This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A COMPREHENSIVE APPROACH FOR MEDICAL DEVICE DEVELOPMENT:
INCORPORATING REGULATIONS, CRITICAL FACTORS AND DESIGN FOR
X IN MODELING A CONCEPTUAL FRAMEWORK
A Dissertation in
ii
The dissertation of Lourdes A. Medina was reviewed and approved* by
the following:
Richard A. Wysk
Dissertation Co-Advisor
Dissertation Co-Advisor
Irene Petrick
Paul Griffin
Peter & Angela Dal Pezzo Department Head Chair of the Harold
and Inge Marcus
Department of Industrial and Manufacturing Engineering
*Signatures are on file in the Graduate School
iii
Abstract
This thesis presents the development of a model-based framework
that can be used to
manage the complexities of medical device development (MDD). MDD is
considered
complex given the (1) wide range of technologies with diverse
intended uses, (2) multiple
stakeholders with conflicting objectives, and (3) development
environment impacting the
development process. However, big part of this complexity is
attributed to the highly-
regulated nature of these products. Medical devices are regulated
worldwide by a wide
range of government agencies, with the Unites States recognized as
the global leader. The
Food and Drug Administration (FDA) regulates medical devices to
assure that these
products are safe and effective before their release into the U.S.
market. Therefore, FDA
is identified as the first factor influencing the industry
decisions and priorities for MDD.
In general, published work in this area either addresses the MDD
with a specific
focus on regulation or provides proposals for various MDD
approaches; however, these
works are fragmented in that they fail to consider the relationship
between the two foci.
This research proposes the use of a model-based framework along
with Design for X
(DfX) concepts to develop a stronger relationship between the
development process and
the regulations. A model-based framework provides a comprehensive
view on the
different elements that should be considered for MDD process. DfX
methods are
recommended to drive the development process with a focus on
important objectives, in
this case – the regulations. The number of studies addressing DfX
for MDD is limited;
however, these methods are relevant to the objectives of developing
safe and effective
iv
medical devices. Accordingly, this research develops a Design for X
(DfX) approach for
MDD that incorporates the FDA regulation for medical devices.
DfX for MDD involves the creation of a product design process
model, definition of
critical factors by hierarchical levels and development of the
Design for FDA (DfFDA)
concept. The product design process model provides the model-based
framework and a
set of formalisms to define the MDD landscape. The model is build
from documentation
analysis and validated by subject matter experts (SME) and a use
case. Critical factors for
MDD are defined in this framework and organized based on their
relevance on different
hierarchical level: organization, development environment, and
development process.
The critical factors identified at the various hierarchical levels
include: (1) organization
level: communication and experience of the development team; (2)
development
environment level: FDA/international regulatory requirements,
customer‘s demand for
cost-saving/cost-effective technology and price-sensitivity, costs
– R&D, clinical
research, litigation, and availability of capital funds, coverage
and reimbursement
requirements, and intellectual property protection issues; and (3)
development process
level: NPD process completeness and proficiency, preliminary market
and financial
analysis, customer involvement, and complexity/technical
challenge.
An additional level is identified, the product level, where an
empirical investigation is
performed to study the impact of various factors in FDA‘s decision
time. This analysis
identifies critical factors at the product level and is the basis
to develop DfFDA. The
critical factors identified include regulatory components, product
characteristics and the
historical reference. Significant regulatory components include the
submission types
(PMA, traditional 510(k), special 510(k) and abbreviated 510(k))
and the different types
v
product characteristics included the factors specific to hip
devices (cemented,
constrained) and generalized factors applicable to most medical
devices (intended use,
context of use, function and material). The importance of
historical reference, as an
indication of various types of experience, showed the significance
of company's
experience with FDA, and FDA's experience in terms of product codes
and product
characteristics (body part, function, material and et
cetera).
The proposition of a product design process model and analysis of
critical factors are
both integrated into DfX for MDD. In this process DfX methods
applicable to MDD and
overlapping with FDA objectives are identified. The DfX methods
identified include:
Design for Validation (DfV), Design for Reliability (DfR), Design
for Quality (DfQ),
Design for Manufacturing (DfM), Design for Assembly (DfA) and
Design for Usability
(DfU). The DfFDA concept is proposed to extend on the existing
concepts in order to
increase awareness about regulatory compliance and promote
designers to consider the
regulations throughout the MDD process. The proposed DfFDA provides
guidelines
based on the analysis of critical factors. The defined guidelines
includes both generalized
and product specific guidelines. Generalized guidelines focuses on
the (1) consideration
of FDA early and throughout the development process, (2) use of
effective
communication with FDA, (3) consideration of human factors, (4)
consideration of
standards and guidance documents, and (5) explosion of FDA
resources while being
attentive to changes. These guidelines are further divided into the
hierarchical
framework, with impact to the organization (1 and 2), development
environment (4 and
5), and development process (3). Product specific guidelines, at
the product level of the
vi
hierarchical framework, include highlights of the importance of
product codes and the
development of a predictive model (R 2 of 40% and 35%) for FDA‘s
decision time.
DfFDA complements the DfX framework for medical devices.
In summary, DfX for MDD is proposed as an aid for designers to
proactively use in
handling the complexities of MDD. DfX for MDD is intended to help
as (1) training
material/best practices, (2) framework for process improvement (to
enhance
completeness and environment considerations such as the
regulations), (3) process
guideline for implementation, and (4) assistance for regulatory
compliance.
vii
1.4 Research Objectives
................................................................................................
10
2.1 Overview
.................................................................................................................
14
2.2.1 Industry
.............................................................................................................
15
2.3.2 Review of the FDA Regulation
........................................................................
23
2.3.2.1 Device Classifications
...................................................................................
24
2.3.2.3 QS Regulation and CGMP
............................................................................
36
viii
2.4.1 MDD Process Models
.......................................................................................
39
2.4.2 DfX
...................................................................................................................
54
3.3.2 Analysis of Critical Factors for MDD
..............................................................
79
3.3.3 Development of a Design for FDA concept for MDD
..................................... 83
3.4 Summary
.................................................................................................................
85
4.1 Overview
.................................................................................................................
86
4.3 Model Structure and Use
.........................................................................................
95
4.4 Preliminary Case Study
.........................................................................................
105
4.5 Summary
...............................................................................................................
112
5.1 Overview
...............................................................................................................
113
5.2 Background
...........................................................................................................
114
6.1 Overview
...............................................................................................................
139
6.2.1 Critical Factors in the MDD Process and Environment
................................. 140
6.2.2 Critical Factors for MDD by Hierarchical Levels
.......................................... 144
6.2.3 Analysis of Product Design Level
..................................................................
146
6.3 Hypotheses and Methods
......................................................................................
147
6.4 Results
...................................................................................................................
157
Chapter 7: Design for Food and Drug Administration - DfFDA
.................................... 178
7.1 Overview
...............................................................................................................
178
7.2.2 Design for Reliability
.....................................................................................
182
7.2.3 Design for Validation
.....................................................................................
183
7.2.4 Design for Quality
..........................................................................................
184
7.2.5 Design for
Usability........................................................................................
185
7.4 Summary
...............................................................................................................
214
8.1 Overview
...............................................................................................................
215
8.4.2 Complexity of the Models – Software Tool
................................................... 224
8.4.3 Challenges to Enhance the Model Use – Case Studies
.................................. 225
Acronym Glossary
..........................................................................................................
226
Appendix 4.2: Content validation – IRB, consent form and survey
........................... 266
Appendix 5.1: Case study – IRB, consent form and model assessment
questionnaires
.....................................................................................................................................
279
Appendix 5.3: Market
analysis....................................................................................
290
Appendix 5.5: Medical device records
........................................................................
293
Appendix 6.1: Frequency tables
..................................................................................
296
Appendix 6.2: Assumptions verification
.....................................................................
303
Appendix 7.1: Estimates for full model
......................................................................
305
Appendix 7.2: Estimates for PC (split) reduced model
............................................... 315
Appendix 7.3: Estimates for RN reduced
model.........................................................
317
xi
List of Tables
Table 2.1: Examples of class I, II and III devices
......................................................... 25
Table 2.2: FDA product classification examples of exempt medical
devices .............. 28
Table 2.3: FDA product classification examples of medical devices
requiring 510(k)
clearance
........................................................................................................................
30
Table 2.4: FDA product classification examples of medical devices
requiring PMA
approval
.........................................................................................................................
31
Table 2.5: Types of PMA supplements (FDA, 2008a)
................................................. 33
Table 2.6: Summary of reporting requirements for user facilities
(FDA, 1997b) ........ 38
Table 2.7: Summary of reporting requirements for manufacturers
(FDA, 1997b) ....... 39
Table 2.8: Summary of literature survey on proposed steps for
engineering design
process (Ogot and Okudan-Kremer, 2004)
...................................................................
41
Table 2.9. Summary of comparison of MDD processes in the literature
...................... 42
Table 2.10: Comparison of MDD processes in the literature
........................................ 43
Table 2.11: MDD examples
..........................................................................................
48
Table 2.12: Phases 0 and 1 for the MDD
......................................................................
49
Table 2.13: Phases 2 and 3 for the MDD
......................................................................
50
Table 2.14: Phases 4 and 5 for the MDD
......................................................................
51
Table 2.15: Summary of Ernst (2002) identification of success
factors of NPD .......... 58
Table 2.16: Summary of MDD empirical research
....................................................... 59
Table 2.17: Summary of internal factors in MDD
........................................................ 63
Table 2.18: Summary of external factors in
MDD........................................................
65
Table 4.1: Model relationships
......................................................................................
88
Table 4.2: Document analysis results example for FDA-related
information. ............. 89
Table 4.3: Document analysis results example for medical device
records.................. 90
Table 4.4: Model clusters
..............................................................................................
93
Table 4.5: Content validation results by cluster
............................................................
94
xii
Table 4.6: Content validation results for product development and
introduction
processes........................................................................................................................
95
Table 4.8: Use of standards for MRA device classification
........................................ 110
Table 5.1: Model implementation examples
...............................................................
119
Table 5.2: Mid-project (M) / final-project (F) assessments - SUS
results .................. 124
Table 5.3: Mid-project (M) / final-project (F) assessments-model
use results part I .. 127
Table 5.4: Mid-project (M) / final-project (F) assessments-model
use results part II 129
Table 5.5: Model implementation comments from users
............................................ 130
Table 6.1: Data sets for analysis of critical factors
..................................................... 148
Table 6.2: List of variables
..........................................................................................
149
Table 6.3: Analysis of continuous variables for the preliminary (P)
data set ............. 160
Table 6.4: Analysis of continuous variables for the complete (C)
data set ................. 160
Table 6.5: Frequency table for PC * RC
.....................................................................
162
Table 6.6: Output for ANOVA with PC – preliminary (P) data set
............................ 163
Table 6.7: Summary of variables for mixed ANCOVA
............................................. 165
Table 6.8: Mixed ANCOVA results - P and C data sets
............................................. 167
Table 6.9: Mixed ANCOVA solution for continuous fixed effects-P / C
data sets .... 173
Table 7.1: List of variables
..........................................................................................
199
Table 7.2: Models 1-9
.................................................................................................
202
Table 7.3: Full model (model
9)..................................................................................
205
Table 7.6: Reduced model – PC
(Split).......................................................................
210
Table 7.8: Reduced model – RN
.................................................................................
213
Table 7.9: Statistics for reduced model – RN
.............................................................
213
xiii
Figure 1.2: MDD stakeholders
........................................................................................
3
Figure 1.3: Relationship between regulations, standards and
guidance documents –
Modified from Alexander and Clarkson (2000)
..............................................................
5
Figure 1.4: Research
questions......................................................................................
13
Figure 2.1: Distribution of medical technology companies by size
for 2001 - Modified
from AdvaMed (2004)
..................................................................................................
16
Figure 2.2: Number of medical device patents (1989-2003) – Modified
from AdvaMed
(2004)
............................................................................................................................
18
Figure 2.3: Recognized standard organizations by FDA (FDA, 2011a)
....................... 20
Figure 2.4: Relationship between the reviewed literature and good
design practice –
Modified from Alexander and Clarkson (2000)
............................................................
56
Figure 2.5: Top ten factors affecting companies' ability to develop
new medical
technologies – Modified from AdvaMed (2003)
.......................................................... 65
Figure 2.6: Top ten factors influencing companies' product
development priorities –
Modified from AdvaMed (2003)
..................................................................................
65
Figure 3.1: Global markets for 2000 - Modified from AdvaMed (2004)
..................... 70
Figure 3.2: Market speed comparison - Modified from AdvaMed (2004)
................... 70
Figure 3.3: Medical device specialization
branches......................................................
71
Figure 3.4: Methodology for the development of a DfX approach for
MDD ............... 73
Figure 3.5: Methodology for the development of a product design
process model for
MDD..............................................................................................................................
75
Figure 3.6: Case study methodology
.............................................................................
78
Figure 3.7: Methodology for the analysis of critical factors for
MDD ......................... 80
Figure 3.8: Methodology for the development of a Design for FDA
(DfFDA) concept
for MDD
........................................................................................................................
84
Figure 4.2: Model example for medical device records
................................................ 90
Figure 4.3: First model iterations
..................................................................................
91
Figure 4.4: FDA‘s and medical specialties clusters
...................................................... 97
Figure 4.5: Standards cluster
.........................................................................................
97
Figure 4.6: Patents cluster
.............................................................................................
99
Figure 4.7: Development/Introduction cluster
..............................................................
99
Figure 4.8: Development/Introduction cluster – concepts of the
product definition
process
.........................................................................................................................
100
planning
processes.......................................................................................................
102
process
.........................................................................................................................
102
Figure 4.12: Development/Introduction cluster – medical device
records ................. 102
Figure 4.13: Final product design process model for medical devices
....................... 103
Figure 4.14: Scenario - FDA‘s and medical specialties‘ clusters
............................... 106
Figure 4.15: Scenario – Development/Introduction cluster – concepts
of the product
definition process
........................................................................................................
111
Figure 5.1: Laparoendoscopic single-site surgery example -
transabdominal magnetic
anchoring and guidance system (MAGS) (Medscape, 2008; Park et al.,
2007) ......... 114
Figure 5.2: Laparoscopic surgical instrument (Learning Factory,
2011) .................... 115
Figure 5.3: Weekly MDD progress
.............................................................................
118
Figure 5.4: A comparison of the adjective ratings, acceptability
scores, and school
grading scales, in relation to the average SUS score – Adopted from
Bangor et al.
(2009)
..........................................................................................................................
124
Figure 6.1: Critical factors in the MDD process and environment
............................. 143
Figure 6.2: Hierarchical framework for the analysis of critical
factors for MDD ...... 145
Figure 6.3: Research methods to identify significant factors
...................................... 157
Figure 6.4: Histograms of DT and Log(DT)
...............................................................
158
xv
Figure 6.7: Box-plot of Log(DT) for each PC
............................................................
163
Figure 6.8: Box-plot of Log(DT) for each ST
.............................................................
163
Figure 6.9: Critical factors for MDD – product level
................................................. 175
Figure 7.1: Design for X for medical devices
.............................................................
180
Figure 7.2: DfFDA for MDD
......................................................................................
188
Figure 7.3: Interaction of human factors considerations – Modified
from FDA (2000)
.....................................................................................................................................
191
Modified from (Alexander and Clarkson, 2000)
......................................................... 192
Figure 7.5: Box-plot of Log(DT) for each PC
............................................................
194
Figure 7.6: PCs example part 1
...................................................................................
195
Figure 7.7: PCs example part 2
...................................................................................
196
Figure 7.8: PCs example part 3
...................................................................................
196
Figure 7.9: Relationship of PC with FDA classification and product
characteristics . 197
Figure 7.10: ASE plot for full model (model 9)
..........................................................
206
Figure 7.11: ASE plot for reduced model – PC (Split)
............................................... 212
Figure 7.12: ASE plot for reduced model – RN
..........................................................
213
xvi
Acknowledgements
The completion of this dissertation, along with all the
requirements of this degree, has
been a discovery process of challenges that has helped me grow
professionally and
personally. These accomplishments would not have been possible
without the support,
patience and guidance of exceptional people that have helped me in
this journey.
My deepest gratitude is to my advisors, Dr. Gül E. Okudan-Kremer
and Dr. Richard
A. Wysk. I thank Dr. Gül E. Okudan-Kremer who for the last two
years has supported my
work with enthusiasm and has inspired and motivated me in many
aspects of my work.
Her commitment, expertise and passion have been of great
contribution to this work and
to my career as a whole. I thank Dr. Richard A. Wysk who supported
my development
from early on and continued motivating me despite the distance and
his many other
academic and professional commitments. His wisdom, knowledge and
dedication have
been key factors in my development and in the completion of this
work. I also thank my
committee members, Dr. Andris Freivalds and Dr. Irene Petrick,
whose valuable
comments made a difference in this work.
This work would not have been possible without the economic support
of various
fellowships: Alfred P. Sloan, GEM and NASA Harriet Jenkins
Pre-doctoral Fellowship
Program. I also thank my friends, family and in-laws, to only some
of whom it is possible
to give particular mention – Anita, Zory, Ramon, Canchy, Richard,
Mariita, Luis, Maneli,
Evelyn, Raul, Robert, Maria, Wilmarie, Aixa, Davids and those to
whom I include in the
dedication of this work. Thanks for supporting me throughout this
journey and being
there for me.
xvii
Dedication
I dedicate this dissertation to God, my husband, parents and
grandparents.
To God, who is the master of my life and accomplishments.
To my husband, Josue R. Crespo, my accomplice in this journey, who
left everything
to follow my dream. Thanks to your support, understanding,
encouragement and
unconditional love – we accomplished our goal. I love you!
To my parents, Pedro L. Medina and Ana A. Aviles, for following the
steps of my
grandparents and encouraging me to always reach higher. Thanks for
your commitment
and support to my education. Thanks for being there when I needed
it the most.
To my grandparents, who have dedicated their life to support the
education of
multiple family generations. Thanks for keeping me in your daily
prayers for all these
years, and for teaching me the importance of having God as the
pillar of all my
accomplishments. (A mis abuelos, quienes han dedicado su vida en
apoyar la educación
de varias generaciones. Gracias por mantenerme en sus oraciones por
todos estos años, y
por enseñarme la importancia de tener a Dios como el pilar de todos
mis logros).
1
1.1 Overview
This chapter provides an introduction to medical device development
(MDD) (See the
Acronym Glossary at the end of this dissertation for a list of
acronyms) with a description
of the challenges, motivations and potential contributions of this
research. Section 1.1
presents background information to explain the complexities of MDD.
Section 1.2
discusses the motivation for this research and research questions.
Section 1.3 describes
the research objectives, and provides an overview on the research
strategy employed.
Finally, Section 1.4 summarizes the chapter.
1.2 Background
This thesis presents the development of a model-based framework
that can be used to
manage the complexities of medical device development (MDD). MDD is
considered
complex given the (1) wide range of technologies, (2) multiple
stakeholders with
conflicting objectives, and (3) development environment impacting
the MDD process.
A device is defined as an object, machine, or piece of equipment
that has been made
for some special purpose (Merriam-Webster, 2011). However, the
definition of a
medical device is more complex. A medical device is defined by FDA
as an an
2
instrument, apparatus, implement, machine, contrivance, implant, in
vitro reagent, or
other similar or related article, including a component part, or
accessory which is: 1)
recognized in the official National Formulary, or the United States
Pharmacopoeia, or
any supplement to them; 2) intended for use in the diagnosis of
disease or other
conditions, or in the cure, mitigation, treatment, or prevention of
disease, in man or other
animals; or 3) intended to affect the structure or any function of
the body of man or other
animals, and which does not achieve any of its primary intended
purposes through
chemical action within or on the body of man or other animals and
which is not
dependent upon being metabolized for the achievement of any of its
primary intended
purposes" (FDA, 2009a).
In addition to its complex definition, medical devices differ from
each other in a
variety of aspects. Medical devices vary in terms of the medicine
disciplines (See Figure
1.1), technological complexity, intended use, context of use and
the type of medical
condition (Shah et al., 2009). Medical devices include a variety of
technologies with
applications ranging from simple therapeutic (e.g., scalpels,
band-aids) to highly complex
(e.g., deep-brain stimulators, implantable defibrillators) (Linehan
et al., 2007).
Figure 1.1: Medicine disciplines for medical devices
General Medical
General/Plastic
Surgery
Obstetrics &
Gynecology
Ophthalmic
Gastroenterology
& Urology
Ear/Nose/Throat
(ENT)
MEDICAL
DEVICES
3
MDD is also a complex problem from the perspective of its
multi-stakeholders and
their conflicting objectives. Stakeholders for medical devices
include the government,
regulatory agencies, insurance companies, patients, physicians,
healthcare professionals,
hospitals‘ administration, industry management, developers and
manufacturers. Figure
1.2 summarizes the MDD stakeholders with examples of the industry
for orthopedic
devices and the government/regulations for the United States. While
the Food and Drug
Administration (FDA) is the part of the U.S. government that
regulates medical devices,
its mission is not to take care of the government‘s interests, but
to protect and promote
the public health of end-users impacted by FDA-regulated products.
The importance of
the FDA is evident; however, prior studies did not explicitly
address the impact of the
regulations on MDD. Therefore, this is a motivation for this
research.
Figure 1.2: MDD stakeholders
As patient care increasingly relies on medical devices (Maisel,
2004) with tens of
thousands of human lives continually depend on the proper
performance and reliability of
medical devices (Bell, 1995); patients‘ and physicians‘ major
interest is to assure device
safety and reliability while using the latest technological
findings to prolong life (Maisel,
4
2005). Meanwhile, cost plays a significant role in the patients‘
ability to benefit from
these technologies along with the problems associated with
insurance coverage. One of
the government‘s major interests is to reduce the cost of
healthcare. From the industry
perspective, R&D seeks practical solutions to clinical problems
that would result in more:
safe, effective, reliable, repeatable, scalable, and profitable
technologies. From a business
standpoint, technologies should rest in unique and patentable
solutions with enough
revenue to overcome the costs (Kucklick, 2006).
Furthermore, this shows that there are multiple and often competing
perspectives
among the different stakeholders impacted by the development of
medical devices. There
are tradeoffs across multiple aspects such as cost, safety,
effectiveness and innovation.
However, there should be no compromise on human life, emphasizing
the importance of
the FDA that attempts to make the best decision from a risk
assessment perspective.
Surrounding MDD and the diverse stakeholders, medical devices are
developed in a
complex environment of intellectual property (IP) protections,
regulations, standards,
guidance documents and other legal issues. Figure 1.3 illustrates
the importance of the
medical device environment. The figure shows the relationship
between the regulations,
standards and guidance documents. Regulation and the consideration
of other legal issues
are imperative to assure a well-regulated practice, while
decreasing the chances of
lawsuits. The use of standards is recommended to facilitate the
compliance with
regulations. IP protections in the form of patents are recommended
to deter others from
pursuing of similar inventions (USPTO, 2010). Guidance documents
are defined as
optional as they assist in understanding the regulations.
5
Modified from Alexander and Clarkson (2000)
While the complexities of MDD are related to the products,
stakeholders and
development environment, a big part of this complexity is
attributed to the highly-
regulated nature of these products. Medical devices are regulated
worldwide by a wide
range of government agencies, with the Unites States recognized as
the global leader. The
FDA regulates medical devices to assure that these products are
safe and effective before
their release into the U.S. market. The following section describes
the research
motivations, where the regulations play an important role.
1.3 Motivations and Research Questions
New products are developed to satisfy consumer‘s needs, respond to
competition and
replace other products with better solutions. Companies also pursue
new product
development (NPD) to obtain competitive advantage through
development of new
technologies and/or creation of new markets. NPD is important for
companies to stay
competitive and grow. However, along with the benefits of
introducing new products
there is a risk of its failure; therefore conducting an effective
NPD process is essential. In
OPTIONAL
IMPERATIVE
RECOMMENDED
Regulations
Guidance Documents
regulation.
6
MDD the problem becomes more complex due to the role of these
products in the
healthcare system and their direct impact on quality of life.
Empirical research has
demonstrated the importance of completeness in the MDD process,
showing a significant
correlation between the number of stages followed in the
development process and the
success of the new devices (Rochford and Rudelius, 1997).
While descriptions of the MDD process exist in the literature,
current models lack in
the comprehensiveness needed to support medical device design
teams. In general,
published work in this area either addresses the MDD with a
specific focus on regulations
(Pietzsch et al., 2007; Maisel, 2004; Hamrell, 2006) or provides
proposals for various
MDD approaches (Alexander and Clarkson, 2002; Pietzsch et al,.
2009; Aitchison et al.,
2009); however, these works are fragmented in that they fail to
consider the relationship
between the two foci.
MDD is a complex activity that requires multidisciplinary teams to
work together for
multiple years. The development team‘s experience (Lucke et al.,
2009), effective
communication of priorities (Brown et al., 2008) and the
requirements of comprehensive
clinical trials (Reed et al., 2008) are some of the factors that
impact the development time
of medical devices. Devices that require clinical trials are
developed in approximately six
years, including: (1) three years for preclinical development, (2)
two years for clinical
testing, and (1) one year for reviews from FDA and the Centers for
Medicare and
Medicaid (Reed et al., 2008). In contrast, the development time is
between one to three
years when clinical trials are not performed for medical devices
under the European
regulation (Cookson and Hutton, 2003).
7
Given the importance of following an effective and complete MDD
process along
with the lack of comprehensiveness of existing models, the
following research questions
are addressed:
What is the MDD landscape? How can the process be
characterized?
Conducting an effective MDD process is also essential from a
business perspective
since by default, a new medical device can be construed as a risky
investment
opportunity with a significant likelihood of failure due to
stringent regulatory processes
and potential litigations. An important part of securing return on
such investments is to
satisfy customer and company needs, however, unlike other
manufacturing markets MDD
is also performed to satisfy severe regulations that provide
additional constraints for the
development, manufacture, marketing and continuous improvement of
medical devices.
Therefore, FDA is recognized as an inhibitory factor for discovery
(Foote, 1996).
FDA approval is a significant milestone for industry developers who
have to
overcome it for the commercial release of their devices in the
market. Complete
knowledge of the FDA requirements and terminology is crucial to
appropriately
demonstrate the safety and effectiveness of medical devices.
Therefore, the FDA has
been reported as the primary external factor affecting a company's
ability to develop new
medical technology and influencing a company‘s product development
priorities
(AdvaMed, 2003). For these reasons it is recommended to consider
FDA regulation
components at every stage of the development process (Pietzsch et
al., 2009).
The literature has paid attention, in particular, to the FDA
approval process, due to
the large number of stakeholders and processes involved in MDD. The
FDA regulation
impacts all the stakeholders of MDD due to their absolute power
over the specification of
8
requirements for medical devices. The literature includes general
reviews of the approval
process (Pietzsch et al., 2007; Maisel, 2004) and the evaluation of
specific stages, such as
the FDA‘s role with clinical studies (Saviola, 2004) and the FDA
regulation of medical
devices with automated control systems (Ciakowski, 2000).
Meanwhile, some authors
relate the FDA regulation to specific issues pertaining to a
product type (Maisel, 2005).
Overall, the literature suggests that assuring the FDA approval is
one of the major
elements in the design of medical devices. Accordingly, FDA is one
of the key drivers for
this research which provides the motivation to formulate the
following research
questions:
What is the impact of FDA on MDD? What are the key guidelines for
Design for
FDA (DfFDA)?
These questions are aligned with FDA‘s objective to increase the
regulatory science
for MDD. Jeffrey Shuren, Director of the Center for Devices and
Radiological Health
(CDRH) in FDA explained, at the 2011 Design of Medical Devices
Conference, that
investment in regulatory science has been limited and that FDA‘s
current efforts include
strengthening the U.S. research infrastructure and promoting
high-quality regulatory
science through (1) partnerships between the FDA, industry and
academia, and (2)
curriculum, training, and hands-on experience (Shuren, 2011).
This research intends to investigate and provide a deeper
understanding on the FDA‘s
role in relation to medical devices, specifically for
considerations of the data FDA reports
to the public domain as part of the 510(k) clearances and PMA
approvals. Scientific
treatment of such considerations has been a void in the literature
prior to the work
presented here.
9
In addition to studying the FDA as a critical factor for MDD, a
complete framework
of critical factors for success in MDD must be defined. It is
generally recognized that
effective execution of the NPD process affects the likelihood of
success for new products
in the market; however, success is not primarily a result of just
developing a good
product, but a combination of various factors. Many researchers
have conducted
conceptual and empirical research in an attempt to define NPD
success factors (Ernst,
2002). Most studies are survey based, and success is the result of
rating how the NPD
process is perceived in general or focused on specific factors,
e.g. achieving the expected
profit, sales, and deadlines. The results from these studies show
that a variety of factors
impact the success of NPD, which include characteristics of the NPD
process,
organization, culture, senior management and strategy.
The majority of the literature addresses critical factors in the
context of NPD in
general, but not for any specific applications. However, this study
identifies the critical
factors for MDD in particular to illustrate the impact of such
regulations on medical
device products. Accordingly, this research addresses the following
research question:
What are the critical factors for MDD?
This research proposes a Design for X (DfX) framework to execute
the MDD process
and subsequently, addresses the importance of the regulations in
combination with other
critical factors for MDD. DfX methods aid the design process to
focus on important
objectives (Alexander and Clarkson, 2000). The number of studies
addressing DfX for
MDD is limited; however, these methods are relevant to the
objectives of developing safe
and effective medical devices, together with following Good
Manufacturing Practices
(GMPs). A DfX framework provides a generalized view on the
different elements that
10
should be considered in the product design and development process.
Depending on the
product scope, the design and development could be focused on
several fundamentals;
including alternating the design based on the following objectives:
manufacture,
assembly, variety, quality, reliability, disassembly,
maintainability and obsolescence
(Chiu and Okudan, 2010). With these motivations, this research
addresses the following
research questions:
What is the relevance of DfX methods for MDD? How are the DfX
methods
related? Where is the gap?
1.4 Research Objectives
The principal objective of this research is to develop Design for X
(DfX) guidelines
for MDD to support medical device companies, new designers and
experienced
developers entering this field. Therefore, this work involves the
following activities:
Develop a standard product design process model for MDD through
a
documentation analysis, content validation with subject matter
experts (SMEs),
and implementation case study in the development of a medical
device. The
product design process model explains the whole context of the MDD
process and
environment.
Identify critical factors for MDD through the evaluation of studies
in the
literature, organization of factors in a hierarchical framework,
and completion of
this framework with a quantitative analysis at the product level.
The analysis of
critical factors for MDD shows the impact of the process on the
success of
products and in particular the implications of specific factors in
FDA‘s decision
time of submissions.
11
Develop the Design for FDA (DfFDA) concept through the
identification of
relevant DfX methods, the definition of guidelines from the
analysis of critical
factors and the development of a predictive model for FDA‘s
decision time.
Relevant DfX methods for MDD are identified from the literature
addressing DfX
for medical devices and the evaluation of their applicability. DfX
methods are
identified given that these are applicable tools that could be
employed to improve
MDD. DfFDA is developed to increase awareness about regulatory
compliance
and to complete the DfX framework for MDD.
Among these objectives, FDA‘s decision time of submissions plays a
significant role
in this research. FDA's decision time is the (1) measure of
significance in the
identification of critical factors, and the (2) dependent variable
in the development of a
predictive model for DfFDA. This measure is appropriate based on
the implications of
FDA in MDD. FDA has absolute power over the marketing of medical
devices in the US.
Consequently, FDA's decision time is an important component in the
time to market of
medical devices. However, this metric was also selected due to data
availability given
that this information is publicly available in FDA reports.
DfX for MDD contributes to numerous aspects of the design process.
It can be used
as a training material for best practices of MDD. It can be used as
a reference tool that
medical device companies with novice designers and/or experienced
designers who are
new to MDD can follow to gain proficiency in the field. Secondly,
DfX for MDD serves
as a framework for enhancing a holistic view of the MDD process,
allowing current
development processes to be examined for completeness and inclusion
of important
considerations such as documentation, regulations, etc. Further,
this work provides the
12
process guidelines for implementing MDD; it offers a template that
designers can follow
to handle the complexities of MDD and identify strategies for a
more efficient product
development. Finally, DfX for MDD offers assistance for regulatory
compliance. It has
been developed for the regulatory environment in the U.S. with
specific attention paid to
FDA regulation requirements. However, a similar approach can be
pursued for other
regulatory bodies such as EU, etc. Since it has been shown that FDA
has implications for
international markets within the medical device and diagnostics
industry (AdvaMed,
2004), we can assume that the model is applicable for all MDD
practitioners.
In summary, DfX for MDD enhances the development of products with a
holistic
approach and is intended to increase the chances of success in MDD
given its
comprehensiveness, involvement of critical factors and
consideration of the regulations.
In addition, the results from this study would motivate other
researchers studying the
MDD process to: (1) follow a comprehensive MDD landscape, (2)
address critical factors
by hierarchical levels, (3) reduce the subjectivity of such studies
with quantitative
approaches, (4) expand the literature at the product level, (5) use
DfX methods, and (6)
study DfFDA.
1.5 Summary
The complexity of MDD is evident from the discussion of the types
of products,
development environment, stakeholders and the industry. This
research is intended to aid
in the management of these varied complexities by focusing on the
following four areas:
MDD process, FDA, critical factors for MDD and DfX. The importance
of these areas is
described in detail with support from the literature and definition
of research questions, as
13
summarized in Figure 1.4. These are addressed with the development
of a DfX approach
for MDD that includes a validated product design process model,
critical factors by
hierarchical levels, relevant DfX methods and a DfFDA
concept.
Figure 1.4: Research questions
Chapter 2
Literature Review
2.1 Overview
This chapter reviews the literature on medical device development
(MDD) (See the
Acronym Glossary at the end of this dissertation for a list of
acronyms) related to
presented research questions. The MDD landscape is described with a
discussion of the
development process and environment. For the environment, this
chapter includes a
description of the industry, intellectual property, standards and
regulations. Meanwhile, a
big portion of this research is related to the Food and Drug
Administration (FDA), which
is the regulatory agency of medical devices in the Unites States.
This is supported with an
analysis of the existing literature focused on FDA and a detailed
description of the FDA
approval process. In addition, the MDD process is explained with
the analysis of studies
modeling this process and the discussion of relevant DfX methods.
Critical factors for
both, the MDD process and environment are reviewed along with the
limitations of
current practices in the identification of these factors.
The chapter is organized as follows. Section 2.2 provides a concise
description of the
MDD environment; however, it is Section 2.3 that presents a
detailed description of the
FDA regulation. Section 2.4 explains the MDD process based on
methodologies from the
15
literature, and specific DfX methods. Section 2.4 discusses the
literature on critical
factors for MDD. Finally, Section 2.5 summarizes the review
findings.
2.2 Medical Device Development Environment
The MDD environment can impact the feasibility of products and
their further release
to the market. Device ideas might not be feasible due to the
capabilities of the company,
existing competition or intellectual property protections.
Likewise, the release of a
product to the market is dependent on its compliance with standards
and regulations to
assure that the device is safe and effective.
2.2.1 Industry
In the last decade, the U.S. medical device industry has shown
significant growth,
with 15,000 manufacturing companies and 100,000 individual products
on the market
registered in 2004 (Bren, 2006). A substantial portion of the
industry is listed with the
FDA, as evident from the 10,433 records of companies registered in
the United States as
of 2010. These records include multiple registries of the same
company with multiple
locations.
The Center for Devices and Radiological Health (CDRH) recognizes
the medical
device industry as one of rapid growth, change and increasing
complexity, which add to
the demands of regulatory bodies such as FDA. Interestingly,
AdvaMed‘s (2004) analysis
of a specific section of the medical industry (only companies
categorized with North
American Industry Classification System (NAIC) codes for medical
device industry)
16
showed that the majority of the industry in 2001 consisted of small
companies (Figure
2.1).
Figure 2.1: Distribution of medical technology companies by size
for 2001 - Modified
from AdvaMed (2004)
In summary, an industry composed of many small companies with a
broad variety of
devices is a challenge for FDA, which is expected to control the
design/manufacture of
devices without becoming the limitation for technological
advancement (Bell, 1995). In
addition to having to review many submissions with different
styles, FDA performs
quality audits to assure the maintenance of high standards in
manufacturing. FDA also
keeps track of product malfunctions and in cases performs recalls.
Therefore, a large
number of companies and products make this task a challenge with
cost implications.
These issues are discussed further in Section 2.3, with a detailed
description of the FDA
regulation.
17
2.2.2 Intellectual Property
The use of intellectual property (IP) protections in the form of
patents for medical
devices is highly recommended to deter others from pursuing similar
inventions (USPTO,
2010). A patent is an official government document defined by the
U.S. Patent and
Trademark Office (USPTO) as providing the right to exclude others
from making, using,
offering for sale, selling or importing the invention for a time
period of up to 20 years.
There are three kinds of patents defined by the USPTO, including:
utility, design and
plant patents.
Acquiring a patent is a time consuming and costly process. An
important step
includes performing a patent search to assure that the device or
technology is useful and
original. The patent documents usually include three main
components: (1) drawings, (2)
invention‘s explanation, and (3) claims (Ogot and Okudan-Kremer,
2004). This process
usually involves multiple revisions given that it is difficult to
make it through a patent
application on the first attempt. In addition, patent attorneys are
usually consulted to
revise the document. Fees are paid to the patent office along with
the application
submission. After a patent has been granted, it should be
maintained by paying the
corresponding maintenance fees.
Gene Quinn, Patent Attorney, provides examples of the patent cost
of different
inventions depending on their complexity (Quinn, 2007). The patent
cost ranges from
$5,000 for relatively simple inventions (e.g., electric switches,
paper clips); $10,000 for
intermediate complexity inventions (e.g., simple software, simple
RFID device); to more
than $15,000 for highly complex inventions (e.g., MRI scanners,
telecommunication
networking systems) (Quinn, 2007). Figure 2.2 illustrates the
number of medical device
number of developments is increasing which makes the environment
more competitive
and complex. Accordingly, patents must be considered for MDD and as
an important part
of the MDD landscape.
Figure 2.2: Number of medical device patents (1989-2003) – Modified
from AdvaMed
(2004)
2.2.3 Standards
Even though standards are not compulsory, many institutions have
made these a
necessary requirement. This was initiated by insurance companies,
who required
Underwriters Laboratory (UL) certification to show the safety of
products, resulting in
the current development of approximately 1000 standards (Bell,
1995). A variety of
standards have been developed by independent agencies to help
facilitate and standardize
the MDD process. Consensus standards are developed by committees
worldwide to
4178 4500
19
provide consensus on the recommendations for test methods,
materials, devices, and
procedures (Duncan et al., 2004). Examples of these standards
include:
Test method standards: ASTM F897 - fretting corrosion of
osteosynthesis
plates and screws
Material or specification standards: ASTM F139 - stainless steel
sheet and
strip for surgical implants
Procedure, or guidance standard: ASTM F86 - surface preparation and
making
of metallic surgical implants
Although the use of these standards is voluntary, their use is
highly recommended
with multiple advantages for medical device applications. Some of
these advantages
include: (1) assistance in the FDA approval process to expedite the
review process, (2)
guidance in the testing and making of materials and devices, (3)
providing a summarized
description of the products for sale purposes, (4) simplifying life
and promoting
harmonization between countries and the industry, (5) adding to the
confidence of the
users of the products, and (6) in the case of standard test
methods, to re-produce, verify or
confirm the results by other researchers (Duncan et al.,
2004).
The FDA believes that conformance with recognized consensus
standards can
support a reasonable assurance of safety and/or effectiveness for
many applicable aspects
of medical devices (FDA, 2007). Supporting this, the FDA website
includes a database
for recognized consensus standards that describe the impact of the
standard, e.g. devices
affected, processes affected and extent of recognition (FDA, 2011).
Figure 2.3 includes
the organization of FDA‘s recognized standards.
20
2.3 FDA Regulation
Compliance with the FDA regulations is a necessary requirement to
market any
medical device in the Unites States. Therefore, FDA is an essential
part of the MDD
landscape. This section summarizes the literature addressing FDA
and reviews the FDA
approval process to be used as background for this research. The
review of the FDA
regulation for medical devices is an essential component of this
research. It provides a
detailed understanding of the environment and to what extent FDA
impacts the MDD
process.
21
2.3.1 Summary of the Literature on FDA
A significant part of the medical device literature includes
reviews and summaries of
the FDA regulations with different objectives, which have provided
the basis for
numerous FDA publications. Meanwhile, several businesses have the
sole objective of
providing services of support for regulatory compliance to medical
device companies.
For instance, Quality and Regulatory Associates, LLC is an example
of such
organizations dedicated to consultation on FDA affairs, which has
also published an
overview of the FDA‘s medical device regulation (Syring, 2003).
Summaries of FDA
regulations are commonly generated with the objective of informing
specific stakeholders
such as practicing physicians (Maisel, 2004), manufacturers
(McAllister and Jeswiet,
2003), or medical device engineers/designers (Munzner, 1988). FDA
regulations are also
reviewed in the context of specific facets or particular
applications. Saviola (2005)
looked at the FDA‘s role in clinical studies with human subjects
and retinal visual
prosthetic devices, while Ciarkowski (2000) described the FDA
regulation for medical
devices having automatic control systems. Other relevant topics
involve the legal and
ethical issues of regulation (Malloy 2006); risk management to
anticipate failures,
manufacture safer products, and reduce liability cost (Bartoo,
2003); ensuring the safety
and effectiveness of medical devices through market surveillance
(Feigal et al., 2003;
Bren, 2006); and identifying improvement opportunities for medical
device regulations
(Watson, 1994; Maisel, 2005; Foote, 1996).
Pietzsch et al. (2007) provided one of the most general reviews to
date, with a
summary of several concepts related to medical device regulation
and a discussion of
contemporary issues such as combination products (those that
combine drugs and
22
devices, or biologics and devices) and the increasing efforts
toward harmonizing the
regulations among countries. Several publications have discussed
these issues in great
detail. For instance, Eidenberger (2000) summarized the regulation
of medical devices in
both European and U.S. markets, focusing on agreements for mutual
recognition and
their implications on global harmonization 1 . Hamrell (2006)
highlighted the need for a
more comprehensible review process and a better FDA organization
(multiple centers
may have to review one combination product) for combination
products.
Combination products may incorporate nanotechnology implications.
Paradise et al.
(2009) evaluated FDA regulations for drugs and medical devices that
involve
nanobiotechnology using a historical case study approach. They
concluded that: (1) the
FDA is more reactive than proactive, making changes largely in
response to problems;
(2) continuous creation of new developments involving nano-products
may require the
FDA to make decisions about the applicability of existing
frameworks and/or the need for
a new nanotechnology scheme, calling into question the
appropriateness of the FDA
structure for nano-products; (3) there exist potential FDA-related
problems of
transparency, relevant stakeholder representation, a lack of
financial resources, and the
need for a clear definition of nano; and (4) nanotechnology as an
entity may become a
critical challenge that the FDA will have to contend with at some
point, suggesting the
potential for a new product classification (neither a drug nor a
device).
The incorporation of software in medical devices has become a
popular topic
motivating the review of FDA regulations. Lin and Fan (2009)
concentrated on the issue
of software development for FDA-compliant medical devices,
providing a hybrid
1 Global harmonization is mainly stimulated by the European Union
along with North America, Asia,
Australia and Eastern Europe (Eidenberger, 2000).
23
methodology for the application of a Digital Subtraction
Angiography device.
Sudershana et al. (2007) addressed the software development from
the perspective of
collaborative distributed software projects and the challenges it
presents in an FDA-
regulated environment, which include strict measures of
verification and validation.
McCaffery and Coleman (2007) proposed a software process
improvement model, the
Configuration Management Capacity Model (CMCM), for use in local
and distributed
development scenarios.
While the impact of the regulatory environment on the MDD is
evident, prior work
focused on regulations for specific applications and on the
regulatory environment in
general, while neglecting to address how and at what stage of
product
design/development the regulations should be considered.
2.3.2 Review of the FDA Regulation
While an updated review of the regulations is necessary to consider
their evolution,
this research develops a Design for FDA (DfFDA) concept based on
the analysis of the
regulation components. Accordingly, this section summarizes the FDA
regulation for
medical devices based on the compilation of multiple FDA
publications, FDA (1994) –
FDA (2011b). This review includes the explanation of device
classifications, regulatory
pathways, the Quality Systems (QS) regulations, and post-market
requirements.
The regulation of medical devices is described in terms of General
Controls, Special
Controls and the requirement of Premarket Approval (PMA). General
Controls refer to
the basic requirements that all medical devices should follow,
unless particular exemption
rules apply. These General Controls include: (1) establishment
registration; (2) medical
24
device listing; (3) complying with the Good Manufacturing Practices
(GMP); (4) labeling
regulations; and (5) Premarket Notification (510(k)) (FDA,
2009c).
Establishment registration refers to the registration of companies
with the FDA, while
medical device listing is the registration of devices by the
registered companies (FDA,
2009v). Companies that must be registered include: medical device‘s
manufacturers,
distributors, re-packagers and re-labelers. GMPs provide an overall
tracking and quality
approach to the manufacturing of medical devices (FDA, 1996). In
contrast, the label and
labeling regulation reinforces the objective of having safe and
effective medical devices
by dealing with issues such as misbranding, misleading labels,
label information
specifications, directions for use, prescription device
requirements and special labeling
for particular devices (FDA, 1997a). Meanwhile, the 510(k)
application consists of
making the FDA aware of a plan to market a particular device and
provides the necessary
information to obtain clearance (FDA, 2009h).
Special controls provide additional rules for specific groups of
devices. Examples of
special controls include particular labeling requirements,
mandatory performance
standards and post-market surveillance (FDA, 2009d). Special
labeling requirements and
mandatory performance standards have been developed in the context
of specific device
applications, while post-market surveillance is generalized for the
compliance of all
medical devices with Special Controls.
2.3.2.1 Device Classifications
The FDA classifies medical devices in multiple categories, which
are the framework
for the regulation process. Medical devices are also divided
between medical specialties.
25
This classification provides a logical structure for the assignment
of panels during the
revision process. As a result, 1700 different generic types of
medical devices have been
clustered into 16 medical specialties (FDA, 2010b). A risk
classification (RC) is also
used to demonstrate the level of control necessary to assure the
safety and effectiveness
of medical devices, which consist of three categories (FDA, 2009f;
FDA, 2006b): class I,
II and III. In addition to the risk that a device could present to
its patients/users, the
assignment to a specific classification considers the device‘s
intended use and
indications for use (FDA, 2006b). Table 2.1 show examples of
orthopedic and
cardiovascular devices for these three classifications compiled
from FDA (2010b).
Table 2.1: Examples of class I, II and III devices
These classifications are highly correlated to the controls and
regulatory pathways
followed for FDA approval/clearance. Exemptions to comply with the
510(k) may apply
for class I and class II devices, while class III devices are
always required to perform a
510(k) or PMA submission (See FDA (2010d) for more details on
exemptions). Class I
devices are simpler with minimum risk to users when compared to
Class II and III
devices. These devices should comply with the General Controls
(with and without
exemptions), and are subject to minimum regulations (FDA, 2009c).
Class II devices
represent higher risk and complexity, with Special Controls in
addition to the General
Review Panel Class I Class II Class III
component, cast bone cement
cannula, catheter replacement heart-valve
26
Controls (FDA, 2009d). The devices with the greatest risk and
complexity are classified
as Class III, with General Controls and PMA requirements. A Class
III device supports
or sustains human life or is of substantial importance in
preventing impairment of human
health or presents a potential, unreasonable risk of illness or
injury (FDA, 2009g).
Other classifications of medical devices, within the FDA taxonomy,
include the use
of generic types for product codes (PC) and the regulatory number
(RN). The generic
types are defined as a grouping of devices that do not differ
significantly in purpose,
design, materials, energy source, function, or any other feature
related to safety and
effectiveness, and for which similar regulatory controls are
sufficient to provide
reasonable assurance of safety and effectiveness (FDA, 2009g). PCs
are used to identify
the generic categories and serve as a classification method for the
medical device listing.
These codes consist of three letters with no other meaning that the
product characteristics
these represent. Moreover, the letter code does not necessarily
relate to the words used to
describe the products in the group. For instance, examples of PCs
for orthopedic devices
include: LOD for bone cement, HXC for wrenches and JDD for
prostheses upper
femoral.
RNs are used as an association of specific medical device
requirements with the Code
of Federal Regulation (CFR). An RN consists of eight-digit number,
where the first three
digits represent the type of device in terms of the medical
specialization/panel. An RN
example is 888.3030, where 888 refer to the fact that it is an
orthopedic device. The
remaining digits refer to the specific section within the medical
device specialization. The
code includes detailed information for identification of the type
of device and its risk-
based classification. It should be noted that multiple PCs may have
the same RN.
This section discusses the different pathways and requirements that
are applied for
medical devices based on their risk classifications (RCs). Both,
pathways and risk
classifications are important components of the FDA regulation to
be analyzed in this
research through the identification of critical factors and the
development of Design for
FDA (DfFDA).
2.3.2.2.1 Exemptions
Exempt devices are those which do not require a 510(k) submission;
although they
need to comply with other General Controls for their legal
marketing. These are the
minimum requirements any medical device should satisfy for legal
marketing, while
additional limitations for exempt devices may be defined by FDA.
The exempt status
along with its limitations may be found in the PC
information.
The devices that benefit from exemption include pre-amendment
devices and Class
I/II devices that are exempt by regulation. Pre-amendment devices
are defined by FDA
(2009h) as devices legally marketed in the U.S. by a firm before
May 28, 1976 and
which have not been significantly changed or modified since then;
and for which a
regulation requiring a PMA application has not been published by
FDA. As of May
2009, the FDA considers most of the Class I devices as exempt (74%
of all medical
devices with 800 generic types), and only a portion of Class II
devices are exempt (60
generic types). Some Class I devices are exempt from the Good
Manufacturing Practices
(GMP) requirements; however, they have to comply with complaint
files and general
record keeping (FDA, 2009i). Examples of exempt generic devices
compiled from FDA
(2010b) are shown in Table 2.2.
28
Table 2.2: FDA product classification examples of exempt medical
devices
2.3.2.2.2 Premarket Notification
The Premarket Notification (510(k)) objective is consistent with
the FDA‘s role of
assuring safe and effective medical devices, demonstrated through
substantial
equivalence (SE) with other legally marketed devices that do not
require PMA (FDA,
2009h). Applicants are expected to provide SE claims to show that
the device is at least
as safe and effective as a predicate device. Equivalence includes
having the same
intended use with or without the same technological
characteristics. In the case of
different technological characteristics, the submission should
still show the device‘s
safety and effectiveness as comparable to the precedent device,
while minimizing the
potential to doubt the device‘s safety and effectiveness.
The 510(k) submission for a device should be completed at least 90
days prior to its
planned date of market introduction, which applies to: (1) the
first time introduction of a
device into the market, (2) a modification of the intended use of
an approved device, and
(3) a change on the device with potential impact in its safety
and/or effectiveness (FDA,
2009h; FDA 2009j). The areas of the industry subjected to this
format of submission
510 (k) & GMP Exempt
component, cast caliper goniometer with
electrodes
29
include those (1) involved in the device market introduction -
e.g., domestic
manufacturers, specification developers, and/or foreign
manufacturers/exporters, and (2)
having a significant impact in the device - e.g., re-labelers who
make labeling changes.
There are no specified standard forms to complete a 510(k);
therefore, it is the
submitter‘s responsibility to present that information.
In addition to the Traditional 510(k) (with details in FDA
(2008t)), there are Special
510(k) and Abbreviated 510(k) that serve as alternate methods
developed to facilitate the
review process for certain situations. The Special 510(k) should be
used for 510(k)
cleared devices that have undergone some kind of modification that
requires a new
clearance and compliance with the design control provisions of the
Quality Systems (QS)
regulation (FDA, 2009k). The benefit of using a Special 510(k)
submission consists of
not having to provide data about the design controls by providing a
Declaration of
Conformity with the design control provision of the QS regulation.
Reference to the
original 510(k) cleared device is necessary. The modification of
pre-amendment devices
may be cleared with a Special 510(k).
The Abbreviated 510(k) should be used when a guidance document
exits, a special
control has been established or FDA has recognized a relevant
consensus standard
(FDA, 2009l). Abbreviated 510(k) expedites the review process. The
process is facilitated
through a reduction in the documentation work by allowing
manufacturers to provide
summary reports based on the employment of these guidance
documents, special controls
or conformity with standards. Likewise, this process could involve
a release from the
requirements of submitting test data for specific situations.
30
Other methods to accelerate the review process include the use of
reviews from third
parties and expedited review option. The third party review option
allows contracting an
accredited person (authorized by FDA) for eligible devices to
review the 510(k)
submission and forward it to the FDA along with their review and
recommendations. The
FDA has 30 days to review and provide a final decision (FDA,
2009m). The expedited
review means having priority review over other devices in queue,
which may include
devices that treat/diagnose life-threatening or irreversible
disease/condition, or that
addresses an unmet medical need (FDA, 2008b).
The FDA 510(k) clearance for device commercialization consists of a
letter sent by
the FDA confirming the SE and provision of consent for marketing.
Other results from
the FDA review representing no clearance for the device may
include: stating no SE,
requesting additional information, withholding the decision to wait
for a disclosure
statement or informing that the 510(k) was not required. Table 2.3
shows some examples
of generic device types with 510(k) clearance requirements compiled
from FDA (2010b).
Table 2.3: FDA product classification examples of medical devices
requiring 510(k)
clearance
bone cement spacer, cement
degenerative disc disease
FDA Product Classifications
2.3.2.2.3 Premarket Approval
Premarket Approval (PMA) application is submitted to request FDA
approval for
Class III devices (see Table 2.4 for examples compiled from FDA
(2010b)), which
require the strictest regulation due to their high risk nature and
the need for additional
information to prove their safety and effectiveness.
Table 2.4: FDA product classification examples of medical devices
requiring PMA
approval
In theory, the FDA requires 180 days to review and provide a
decision on any PMA
submission. However, the FDA (2009n, 2009e) recognizes that the
process in practice is
much longer as it requires the advisory committee‘s recommendation,
final FDA
decision, and announcement in the Internet, in addition to the
opportunity of
reconsidering petitions within 30 days. Other reasons for delay in
the review process that
may result in denial, in some circumstances, include PMA
submissions that are:
incomplete, inaccurate, inconsistent, missing critical information,
or poorly organized
(FDA, 2009e). Examples of incomplete submissions include missing
elements in the
PMA
analysis. Scientific reports may indicate problems with study
design and conduct, data
analysis, presentation and conclusions. For these reasons, FDA
recommends applicants to
perform a quality control audit of their PMA application before
submission.
Along with a PMA approval, the FDA may require additional
conditions to comply
with, as in the case of post-market studies. For a faster review,
the expedited review
option is mostly used for PMA applications. After an original
(traditional) PMA has been
approved, supplemental PMAs are used for the approval of device
modifications. The
different types of PMA supplements (Table 2.5) include:
panel-track, 180-day, real-time,
special PMA, 30-day notice and manufacturing site change supplement
(FDA, 2008a).
Some Class III devices may be allowed to submit a 510(k) under
certain conditions if:
(1) it is a pre-amendment device with no indication of requiring a
PMA (an effective date
for the PMA requirement should be on file through regulation number
and/or PC
information), (2) the applicant is able to demonstrate substantial
equivalence (SE) with
other Class I/II or 510(k) approved Class III device, or (3) a
device not substantially
equivalent (NSE) to other Class I/II or 510(k) approved Class III,
but that is eligible for
the novo process as a Class I/II device (FDA, 2009h). The novo
process, officially known
as the Evaluation of Automatic Class III Designation provision, is
for low-risk devices
classified as Class III due to their NSE nature (FDA, 1998). The
process allows the re-
classification of these devices as either Class I or II and
subsequently, to submit a 510(k)
application.
33
34
The main element of a PMA submission consists of providing
scientific evidence of
the device‘s safety and effectiveness in terms of its intended use.
Good science and
scientific writing is key to the approval of PMA applications (FDA,
2009n). The
technical sections in the PMA application include data and relevant
information that can
be used as scientific evidence; these are sub-divided into two
sections: the non-clinical
laboratory studies and the clinical investigation. The non-clinical
laboratory studies
should be performed in compliance with the Good Laboratory Practice
for Nonclinical
Laboratory Studies, and may include laboratory/animal tests to
provide information
about microbiology, toxicology, immunology, biocompatibility,
stress, wear and shelf life
(FDA, 2009n; FDA 2009o).
A clinical investigation should include the fulfillment of
institutional review board
(IRB) requirements, whose role is to protect the rights and welfare
of human subjects in
clinical investigations (FDA, 2010c). Some of the information that
should be provided
with regards to clinical investigations include: study protocols,
safety and effectiveness
data, adverse reactions and complications, device failures and
replacements, patient
information, patient complains, data tabulation for all subjects,
and statistical analysis
results (FDA, 2009n; FDA, 2009o). It should be noted that FDA has
device-specific
guidance documents for some devices.
2.3.2.2.4 Other
Other FDA processes include the Humanitarian Device Exemptions
(HDEs), the
Investigational Device Exemptions (IDEs) and the Product
Development Protocols
(PDPs). The HDEs were developed for FDA approval of medical devices
that
treat/diagnose a disease/condition affecting less than 4,000
individuals per year (FDA,
35
2006a). The application process is very similar to the PMA (and PMA
supplements),
while not having to comply with the effectiveness
requirements.
While the HDE is an option FDA provides to facilitate the approval
based on
humanitarian circumstances, the IDE is an additional requirement
that the FDA imposes
on clinical investigations with significant risk to the subjects.
In addition to the
compliance with the IRBs, these investigations should possess FDA
approval through an
IDE application, which includes information about: the
investigation plan, prior
investigation reports, device manufacture, IRB actions,
investigator agreements, subject
informed consent forms, device labeling, and device cost, among
others (FDA, 2010a).
There are also IDE amendments and supplements.
The PDP is an alternate product pathway to the PMA for Class III
devices, wherein
the FDA is directly involved in the device development and testing
plan from early stages
of the process. This process was developed to minimize companies‘
investment of capital
in devices that will not be approved, and to assure early approval
of the devices if it
achieves the expected outcome. Four stages that define the PDP
process are: (1) PDP
summary outline, (2) FDA/Advisory panel review of the full PDP, (3)
consideration/pre-
approval of design modifications and protocol revisions, and (4)
action on the sponsors
Notice of Completion (FDA, 2010a). The sponsor‘s Notice of
Completion is the last
stage where the sponsor defends the application after finishing all
the PDP requirements.
At this stage, the FDA may perform an inspection of compliance with
the QS regulation.
FDA‘s decision declares the PDP as complete or incomplete. A device
that is declared by
FDA as PDP completed is considered to be a PMA approved (FDA,
2009e).
36
The Current Good Manufacturing Practices (CGMP) for medical devices
have
undergone several changes, from subjecting the same regulation for
all FDA-regulated
devices (food, drugs, biologics and devices) to having specifically
defined requirements
for the medical device industry. The CGMP currently enforced forms
a majority of the
QS regulation, with the latest revision in 1996 and effectiveness
in 1997, which consists
of requirements for the methods/facilities/controls of medical
devices‘ designing,
manufacturing, packaging, labeling, storing, installing, and
servicing of medical devices
(FDA, 1996).
The latest revision includes the incorporation of preproduction
design controls and
definition of a regulation which is more cohesive to promote
international harmonization
with quality systems requirements worldwide, which include strong
consideration of
international standards such as the International Organization for
Standards (ISO)
9001:1994 (Quality Systems – Model for Quality Assurance in Design,
Development,
Production, Installation, and Servicing).
The QS regulation is flexible as it does not provide specific
details on how a device
should be manufactured. This is due to the wide variety of medical
devices that exist with
major differences in terms of clinical solution, function, and
specifications. Moreover,
this flexibility represents a major responsibility for medical
device manufacturers who
should follow FDA‘s defined framework to develop their own quality
system and
requirements in order to guarantee the device‘s safety and
effectiveness in the context of
the defined production processes (FDA, 2009p). Manufacturers should
incorporate all
37
relevant components of the QS regulation for the specific product,
operations, and
maintain evidence of their compliance.
The compliance with the QS regulation is a requirement for finished
device
manufacturers planning to introduce devices into the market. A
finished device is defined
by the FDA (2009p) as any device or accessory to any device that is
suitable for use or
capable of functioning, whether or not it is packaged, labeled, or
sterilized, which makes
manufacturers of device accessories also comply with the QS
regulation.
In summary, the QS regulation consists of the following major
components: (1)
quality system requirements (management responsibility, quality
audit and personnel);
(2) design controls; (3) document controls; (4) purchasing
controls; (5) identification and
traceability; (6) production and process controls
(production/process controls, inspection,
measuring/test equipment, and process validation); (7) acceptance
activities (receiving/in-
process/finished device acceptance, and acceptance status); (8)
nonconforming products;
(9) corrective and preventive action; (10) labeling and packaging
control; (11) handling,
storage, distribution and installation; (12) recording (general
requirements, device master
record, device history record, quality system record, and complaint
files); (13) servicing;
and (14) statistical techniques (FDA, 1996).
2.3.2.4 Post-market Requirements
FDA approved devices should satisfy different types of post-market
requirements that
consist of: post-approval studies, post-market surveillance
studies, tracking systems,
reporting of device malfunctions, serious injuries or deaths, and
registering the
establishments where devices are produced or distributed (FDA,
2009q). Post-approval
38
studies are conducted for Class III devices as stipulated
conditions which are part of the
PMA, HDE or PDP. FDA may require Post-market surveillance studies
for Class II or
Class III devices which: (1) failure may have potential adverse
health consequences, (2)
intended use may be for a pediatric population, (3) type of contact
may involve an
implantation in the body for more than a year, or (4) function may
be life-sustaining/life-
supporting (FDA, 2009u). The inclusion of these devices (with the
exception of the
pediatric case) with a tracking system may be required for better
identification of
problems and to maintain traceability of the devices at the user
level. As part of the
Medical Device Reporting (MDR) regulations; companies should report
to FDA instances
of device malfunction complaints, injuries and deaths. Summaries of
reporting
requirements for user facilities and manufacturers are included in
Tables 2.6 and 2.7,
respectively.
Table 2.6: Summary of reporting requirements for user facilities
(FDA, 1997b)
39
Table 2.7: Summary of reporting requirements for manufacturers
(FDA, 1997b)
2.4 Design, Development and Manufacture of Medical Devices
A robust and well organized MDD process model should be followed
for an effective
device development and for the compliance with regulatory
requirements (Rochford and
Rudelius, 1997; Pietzsch et al., 2009). Design for X (DfX) concepts
may be applied to
achieve specific objectives, e.g. manufacturability, usability and
et cetera. This section
discusses MDD process models and DfX methods in the literature for
MDD.
2.4.1 MDD Process Models
Numerous procedures have been proposed for the design and
development of generic
products. Ogot and Okudan-Kremer (2004) performed a comparative
analysis of relevant
literature (included in Table 2.8) and had defined five steps for
the normative design
40
and (5) production.
Although the MDD process is very similar to this framework, the
extensive
regulatory control over medical devices presents a need to develop,
separate, and adapt
models for MDD. Medical devices impact human life by prolonging,
sustaining,
improving and/or supporting it. They can also represent a threat to
human health because
of the potential risk of illness or injury. Consequently, medical
devices are regulated by
government agencies worldwide. In addition to presenting varied
submission
requirements for device approval/clearance into the market, FDA
regulations stipulate
that the definition of design controls (as part of their Quality
Systems (QS) regu