-
emies Press.
———. 2011. Public health effectiveness of the FDA 510(k) clearance process: Measuring
postmarket performance and other select topics. Washington, DC: The National Acad-
emies Press.
ITA (International Trade Administration). 2010. Medical device industry assessment updated
March 24, 2010. Washington, DC: United States Department of Commerce.
Jorgensen, M. 2010. Identification of more risks can lead to increased over-optimism of and
over-confidence in software development effort estimates. Information and Software
Technology 52(5):506-516.
Kalorama Information. 2010. The global market for medical devices. Market Research Group,
LLC.
Kaplan, A. V., D. S. Baim, J. J. Smith, D. A. Feigal, M. Simons, D. Jefferys, T. J. Fogarty, R. E.
Kuntz, and M. B. Leon. 2004. Medical device development: From prototype to regulatory
approval. Circulation 109(25):3068-3072.
Lee, S., B. Ransford, K. Fu, T. Kohno, and W. H. Maisel. 2008. Electromagnetic interference
(EMI) of implanted cardiac devices by MP3 player headphones. Circulation 118:S596.
Linehan, J. H., and A. Chaney. 2010. Academic/industry challenges for medical device develop-
ment. Science Translational Medicine 2(63):63mr6.
Long, D. E. 2008. Crossing the innovation divide. Temple Law Review 81(2):507-554.
Makower, J., A. Meer, and L. Denend. 2010. FDA impact on U.S. medical technology innova-
tion: A survey of over 200 medical technology companies. Mountain View, CA.
McCulloch, P., D. G. Altman, W. B. Campbell, D. R. Flum, P. Glasziou, J. C. Marshall, J.
Nicholl, J. K. Aronson, J. S. Barkun, J. M. Blazeby, I. C. Boutron, P. A. Clavien, J. A.
Cook, P. L. Ergina, L. S. Feldman, G. J. Maddern, B. C. Reeves, C. M. Seiler, S. M.
Strasberg, J. L. Meakins, D. Ashby, N. Black, J. Bunker, M. Burton, M. Campbell, K.
Chalkidou, I. Chalmers, M. de Leval, J. Deeks, A. Grant, M. Gray, R. Greenhalgh, M.
Jenicek, S. Kehoe, R. Lilford, P. Littlejohns, Y. Loke, R. Madhock, K. McPherson, J.
Meakins, P. Rothwell, B. Summerskill, D. Taggart, P. Tekkis, M. Thompson, T. Treasure,
U. Trohler, and J. Vandenbroucke. 2009. No surgical innovation without evaluation: The
ideal recommendations. Lancet 374(9695):1105-1112.
MDMA (Medical Device Manufacturers Association) and NVCA (National Venture Capital
Association). 2009. Medical technology and venture capital: A fruitful yet fragile ecosys-
tem. Washington, DC
Merriam Webster. 2011. http://www.merriam-webster.com/dictionary/innovation (accessed
05/26/2011).
NANDO (New Approach Notified and Designated Organisations). 2011. European Com-
mission: Enterprise and industry: NANDO (new approach notified and designated
organisations) information system. http://ec.europa.eu/enterprise/newapproach/nando/
(accessed 02/07/2011).
NRC (National Research Council). 1999. Trust in cyberspace. Washington, DC: National
Academy Press.
Oracle. 2010. Oracle solaris patching strategy and best practices. Presented at Oracle Open
World. September 29, 2010
OTA (Office of Technology Assessment). 1984. Federal policies and the medical devices indus-
try. Washington, DC: Congress of the United States, Office of Technology Assessment.
OCR for page 188
188 MEDICAL DEVICES AND THE PUBLIC’S HEALTH
Oxford Dictionaries. 2011. http://oxforddictionaries.com/view/entry/m_en_us1258230#m_en_
us1258230 (accessed 05/26/2011).
Pfleeger, S. L. 2005. Soup or art? The role of evidential force in empirical software engineering.
IEEE Software January/February 66-73.
Pfleeger, S. L., L. Hatton, and C. Howell. 2002. Solid software. Upper Saddle River, NJ:
Prentice Hall.
Platzer, M. 2006. Patient capital: How venture capital investment drives revolutionary medical
innovation. Washington, DC: National Venture Capital Association.
Privitra, M. B., M. Design, and J. Johnson. 2009. Interconnections of basic science research
and product development in medical device design. Presented at 31st Annual Conference
of the IEEE EMBS, Minneapolis, MN.
Rogers, E. M. 2003. Diffusion of innovations. New York: Simon & Schuster, Inc.
Schum, D. 1994. E vidential foundations of probabilistic reasoning . New York:
Wiley-Interscience.
Shuren, J. 2011. On the impact of medical device regulation on jobs and patients. Testimony
before the Subcommittee on Health, Committee on Energy and Commerce, House of
Representatives. February 17, 2011. Washington, DC
Teece, D. J. 1986. Profiting from technological innovation: Implications for integration, col-
laboration, licensing, and public policy. Research Policy 15:285-305.
