Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
An analysis of third -party regulatory audit systems for medical device safety
(USC Thesis Other)
An analysis of third -party regulatory audit systems for medical device safety
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
AN ANALYSIS OF THIRD-PARTY REGULATORY AUDIT
SYSTEMS FOR MEDICAL DEVICE SAFETY
by
Lillian J. Gill
A Dissertation Presented to the
FACULTY OF THE SCHOOL OF POLICY, PLANNING,
AND DEVELOPMENT
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PUBLIC ADMINISTRATION
December 2004
Copyright 2004 Lillian J. Gill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3155410
Copyright 2004 by
Gill, Lillian J.
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3155410
Copyright 2005 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
This dissertation is dedicated to the following people, to whom I am eternally
grateful for their never-ending advice, love and support. I can never repay them for
all the good I have accomplished and so readily enjoy.
To my mother, Mildred Benton, who serves as my role model for her courage
and grace.
To my children, Kristen Hairston and Lauren Montgomery who often
reminded me of their childhood code: “Rest if you must, but don’t you quit.”
To my anchor, friend and husband, Kenneth Gill who is the wind beneath my
wings.
To my new grandson, Joseph Solomon Hairston, who I hope will continue on
this path.
To those loved ones who departed before seeing me cross this finish line:
Father, Martin Benton; Grandmother, Ava Harris; Sister, Anita Wingfield, and
Mikkie.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
Reaching this pinnacle has been a blessed and incredible journey. Although
there have been many bumps on this path, there have always been faithful
cheerleaders supporting and encouraging me along the way. I want to take this
choice opportunity to thank those who have helped me to walk this trail, who kept
me on track when I felt like I couldn’t go another step, and who carried me to the
finish line.
I want to first give thanks to my Lord and Savior Jesus Christ, for it is only by
His grace that this project has been completed.
This phase of the journey would not have been possible without the guidance
and support of my dissertation chair, Dr. Elizabeth Graddy. Many thanks for your
help and patience. I also give my sincere thanks to my other committee members and
longtime supporters in this doctoral program, Dr. Joe Wholey and Dr. Beverly
Daniel.
A special thanks to the many friends, coworkers, and fellow students whose
advice, encouragement, fellowship and support sustained me throughout this
educational journey. While there were many who provided a stimulating
environment and thought-provoking discussions, I would like to thank Dr. Bernard
Barnes, Dr. Olatokunbo Fashola, Dr. Todd Hairston, Debra Adams, Daniel
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Montgomery, Chet Reynolds, Linda Suydam, and Cherie Wells. To my mentor and
friend, Dr. Edappallath Radha, your guidance and support mean more than you will
ever know. Thank you for never allowing me to consider stopping.
Finally, I acknowledge the love, encouragement, and support of my family. I
owe special thanks to my children, parents, in-laws, and church family who never
failed to believe in me. I give a most heartfelt and deep acknowledgement to
Kenneth, my husband and life companion who has been so patient and supportive
through the many struggles, the starts and stops, and finally the success of a
completed dissertation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
DEDICATION ......................................................................................................... ii
ACKNOWLEDGEMENTS..................................................................................... iii
LIST OF TABLES.................................................................................................... vii
LIST OF FIGURES................................................................................................... viii
ABSTRACT ......................................................................................................... x
Chapter
I. INTRODUCTION...................................................................................... 1
Research Concerns............................................................................. 8
Purpose of the Study........................................................................... 10
Research Questions............................................................................. 11
Limitations........................................................................................... 12
Methodology....................................................................................... 13
Terminology of Auditing................................................................... 16
II. LITERATURE REVIEW........................................................................... 24
Introduction........................................................................................ 24
Privatization........................................................................................ 26
Accountability.................................................................................... 36
Summary..................................................................................... 43
The Role of Government................................................................... 45
The Rise of New Public Management.............................................. 48
The New Public Service Alternative................................................. 53
Summary..................................................................................... 55
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
III. MODELS OF THIRD-PARTY AUDITS.................................................. 57
Introduction ........................................................................................ 57
An International View on Regulatory A udits.................................. 60
European Audit Model .............................................................. 63
Canadian Audit Model .............................................................. 76
Australian Audit M odel............................................................. 83
FDA’s Regulatory Inspection Process............................................. 89
IV. AN FDA THIRD-PARTY AUDIT PROGRAM..................................... 108
Introduction ......................................................................................... 108
States as a Third-Party Auditor.......................................................... 108
Comparison of Inspection Results.......................................................117
V. SUMMARY AND CONCLUSIONS ....................................................... 124
SELECTED BIBLIOGRAPHY............................................................................... 137
APPENDICES
A. List of Study Interviewee and Discussants ...................................... 146
B. Interview/Discussion Areas and Questions ..................................... 147
v i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
1. Third-Party Inspection Accomplishments...................................................... I ll
2. Comparison of FDA and Third-Party Inspection Costs................................. 113
3. Comparison of Form FDA 483 Observation Findings.................................. 119
4. Third-Party Inspection Classifications.............................................................121
5. Percentage of Third-Party Inspection Classifications................................... 122
6. Comparison of Top Ten Deficiencies Found................................................ 123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
1. Auditing Areas in a Medical Device Manufacturing Process...................... 17
2. Relationship Between Cost and Accountability and Government
Production and Control.................................................................... 26
3. Advantages and Disadvantages of Privatization.......................................... 45
4. Changing Role and Responsibility of Government ..................................... 55
5. EU Audit Process........................................................................................... 69
6. Canadian Audit Process ................................................................................. 81
7. Australian Audit Process................................................................................ 87
8. Laws that Increase FDA Responsibility....................................................... 91
9. Growth of Registered Manufacturers............................................................ 92
10. FDA Inspection Process................................................................................. 94
11. Accomplishments of Domestic and Foreign Medical Device
Inspections ........................................................................................ 96
12. Five-Year Recall Statistics............................................................................... 101
13. Medical Devices Most Frequently Recalled ................................................ 102
14. Violation Rates for Domestic Inspections ..................................................... 104
15. Violation Rates for Foreign Inspections ...................................................... 104
16. Costs Associated with an FDA Inspection...................................................... 113
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17. Relationship of Four System Types to Accountability and
Efficiency Costs ..................................................................................125
18. Relationship of Four System Types to Government Production
and Control......................................................................................... 126
19. Characteristics of Government Production versus Control........................... 128
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Auditing the quality process used to manufacture medical products provides
benefits for multiple stakeholders. Assurance that the manufacturing process follows
acceptable standards of quality provides stakeholders with confidence that
manufacturing-related risks and adverse events are minimized. When government
provides this protection, the resources required to audit and assure regulatory
compliance proves to be costly. Foreign governments have adopted private auditing
systems in response to reduced government budgets and global markets’ demands for
uniform auditing systems.
The purpose of this exploratory, descriptive research study was to help
government officials, public administrators and stakeholders become more effective
decision makers, managers and participants by enhancing their knowledge of the
implications of cost and accountability when privatizing regulatory audits for quality
manufacturing of medical devices. The objectives of the study were: (1) to examine
the costs and benefits of privatization with respect to efficiency and accountability,
(2) to evaluate privatized models in use by other governments, and (3) to evaluate the
use of third parties in conducting Food and Drug Administration (FDA) medical
device inspections.
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This study utilized a combination of qualitative and quantitative research
methodologies and secondary data analysis. The information and data were collected
from a critical review of the literature involving privatization of audits, in-depth
interviews and discussions with representatives from the FDA and international
governing bodies that audit medical device manufacturing quality, and various
government documents.
This study concluded that with training and experience, third parties can
conduct inspections comparable to the FDA with respect to detecting manufacturing
deficiencies in an equivalent inspection time period. Findings also suggest that while
production costs may not be significantly different, transaction costs associated with
the need for continuous communication, training and monitoring may cancel savings
realized or increase the cost of having third parties conduct inspections. The study
also indicated that concerns about accountability when cost savings are the driving
force for privatization could be reduced if government balances their control and
production in the audit process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER I
INTRODUCTION
By law, various regulatory agencies are responsible for protecting the health
and safety of the public. Agencies, such as the Environmental Protection Agency
(EPA), the Occupational Safety and Health Administration (OSHA) and the Food
and Drug Administration (FDA), are mandated to provide assurance that businesses
adhere to the laws and regulations regarding environmental, occupational, and food,
drug, and medical product safety. Traditionally, agents of the organizations charged
with the responsibilities provide these assurances through the review and audit of
procedures that often require on-site visits. These audits provide valuable
information that improves medical product quality and credibility for developers and
investors, and health care providers and their patients.
Manufacturers can market with confidence a product that passes minimal
quality standards, and investors can feel secure that an audited commodity insulates
them from financial risk. Arens and Loebbecke (1997) indicate that in a market
setting, auditing protects investors from risks by reducing the bias that may be
introduced through the management of the organization, particularly in a financial
setting. Auditing benefits and reduced risks expand beyond the scope of the
financial settings that have been the source of recent media attention. In the health
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
care setting, audits provide assurances that proper procedures for medical care and
patient handling are in place and are being conducted so that the highest quality
product is available to the public health sector. Auditing the quality processes used
in the manufacture of medical products also helps to minimize risk to public health
and safety. This oversight provides both safety and economic benefits, as the risk of
a poor quality product may lead to injury and financial loss due to subsequent legal
actions. Purchasers or users of products that pass the scrutiny of an audit can
manage with a sense of credibility that the product behaves as intended.
Auditing also improves what the public, and particularly the buyer, knows
about the credibility of the product. Audit value is derived in part from its ability to
improve the quality of information and, hence, its ability to impact purchasing
decisions in the marketplace. Boylan (2000) points out that the introduction of
auditing and the knowledge that a process has undergone an audit improves the
quality of information available in the marketplace and mitigates the need for buyers
to price-protect themselves because the uncertainty about their asset quality is
reduced. When viewed from an economic theory perspective, auditing impacts
market economy and its participants by improving market performance.
Benefits of auditing can be prospective as well as retrospective. Auditing the
design and production of medical products produces valuable knowledge and
insights into how and why potential failures of control or prevention can lead to
adverse incidents. Intelligence gathered through the observation and investigative
process can uncover invisible risks, determine the scope and nature of emerging
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
problems, identify the components or vulnerabilities of a particular noncompliance
trend, and suggest remedies. In an enforcement setting, information about what the
violators knew or did not know and what factors or incentives produced unfavorable
behavior is vital to improving control (Sparrow, 1994). Despite the potential
negative or costly consequences that unfavorable findings have on businesses, audit
information, when fed back into the control operation, may be significantly more
valuable than the immediate disposition of the failure itself (Sparrow, 2000).
In addition to the economic and marketing benefits that auditing provides to
the private sector, an oversight audit of manufacturing processes and procedures
provides a means of assuring the public that the producer and the medical device
produced is adhering to some level of quality. The ultimate consumer or patient can
feel assured that the product provides the benefit required for improved health
without adverse medical consequences. These audits reduce the public’s
vulnerability to adverse medical events caused by products intended to treat or
products that fail during treatment and seek to minimize unethical business practices
that can result in monetary losses due to economic fraud. Traditionally, government
has been responsible for the oversight of risk reduction for these commodities, as the
public has come to expect a vigilant and responsive government that provides
individual protection from the negative consequences of market failures. This
increasing expectation that government provide a protective shield for all
stakeholders proves to be costly, as significant numbers of resources and regulations
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
required to meet that public expectation result in the growth of government
responsibility and size.
Over the last two decades, the movement toward reducing big government
left public health agencies with smaller budgets and fewer human resources.
Concurrently, technological innovation, particularly in the development of medical
products, grew exponentially; thus, creating a widening gap in the government1 s
ability to oversee the safe manufacture of medical products and diminishing its
capability to limit adverse risks from new technology. For example, the Food Drug
and Cosmetic Act of 1976 mandates biennial investigations or audits of 13,000 to
16,000 manufacturers and distributors of moderate to high-risk medical products,
such as heart valves, surgical lasers, and artificial tissue. The FDA estimates that
there may be as many manufacturers outside the U.S. market that provide medical
products to U.S. consumers, only a fraction of which register with the Agency.
Reform efforts designed to reduce the size of government and cut burdensome
regulatory requirements reduced the FDA auditing capability, beginning in 1989,
from once every 2Vz years to a projected coverage in 2003 of once every 9 and 11
years for domestic and foreign manufacturers, respectively. The Agency and the
market feel the impacts of audit reductions. Data developed by the FDA in a 1998
budget request for additional auditing resources appears to confirm the opinion of
Kachelmeier (1991) that lapses in time between audits or limited and abbreviated
auditing alters the qualitative nature of products and reduces the level of assurance
about asset quality. According to an FDA executive, in a number of public meetings
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of regulatory affairs executives, many manufacturers express the concern that
although internal audits are performed as part of their routine quality process, they
note an association between the length of time between FDA audits and the increase
in the number and severity of manufacturing violations noted by the FDA
investigator. FDA data also indicates a connection between the lack of investigator
availability and the delay in bringing innovative medical technologies to market.
Several reasons contribute to this delay, including the lack of interaction with or
guidance given to the manufacturer on audit findings or the reduction of training or
instruction that the FDA provides during the audit process.
The customary way that the government addresses these resource shortages is
to request an increase of staff. One might expect that given the safety and risk
significance of the products under FDA’s responsibility, increased staffing
authorization would face little challenge. Congressional response to the potential
increase in “command and control” over the regulated industry, if additional
resources dedicated to auditing the device manufacturing market are allocated, is
divided. Often, there is agreement that the FDA should do more to protect the end-
user of manufactured products, but disagreement over who should provide that
protection. Some elected representatives call for a reduced but more responsive
FDA with greater accountability for meeting statutory obligations and assuring
minimal risk to the health and safety of users and patients. Others insist that given
the complexity of technology, the increase in size of the industry, and the diffuse and
global manufacturing market, FDA adopt the international model of relying on
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
industry self-audits (first party) or market-based third-party audit oversight. The
regulated industry also appears to have divergent opinions. Many consider FDA’s
involvement in the audit process to be a deterrent to free market trade and call for
market-based approaches that are independent of any FDA reporting requirement.
Still others consider the FDA presence and role critical to assuring high standards of
manufacturing quality and behavior and consider the role that the FDA plays in
providing a protective, level playing field across the manufacturing community to be
a key government function.
The convergence of political, economic and global forces exerts considerable
pressure on government to alter traditional means of delivering audits for quality
manufacturing practices. Mounting and conflicting pressures driven by the push for
a smaller more responsive government, the demand for more accountability in
meeting statutory obligations for audit oversight, an increase in the public
expectation that the government ensure minimal risk to health and safety from
medical products, and the need to contend with the global manufacturing market
prompts the FDA and the legislature to consider privatizing those regulatory audit
functions once considered sacred to the government. Privatization has become the
popular answer for this dilemma because it allows market-based involvement with
the promise that the government will realize greater cost savings without eliminating
the value that auditing provides. While privatizing these functions may result in cost
savings, it may also signal a change in the Agency’s accountability to the market,
Congressional leaders, and the public. Services provided to all industries that
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
increase the understanding of manufacturing standards and requirements, provide
transparency of the process to consumers, respond to accountability expectations of
legislative leadership, and make available data and information used to correct
market problems are likely to become casualties of privatization. Unlike a typical
principal-agent contractual agreement that would hold the contractor accountable in
all manners to the FDA, one of the most noted third-party programs under
consideration calls for a primary relationship between the manufacturer and the
independent auditor. This monetary agreement is exclusive to the FDA and provides
no direct government measure of control or accountability for the audit information
generated or the resolution of concerns identified through the audit.
While privatizing audit responsibilities in such areas as trash collection,
management of prisons, and banking is an acceptable practice in the United States,
the government has not relinquished responsibility for regulatory audits that
determine the safe manufacture of medical devices to third parties. Faced with the
similar challenges of increasing government powers, expanding budgets, and global
distribution of medical products, international governments, such as the European
Union, Canada and Australia, have turned to various forms of privatizing this
function to reduce cost and regulatory burden. The models adopted by these three
governments provide different approaches to privatizing regulatory audits.
Alternative options to the government performing this task raise questions of
whether the function of auditing the quality manufacture of medical devices to
minimize risks is solely a government responsibility for the United States, how that
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
responsibility might be shared with the market, and whether privatizing actually
saves cost. Although the three other governments have limited experience with
privatizing audits, there is little in the published literature that examines the impact
privatization has on the cost to government or the accountability for ensuring that
public safety is preserved.
Research Concerns
With the strong possibility of radical changes in the U.S. practice of auditing
medical devices for manufacturing quality and risk reduction, there are a number of
concerns which serve as the basis for conducting the exploratory, descriptive
research undertaken in this study. First, the fact that after 25 years of FDA
“inspection” oversight for the manufacture of medical products and at least three
legislative changes to the original Medical Device Amendments of 1976, all of
which strengthened the government’s regulatory authority over products because of
deaths and adverse events, privatizing the audit function is part of the new legislative
agenda. One factor driving this Congressional consideration is the push from U.S.-
based multi-national corporations to move the FDA system of auditing toward the
international model, one that utilizes non-government auditors and has little
government oversight. The global marketing of products prompts those that move
commerce across nations to adopt a philosophy of audit once and market anywhere.
Should this legislation become a reality, this will certainly change the management
and oversight role of the auditing program in the FDA.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Second, after multiple attempts to develop administrative budget proposals
that address the unfunded auditing mandates of the original device law and the
subsequent safe medical devices amendments enacted by the legislature, information
evaluating the cost and benefit of retaining this function in the FDA as opposed to
using alternative options is sparse. It is the assumption of this writer that the lack of
information on the impact of using alternative approaches, which includes data on
cost, accountability and other unidentified benefits, has been a major reason for the
lack of an informed legislature and public.
The third concern stems from the fact that although the emphasis on
measuring performance of programs is not new to government, there is a renewed
emphasis on the accountability for programmatic outcomes. In fact, under the
current Bush administration, the Office of Management and Budget uses the
Performance Assessment and Rating Tool (PART) to determine if government
agencies are accomplishing stated goals. Accountability is unique in this process
because of the link between the PART score a government agency earns and the
budget it receives. If third parties are used to accomplish the mandated goals of
auditing medical device manufacturing quality, the FDA must be knowledgeable
about the impact that this different approach has on ensuring that the legal and
societal requirements are met and that there is an appropriate allocation of
responsibility and accountability.
The scarcity of resources available from the legislature and the elevation of
budget-linked accountability will undoubtedly call for a closer scrutiny of how
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
government resources are used and will likely result in changes to practices
traditionally considered the government’s domain. Nonetheless, the public will
continue to expect government assurances and, therefore, accountability for public
health protection from medical-device-induced risks, and adverse events will
probably remain the responsibility of FDA.
Purpose of the Study
The purpose of this exploratory, descriptive research study is to better
understand the benefits and risks to efficiency and accountability when government
transfers or relinquishes control and involvement in the audit for manufacturing
quality and production of safe products to the private sector. Information gained
through this effort is intended to help legislators, public administrators and
stakeholders become more effective decision makers, managers and participants by
enhancing their knowledge of the many considerations that factor into adopting,
implementing and evaluating the use of third parties to conduct audits that determine
a medical device market’s compliance with the regulations governing good
manufacturing practices.
The primary goal is to analyze the costs and benefits of having the
government conduct these audits as opposed to third parties. In line with this study
goal, the objectives of the study are: (1) to examine the costs and benefits in terms of
efficiency and accountability of the FDA medical product audit program and (2) to
evaluate other models of third-party involvement in auditing manufacturers of
medical products.
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This exploratory research study on the privatization of regulatory audit
functions provides a rare opportunity for public administrators charged with this
responsibility to examine and review their auditing programs in an efficiency-
accountability framework. As a result of this examination, they may be better
equipped to plan for or defend against partial or total privatization of their audit
responsibilities. This research also provides an opportunity for public administrators
of regulatory programs to place the auditing practices of a U.S. regulatory body, such
as the FDA, within a global perspective.
Research Questions
The data collection methods are guided by the following research questions:
1. Is there a significant difference in the cost and efficiency of regulatory
audits of medical device manufacturers when conducted by the traditional
regulatory authority or the independent third party?
2. Are there differences in benefits to customers in terms of proactively
identifying problems when audits are conducted by the regulatory
authority versus the third party?
3. What are the appropriate accountability mechanisms for government and
third-party auditors? Is there a tradeoff on accountability when efficiency
increases and cost decreases?
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Limitations
This study is limited to the following scope:
1. The study focuses on the processes used to conduct regulatory audits of
medical device manufacturing facilities. It does not cover the educational
or training requirements of auditors or the actual steps in a quality
manufacturing audit.
2. The study sample of audit processes or systems is limited to the United
States, the European Union (EU), Canada, and Australia.
3. The case study is limited to an FDA-State contractual agreement for
third-party audits conducted during the period between 1994 and 1996.
This third-party inspection program is examined to provide actual
information on where efficiencies are likely to be realized and where
questions of accountability may need to be addressed in contemplating an
alternative inspection model.
4. The study does not include data from nongovernment or independent
third party standards development organizations.
5. Limitations on the ability to obtain significant data on the cost of audit
programs for any of the government systems analyzed prevents
performing a benefit-cost analysis of privatizing audits or maintaining
audit functions in-house.
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Methodology
Since the FDA will continue to confront issues of cost efficiency and
accountability they must consider alternative approaches to governing—ones that
balance the advantages and disadvantages of privatizing the medical device
manufacturing quality process. To achieve the goals of this study, information and
data will be obtained from four sources:
• Literature review on the impact of privatization on efficiency costs and
accountability, the changing role of bureaucracy, the use of third parties
in general auditing practices and in public health settings;
• Case study analysis of historical information and performance data from
an FDA third-party program in which third-party state inspectors were
used to conduct inspections;
• Interviews and discussion with FDA staff and international medical
device regulators responsible for leadership and policy development; and
• Data from FDA annual reports, speeches, publications, manuals, and
databases.
The literature review will examine the impact privatization has on cost
efficiency and accountability. This dissertation will also discuss the underlying
theory, role, responsibility, and accountability of public administrators in three
models of government operation: (1) traditional bureaucracy, where government
functions in a controlling mode; (2) the New Public Management, where government
functions as the “steerer”; and (3) the New Public Service model, where government
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
acts as the director/server. These three models highlight the changing roles and
responsibilities of government in response to the pressures that force government to
operate like business and the impact of embracing those market values on the service
provided.
Audit procedures, laws and requirements for conducting medical device
audits in three countries are examined to contrast the traditional approach used by the
FDA. To complement the literature on the use of third parties by foreign
governments, executives from Australia, Canada, and the EU that are involved in
their audit programs will be interviewed. These countries were selected because the
role their government plays in the unique approach to using third party auditors—
government as conductor, government as participant and government as observer—
provides insight into how the FDA might structure a future market-based audit
program. Since the level of involvement and control exercised in the three
bureaucratic operating modes mirror the auditing models used by the EU, Canada,
Australia and the United States, each of the audit systems will be examined and
discussed in light of the influence and control exercised by the respective
governments and the impact each model has on efficiency, cost, and accountability.
Finally, this paper will examine the only third-party program that the FDA
used to conduct regulatory audits of a medical device manufacturer’s compliance to
quality manufacturing requirements. This examination specifically looks at program
cost, the ability of contractual inspectors to conduct FDA inspections and find
critical deficiencies, resource savings or other benefits realized from using contract
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inspectors. Any areas of difficulty that result from using this approach also will be
examined.
Data and information from a number of sources were examined during this
review, including files from the contract program, individual investigation reports,
quarterly reports from participating states, written comments from device
headquarters staff involved in the program, information from the database on
inspection findings (CDRH FDA 483 Data Base), and discussions with analysts and
investigators involved in the program. This examination specifically looked at
program cost and the ability of contractual inspectors to conduct FDA inspections
and find critical deficiencies, resource savings or other benefits realized from using
contract inspectors. Any areas of difficulty that result from using this approach will
also be examined.
Records, speeches and information found in the annual report of the FDA
Center for Devices and Radiological Health and the Center’s databases provide
quantitative data on the numbers of inspections conducted, monetary and human
resources expended for the inspection program, interactions with a variety of
stakeholders, enforcement interactions, and other types of assistance provided to the
manufacturing and consumer community. This data will be used to develop a simple
cost estimate for inspections conducted by the FDA.
The discussion of this analysis and the examination of the three bureaucratic
operating models are intended to provide useful guidance for public administrators
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
charged with overseeing the regulatory audit program that determines medical device
manufacturing quality.
Terminology of Auditing
Many terms, such as audits, inspection, investigation, and regulatory
compliance audit, are used to describe the manufacturing quality assurance oversight
process. These terms describe functions that are carried out by both government and
the private sector to examine various stages of the manufacturing process in order to
verify that the product is designed to operate safely and effectively. A medical
device or product is considered safe if the use does not present any hazard to the user
or to the patient. The product is effective if it performs as expected for its intended
purpose. The elements of the manufacturing process subject to an audit are shown in
Figure 1. While regulatory audits are typically connected to those conducted by
government entities or are conducted to ensure compliance with government
regulations, the various “party” audits signal external involvement in the auditing
process. A description and discussion of the terms used is provided to increase the
understanding of when each sector is involved in this process.
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Device Approval— audit manufacturing capability
Pre-market Testing— audit clinical data integrity
Device Manufactured— audit quality system
Device Used— audit for safety surveillance and communication
Device Available for Sale— audit uniform production capability
Device Design— audit validation and verification o f design
Figure 1. Auditing areas in a medical device manufacturing process.
Auditing is a process used to assess the effectiveness of quality assurance
efforts and to evaluate compliance with applicable quality standards. During the
performance of an audit, auditors must objectively and independently collect and
verify the evidence, evaluate it against audit criteria, and report the findings.
An inspection or administrative search is the most important and
commonplace method of monitoring and enforcing health and safety standards. It
represents a formal and careful examination of a product, business, or premises to
determine its authenticity (e.g., possession of a valid license), quality (e.g., purity
and fitness for use), or condition (e.g., safe and sanitary) (Gostin, 2000). The FDA
uses the terms “inspection” and “investigation” in their assessment of a
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
manufacturer’s adherence to quality system principles found in the law. As
referenced in the FDA Investigations Operations Manual 2003:
An establishment inspection is a careful, critical, official
examination of a facility to determine its compliance with
laws administered by FDA. Inspections may be used to obtain
evidence to support legal action when violations are found, or
they may be directed to obtaining specific information on new
technologies, good commercial practices, or data establishing
standards or other regulations.
FDA inspections may cover a comprehensive look at everything manufactured under
FDA jurisdiction or may focus on specific areas of interest.
An investigation is an information gathering activity designed to determine
and document facts concerning a particular issue in order to facilitate an informed
and sound decision. Examples of investigation types include complaint, disaster,
health fraud, and product tampering. Investigations conducted by the FDA can be
distinguished from other inspections by the nature of interaction (report of findings
not issued) and the location (retail establishment or consumer location versus a
manufacturing facility). Although the terms “inspection” and “investigation” are
used interchangeably, an FDA inspection is more closely aligned with the general
definition of an audit. Because the term “audit” is more universally used than
“inspection,” in this paper, “audit” is used to describe the process of assessing the
quality of manufacturing procedures used by government and nongovernment
bodies, except when discussing the process that the FDA uses.
A regulatory compliance audit assesses the operations of an existing
manufacturing facility to determine the degree of compliance with applicable laws
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and regulations. Operating practices and procedures are inspected in light of relevant
requirements by in-house legal or technical staff or by an independent consultant.
The failure to comply with regulations and laws can result in significant sanctions,
including heavy administrative, civil, and/or criminal penalties. Manufacturers use
many types and levels of audits to assess manufacturing quality because audit results
can have a dramatic impact on future business decisions and viability.
When not conducted by a government body, regulatory compliance auditing
by external assessors falls into three categories: first, second, and third party. These
categories are differentiated according to the contractual relationship with the
manufacturer. A “first-party” audit is one in which an organization audits itself;
that is, the auditor is employed by the organization that is being auditing. First-party
audits are the preferred mode of operation for organizations. Similar to self
regulation, the first-party audit means that the rules that govern behavior in the
market are developed, administered and enforced by the people whose behavior is to
be governed (Uche, 2001). Self-auditing usually arises out of the desire to look
internally at manufacturing capability, to repel the threat of government-imposed
regulation, to curtail the activities of fringe operators, and to protect industry
reputation. Benefits from this approach include: flexibility in problem solving,
availability and use of those most knowledgeable about the business, attainment of
higher standards than the minimalist provisions described in statute laws, and
potential costs savings to the market and government, as self-audit costs are normally
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
internalized within the regulated industry. Self-auditing can raise questions of
conflict of interest and fraud.
A “second-party” audit is one in which a customer audits a supplier or
contractor. In general, a second-party audit focuses on aspects of the manufacturing
process germaine to the product or service under contract and not on the entire
manufacturing system. The second-party audit can be considered the most powerful
weapon because of the extraordinary purchasing power that a customer has over a
supplier or contractor (Uche, 2001). A customer who includes product safety
requirements that comply with the regulations can clearly define the level of quality
and, therefore, consumer protection that needs to be addressed. For example, an
automobile manufacturer that specifies acceptability criteria for parts used in vehicle
assembly sets a standard of quality for the vehicle purchased. Likewise, a pacemaker
manufacturer that specifies battery quality and audits against those specifications
greatly influences the safety and endurance of the implanted device, thus
contributing to the decreased risk from malfunction or expiration. Quality of the
product purchased from the component manufacturer and the financial impact of not
meeting those specifications can have a far greater impact than a citation given by
the Federal Trade Commission or the FDA.
“Third-party” audit commonly refers to the use of external individuals or
entities whose services are usually paid for by the organization that is to be audited.
Theoretically, these auditors are outside the direct control of the contracting
organization and may have no vested interest in specific products being
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
manufactured. Third-party audits have been used in the oversight of occupational
and environmental compliance health programs for many years. OSHA uses special
nongovernmental employees to help with Voluntary Protection Program inspections,
the American Industrial Hygiene Association members to perform audits of
accredited laboratories, and the National Science Foundation International to inspect
and certify all bottled drinking water marketed in the United States.
The recent elevation in use of third-party auditors in lieu of government
assessments of quality manufacturing processes can be attributed to three factors.
First, a government directive—specifically, language in the National Performance
Review (NPR) of 1993 that called for the potential use of third parties in the
Department of Labor to audit for work-site safety and health. Second, Congressional
legislation—the Food and Drug Modernization Act of 1995—mandated the use of
third parties in the review of scientific and technical applications for new medical
devices seeking entry into the market, an activity traditionally performed by the
FDA. Third, and perhaps the most influential reason, is the impact felt from the
international manufacturing and trade market that utilizes third parties to assess
manufacturer conformance to specified standards. Gaining widespread momentum,
this conformity-assessment structure developed by the International Organization for
Standardization (ISO), as reflected in the ISO 9000 “quality assurance management
system,” is the current system in use among the 25 nations that form the EU. A
program that includes a privatized mechanism for auditing the auditor and the
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
company that hires the auditor governs the worldwide system that uses third parties
to conduct ISO audits.
Conformity assessment includes a wide range of approaches applicable to
sampling, testing, inspecting, and evaluating specific processes as well as reviewing
the certification of quality system assessment and registration using any of these
procedures. Under conformity assessment, products are evaluated to determine
whether they conform to particular standards or specifications. Assessing
conformance to standards is not new or unique to ISO. For many years, third parties
played a significant role in numerous conformity-assessment models for various
sectors of the economy, e.g., Underwriters Laboratory (UL), a well recognized
independent electrical testing body that assesses product safety through its
conformance to electrical standards. Toasters, microwaves, and hair dryers are
examples of consumer products that bear the familiar UL symbol. The use of third
parties to inspect for market compliance to regulations is not the norm for regulatory
bodies in the United States. Although cases exist where the government shifted its
auditing responsibility for risk reduction to third parties, such as the use of
independent auditors by OSHA and the EPA, assessment of conformance to
regulations or standards for manufacturing medical products in the United States
remains primarily a government function.
Much of the current debate about privatizing audits focuses on the efficiency
cost—savings realized when accomplishing the audit activity at the least cost, to
government for providing the service. For this discussion, costs refer to those
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
monetary and human resource expenses required when conducting an inspection or
audit. Despite the merits of cost savings from privatizing audits, a cost-based
argument against government involvement ignores critical contributions that
government auditors make to public health safety and market efficiency. Public
sector investigators provide a degree of independence and impartiality that enhances
the implementation of regulatory requirements. They are needed to ensure the
legitimacy of the process. The investigative arms of governments’ skilled
professionals are needed to independently assess quality or oversee the performance
of private sector auditors who provide data to public agencies so that reliable quality
is assured. The experiences of these investigators can also provide a resource for
developing policy and assessing training requirements. The debate should, therefore,
not be focused solely on cost savings to government, but on the value of having
impartial public employees provide the service and determining the amount and type
of service that government should provide to be accountable to the citizens it
represents.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER II
LITERATURE REVIEW
Introduction
The underlying reason for regulatory audits is the desire to minimize, either
proactively or reactively, risk to public health and safety that may occur from the
manufacture or use of medical devices. The privatization debate in the context of
auditing to minimize those risks revolves around efficiency and accountability
concerns, specifically whether market-based mechanisms provide the same level of
oversight protection against risk at lower costs and with equivalent accountability as
government. Arguments for market-based auditing are similar to the arguments for
privatization in any area—greater efficiency (faster, quicker, cheaper) because the
competitive nature of business is self-regulating, better service because of incentives
and more opportunity for innovation because of less command and control, or
intrusion by government. Arguments against private auditing mechanisms also focus
on general arguments against privatization—accountability for performance,
equitable treatment and, in light of auditing for risk minimization, the focus on cost
savings at the expense of protection against unknown or unnecessary risk.
Government-run auditing programs are least efficient because of the high cost
associated with the resources required to control or manage an audit program and for
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
government’s extensive involvement in producing audits. Control of the program
means that government dictates the structure and management of the program,
including the development of the laws and regulations for auditing, standards of
conduct for auditors and manufacturers, reporting requirements, and procedures for
intervention or enforcement. Significant audit production means that government is
actively engaged in and fully responsible for conducting audits that determine the
quality of manufacturing processes and for assuring that the appropriate actions are
taken to correct manufacturing problems. This level of responsibility requires
having direct access to manufacturers and audit data.
Privatization of the process is desirable because increased use of market-
based systems that require lesser government involvement translates into lower cost
efficiencies. This reduced government involvement and control may also reduce
accountability. An overview of the relationship between cost to government and
accountability with respect to government involvement and control is shown in
Figure 2. The privatization of audits in many ways transforms the largely symbolic
“government inefficient-market efficiency” argument into a debate over the
government’s role in balancing efficiency and accountability and the decisions
government makes regarding whose interest it should serve. This chapter discusses
the benefits of privatizing audits and the challenges that privatization presents to
accountability.
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
►
G overnm ent Control
Figure 2. Relationship between cost and accountability and
government production and control.
Privatization
The process of disengaging the public sector from various production and
regulatory functions while enlarging the scope of the market sector is accomplished
through privatization (Chen & Shin, 2000). In a broad sense, the term
“privatization” is used to categorize a variety of activities and processes that range
from the transfer of ownership from public to private hands to the imposition of user
charges for publicly provided services (Gray, 1998). The term “privatization,” as
commonly used, has two meanings. One meaning describes the shift of government
services to the private sector, thus establishing a direct relationship between the
private provider and the private consumer. That shift may include the sale of
government enterprises and assets, the wholesale movement of government-financed
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
activities to fee-for-service, the transfer of government operations to a market-run
activity commonly known as load shedding, or the demonopolization of government
services to allow private alternatives to emerge (Savas, 1987). Under each scenario
there is no contractual agreement with the government entity involved, although
government may have a regulatory, policing, or indirect financing role.
Alternatively, privatization is “contracting out,” or moving services from direct
supply by government to an entity acting under formal contract on behalf of the
government. Here, the financial relationship exists between three parties:
(1) government as the conduit and monitor, (2) the market representative as the
service provider, and (3) the customer/client as the service recipient. Privatization is
essentially the withdrawal of public responsibility for services that the market may
be better equipped to supply at a cost savings to the government.
The arguments for privatization have emphasized a combination of reduced
costs, improved services, increased management flexibility, specialized expertise,
and decreased public monopoly inefficiencies. As all levels of government
throughout the world experience increased fiscal and political pressure to be more
cost effective in their response to public demands, the response to such pressures is a
growing reliance on adaptations of market-based practices and the use of
privatization in service delivery. Reliance on this concept was so strong that a
Canadian report issued during the Mulroney government, entitled “Structural
Reform: A Blueprint for Change”—a report similar to the National Performance
Review—recommended a 50% reduction in federal departments and a 25% cut in
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cabinet ministers (Fulton, 1993). In the current Bush administration, an estimated
850,000 federal employees—almost half the federal workforce—are identified as
performing commercial work (i.e., janitorial services, printing, logistics,
maintenance, professional services, and network support), defined as work that can
be performed by the private sector. Although similar to the Clinton administration’s
goal of transferring a substantial portion of commercial jobs to the private sector to
reduce cost, the biggest difference between the two is in the type of positions that are
on the target list. Where privatizing or outsourcing was once a primary concern of
blue-collar workers, the Bush administration’s approach now targets large numbers
of jobs held by white-collar government workers, including the auditing of quality
manufacturing systems that produce pharmaceuticals and medical products.
Over the years, every level of government in the United States that used
private contractors recognized the benefits of competitive service delivery, including
lower service costs and equal or better service quality that results from adopting
policies that promote privatization and contracting (U.S. General Accounting Office
[GAO], 1998). This concept spread globally as international governments and large
industries adopted contracting with the private sector to conduct public services. In
Australia, Japan and New Zealand, numerous operations and services conducted by
state-owned businesses, such as utilities, airlines and quality auditing, are expected
to fund their operations entirely from revenues generated from their customers
(Caiden, 1994) or turn those operations over to independent fee-for-service
providers. In a study similar to one conducted by the International City/County
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Management Association (ICMA) on the privatization experiences of America’s
largest cities, 66 of the 456 cities responded to a survey on the scope and benefit of
privatized service functions (Dilger, Moffett, & Struyk, 1997). Although this low
response rate is problematic, an evaluation of the responses provides some valuable
insight. The results show that a total of 106 privatized functions were in the public
works/transportation category, 106 in the support functions category, 98 in the public
safety category, 89 in the health and human services category, and 57 in the parks
and recreation category. City managers agreed that the quality and consistency of
services provided by outside sources improved, as did the cost of services. When
asked the extent to which privatization improved service delivery (from zero to
100%) the responses suggested that service delivery improved across the five
categories from 24.2% to 27.6%. Similarly, when asked about cost savings, on a
scale of zero to 100%, administrators responded that privatization yielded an
estimated cost savings of 16.1% to 20.7%. Although the fiscal benefits of
privatization were encouraging, the survey found that service delivery improvements
and cost savings estimates were much lower than anticipated.
In agreement with opinions that privatization provides substantial benefits in
cost savings, Edington (1997) found that outsourcing “gives an organization the
opportunity to maintain a small cost-effective management team and provide high
quality, high impact services.” Outsourcing, a term invented by the information
management system trade press in the late 1980s and included under the
privatization banner, is also viewed as a way of helping an organization grow both
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
quantitatively and qualitatively by allowing for an investment of resources only in
the most appropriate or core business areas. Outsourcing becomes a means of
supplementing an organization’s capability by acquiring specialized resources and
valuable expertise that cannot be or is not cost effective to be produced and
maintained internally. The literature suggests that privatizing certain key
components or functions of an organization, such as the development of advanced
technologies, should result in cost efficiency benefits because organizations are freed
from having to finance large overhead or start-up costs (Ferris & Graddy, 1986;
Savas, 1987); and better quality services are provided, as suppliers have a direct
monetary incentive for good performance (Prager, 1994). A streamlined
organization yields greater program oversight because the remaining non-privatized
business functions are less diffuse and more focused, thus freeing time for managers
to concentrate more fully on core organizational goals. According to a study by
Keane, Marx and Ricci (2003), contracting out audits while maintaining core
functions enhances organizational flexibility, efficiency, and cost savings.
Although no studies were found describing the benefits of using third parties
to assess compliance to public health standards of quality in the manufacture of
medical devices, the literature contains studies examining the utilization of third-
party inspections to help enforce environmental regulations designed to reduce risk
to public health from environmental contaminants. Belke (2001) found that third
parties enhanced government’s effectiveness because they allowed for more focused
attention by the Environmental Protection Agency (EPA) on those facilities that had
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a history of not committing to regulatory compliance and allowed EPA to better
focus efforts on taking timely traditional enforcement actions. A study on the use of
third parties in the inspection of steam boilers in Hartford, Connecticut, showed that
having adequate inspection coverage successfully reduced accidents in rare events
(Er, 1996). Additionally third party involvement in auditing the implementation of
Occupational Safety and Health Administration (OSHA) workplace safety standards
reduced the fatality rates in the United States by over 30% (Kunreuther, McNulty, &
Kang, 2001a); and hygiene quality improved in restaurants in LA County after the
Department of Health Services instituted a third-party inspection program
(Kunreuther et al, 2001). The concept of using third parties as auditors of quality
and safety was extensively tested and reviewed in 1999 by the Delaware Department
of Natural Resources and Environmental Control during two studies of compliance
with EPA standards for water chlorination and ammonia refrigeration at local
facilities. The McNulty (1999) study showed the value of the existing expertise of
third-party insurance auditors that is brought to environmental safety because of
extensive experience in auditing for loss prevention. A similar study conducted in
the fall of 2000 that used third-party insurance auditors in Pennsylvania confirmed
the McNulty study. In both studies, the local government realized reduced costs for
a number of reasons. By contracting with external experts, both states had access to
the skill level and volume of auditors that neither possessed and from a profession or
market that had a vested interest in public safety. Repeat visits from state auditors
were practically eliminated because industry’s preference for market-based
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
interactions over interactions with command and control regulators encouraged them
to respond quickly and positively to third-party auditors. Finally, because the third-
party insurance audit covered many of the EPA requirements that would have been
examined under other insurance circumstances, Delaware and Pennsylvania saved on
the cost of their audits. The studies conclusively demonstrated that third parties
could successfully conduct regulatory compliance audits with adequate rigor at
facilities required to institute risk management procedures.
Privatization assumes that private sector service delivery is always less costly
and is always of an equal or better quality than public sector service delivery. Many
times, markets or governments view privatization favorably because of the potential
production cost savings experienced from the reduced need for capital assets, such as
human resources, which occurs when contracting with specialists who possess
specific skills (Gargan, 1981; Fiorito, 2003). In essence, privatization is beneficial
because the production or direct cost associated with carrying out the activity is
reduced. The fallacy in this reasoning stems from the fact that the less obvious costs
associated with privatizing could exceed anticipated savings. To evaluate the long
term strategic effects of privatization and the potential for success, the benefits that
might be gained must be weighed against those concerns for quality and for the true
costs that privatizing incurs. Wallin (1997) wrote in “The Need for a Privatization
Process: Lessons from Development and Implementation” that one lesson learned
from instituting a law designed to regulate privatization in Massachusetts was that
cost and performance must be carefully measured before privatization is
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
implemented so that a proper evaluation of privatization’s effect can be made. What
that and other studies found was in addition to the direct or production costs of
conducting an audit, when properly instituted, privatization requires the expense of
managing service delivery for both in-house and contracted service delivery
(Williamson, 1981; Ferris & Graddy, 1991; Madhok, 2002; McNally & Griffin,
2004). Known as transaction costs, these additional costs are associated with
conducting economic exchanges, such as the search, selection, bargaining,
monitoring, evaluation, and enforcement of a private entity. Borcherding (1988)
would suggest that incorporating these transaction costs into the production cost of
an audit may reduce those cost savings often attributed to contracting and may even
outweigh the production costs for market and government.
Reflecting the disappointment in cost savings when contracting with the
private sector, researchers report that productivity and long-term profit gains are
decidedly less than proponents predicted (Kettl & Dilulio, 1995). Transaction costs
associated with the additional managerial time and resources in the form of
paperwork, and oversight capacity (Ramanadham, 1994) as well as the increased
time needed to measure performance cancel any cost benefits realized. Without that
capacity to monitor contractor performance, companies that successfully eliminated
jobs reported that third-party performance produced “mediocre, marginal or failed
results” (Davenport, 1995). A 1996 Deloitte and Touche survey of some 1,500 chief
information officers in a variety of United States and Canadian companies
documented that 69% of those officers were disappointed with the savings gained
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
through outsourcing. Study participants with Coopers and Lybrand reported that
only half of the chief executive officers who turned to outsourcing for cost savings
achieved any, while 29% broke even and 4% lost money.
Local health departments and federal agencies experienced similar
disappointments—the expense of monitoring a third-party auditing effort may well
exceed the cost of a government entity performing the effort itself (Lavelle,
Krumwiede, & Sheu, 2000). In a study conducted by Keane et al. (2001b) of health
department managers, 41% of local health departments privatizing at least one
service experienced an increase in total time spent on administration and program
management. A 1977 Public Health Foundation study found that only one-fourth of
privatization efforts studied yielded any real savings. In the information technology
area where outsourcing is most popular, Dean (1996), a cost analysis expert for
NASA, recently commented that outsourcing to a low quality supplier actually can
increase costs and reduce revenue. Belke (2001) concluded from his study assessing
the success in using third parties to assess compliance with EPA regulations that the
cost of administering a third-party program would require that the EPA invest
additional resources for such activities as developing training and certification
standards, auditing protocols, administering auditor oversight mechanisms, and
maintaining a register of certified third-party auditors. The literature also highlights
the fact that for government, the costs of training, updating and enforcing
requirements of the third parties can increase the costs of privatizing and become so
high that they cancel out any savings. A March 1997 GAO publication,
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
“Privatization: Lessons Learned by State and Local Governments,” looked at the
privatization experiences in six areas: Georgia, Massachusetts, Michigan, New York,
Virginia, and the city of Indianapolis. Among the six lessons identified, the report
emphasized the need for reliable and accurate cost data on government activities in
order to assess the overall performance of activities targeted for privatization.
In an evaluation of whether the government realized real savings from
competitive sourcing, GAO reported on year 2000 data from the Office of
Management and Budget Circular A-76 that “overall program costs still exceed
savings.” The President’s fiscal 2001 budget submission also reported that during
fiscal 1998 and 1999, the overall costs of the A-76 program exceeded the expected
savings. Additionally, in 1999, the House Appropriation Subcommittee on Defense
concluded: “There is no clear evidence that this effort [competitive outsourcing] is
reducing the cost of support functions within DoD, with high cost contractors simply
replacing government employees.” The resulting impact is that the number and
capabilities of public servants required to oversee and monitor contractor
performance increases the number of resources and, therefore, the cost that
government devotes to overseeing contracts and monitoring performance
(Ramanadham, 1994). This may explain why Rehfuss (1989) and Keane et al.
(2001b) found that privatization of public health services increased transaction costs
such that only modest cost savings were realized. Privatizing an audit function
would seem to increase cost because of the need to increase the capacity to manage
and monitor the contractor. Given the push to reduce budgets, however, resource
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
savings from reduced production costs will not be transferred to managing the
process, but will be used as a means to reduce involvement and management
responsibility. Lower production costs translate into budget cuts, reduced staffing,
and eventually reduced capacity. Van Slyke verified this occurrence in a 2003 study
finding that as government increased privatization, it simultaneously reduced its own
public-management capacity.
When government lacks sufficient capacity to oversee private auditors, it is
unable to determine whether the third party actually delivered the service (Milward,
1996), or it may become dependent on unscrupulous service (Kettl, 1993). In other
words, when the public-administrative capacity to monitor and enforce third-party
behavior is reduced, government accountability is reduced; and the public is at
greater risk of being exposed to fraud, waste, abuse, and injury. The advantages of
lower cost must be balanced by effectiveness, retained in-house capacity,
responsiveness, and trust in government (Kettl, 1993).
Accountability
Citizens in a democracy may accept the assumption that shifting to private
ownership or responsibility enhances service delivery of functions at every level of
government; however, most also want their services delivered with effectiveness and
accountability. The public’s desire that government institute creative cost saving
solutions to public problems does not diminish their expectation that accountability
be preserved (Romzek & Dubnick, 1987). While reforms may alter government
involvement in administering safety programs worldwide, they do not alter citizens’
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
concerns that responsiveness, equity, representativeness, propriety, and
accountability be upheld in the process (Durant & Legge, 2002). In his article,
entitled “The Diminishing Publicness of Public Service Under the Current Mode of
Governance,” Haque (2001) writes that while traditional problems of administrative
control have not disappeared, a new set of challenges emerges due to issues related
to the contemporary market-driven reforms in public governance. Closer
partnerships or alliances with private firms change the relationship between citizens
and government, create doubt regarding the transparency of public-private
transactions, increase the public concern regarding kickbacks (Frederickson &
Frederickson, 1995), diminish the degree of government control over private
contractors needed for accountability (Frederickson, 1997), shift accountability to
other under-resourced agencies, and change the functions of government.
Additionally, as privatization sharpens the focus on core government functions and
government adjusts or reduces the scope of those identified functions such that more
of them are privatized, there is increased concern that privatization will undermine
the regulatory authority of governmental agencies.
Bureaucratic accountability is based on close public scrutiny of public service
activities through political debates, legislative committees, and other democratic
means. Unlike reform efforts experienced under the Hoover and Brownlow
Commission, government reforms over the past 25 to 30 years deemphasized the
need for democratic accountability of departmental and agency officers to the
President through the Congress. Legal structures that demand openness, fairness,
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
participation, consistency, and impartiality are not applicable to private entities
conducting government work. The use of nongovernment employees makes it more
difficult for elected officials to retain control of services and to respond to
constituent complaints. In addition, public-private partnerships may disturb the very
chain of accountability because the common citizens “may simply not be able to
determine whether government or its contractors is responsible for the particular
service, and thus officials who want to may be able to evade responsibility easily”
(Peters & Pierre, 1998). Bunker and Davis (1998) also point to a transformation in
the essential character of the relationship between the citizen and the provider, which
those on both sides of the privatization debate generally express in terms of
accountability to some bureaucratic overseer.
In a research study based on their theory of the movement of public
responsibilities to the market, Blanchard and Hinnant (1998) stated that privatization
changes the citizen-govemment relationship by altering accountability arrangements
and threatening the legitimacy of government in general. As such, this suggests that
the accountability linkages between the consumer of medical devices and the
government charged with ensuring safe products is altered as the direct path of
consumer-govemment auditor-political representative is altered, to become one of
citizen — > third-party auditor — > government auditor — > political representative or
citizen — »third-party auditor with a dotted line to legal or political representation.
This weakening of the political accountability mechanism seriously weakens the
contract between citizens and their government. In their examination of the recent
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
trends in adopting market-based delivery of public goods and services, Denhardt and
Denhardt (2000) found that this fundamental change in the traditional relationship
between citizens and government was particularly problematic, with a public that
continues to demand responsiveness from the government. In fact, the literature
would suggest that this demand for responsiveness may very well create an outcome
contrary to that intended and stifle any cost savings that might be realized. For
example, Miller and Simmons (1998) found that in many areas the increased cost of
privatization decreased service, resulting in a “neopopulism” in which the public
demands greater social responsibility and accountability from the business
community. As public legislators seek to increase accountability, greater reporting
requirements are levied on third parties. Increased expectation exacerbates capacity
problems because an increase in reporting requirements increases administrative
costs. The government’s demand for responsiveness creates additional requirements
that limit private sector flexibility and their efficiency. With respect to the device
quality audit process, if the inspection is conducted by a private auditing firm
receiving payment from the manufacturer, the public may demand that the Food and
Drug Administration (FDA) hold that private company more accountable through
increased and more detailed reporting, particularly if the device is perceived as high
risk. This increased demand from a public that wants to minimize personal injury
can result in increased reporting and safeguards that increase the cost to government
and business.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The insertion of a market-based auditing process that alters the citizen-
govemment relationship and creates a demand for accountability also changes the
function of government. This change in public function and responsibility shifts the
role of public service from one of active engagement to a more indirect, supportive
function of facilitating the private sector initiatives (Haque, 2001). Privatizing the
auditing function reduces the interventionist role of government and marginalizes the
role of the public administrator in providing basic services and addressing public
needs (Landry, 1993) as government auditors move from direct involvement in
auditing manufacturing practices to more indirect functions, such as monitoring and
evaluating third party activities. The role of auditing manufacturers changes to
become auditor of the auditor as government seeks to maintain a level of
responsibility for assuring that risks are minimized and because of the perception
that private-sector auditors are less able to withstand client pressures and, therefore,
act more as business advisors rather than guardians of the public interest.
Changing the relationship and the function also affects the auditor and the
work environment as different skills and expertise are required to carry out new
responsibilities. For example, to manage the functional shift with some measure of
accountability, the skills of the government auditor, which include maintaining
current manufacturing processing techniques consistent with technological
innovation, should be equivalent to or exceed those of third-party auditors. Brown
and Brudney (1998) found the opposite to be true. Their study showed that when
states experienced revenue constraints, as will likely be the case in the federal sector,
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
new skills and positions identified as required to monitor privatized activities were
not made available. These priority shifts created yet another set of challenges, as
more of the tasks were delegated to outside sources and the press of everyday tasks
and competing demands created an incentive for the remaining staff to disengage
from the project. When the reduction in the scope of government control over audit
delivery generally falls, feelings of ownership are lost and outcomes suffer (Shleifer,
1998). Daft (1991) noted that organizations relying extensively on outsourcing
frequently experience loss of hands-on control, reduced organizational participation
and weakened employee loyalty. When governments reduce their workforces and
rely more heavily on third parties for expertise in providing public service,
governments become more dependent on external players for information and
services and less able to manage or enforce accountability.
Auditing is also a crucial component of accountability because it legitimizes
the information on which formal decisions rest. Privatizing auditing functions may
not allow ready or easy access to information related to those tasks once performed
by the government. This lack of information hinders the government’s ability to
handle multiple, complex, and often conflicting goals that drive public policies and
processes and minimizes its ability to learn from the environment in order to
influence the course of policy implementation (Mazmanian & Sabatier, 1989).
Normanton (1996) and Mintzberg (1996) argue that commercialization is
characterized by a very strong tradition of secrecy that goes against the idea of public
accountability. Durant and Legge (2002) conclude that the threat of an informed
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
public armed with performance data is the greatest ally in ensuring that both market
and social values inform service delivery decisions. Without information or having
to seek access through other legal routes, the public is essentially denied an
opportunity to become involved in issues that protect their rights and safety. The
loss of institutional involvement and policy expertise threatens democratic
accountability and objectivity, as third parties gain the expertise and can exercise that
capability by engaging in policy formulation and evaluation (Van Slyke, 2003).
Access to public records is overwhelmingly a matter of statutory law in the
United States. Both the federal government, through the Freedom of Information
Act (FOIA), and all fifty states have statutes creating some level of access to public
records. Almost every government activity, including information created during
those activities, except those involving national security, is considered public and is
accessible under FOIA. Unfortunately, neither FOIA nor most state public records
statutes explicitly allow access to the records of private entities performing audits.
Records long open to public inspection now are being created, maintained, and
controlled by private businesses often at odds with the very purpose of public records
laws. Bunker and Davis (1998) concluded that current statutory definitions and
judicial standards that used to draw the line between public and private enterprise
may frustrate the public’s ability to scrutinize the activities of private actors
performing services for the state. Access to records of private operators would
require a Congressional amendment to FOIA; and, without that change, when
auditing of manufacturing processes is privatized, records created as a result of the
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
audit that would have been subject to public disclosure become private records solely
by virtue of the third party’s nongovernmental status. Patients and physicians
affected by a device produced by a poorly operating manufacturer, certified as
acceptable by a non-government auditor, will not have a government agency such as
the FDA with the knowledge and information necessary to recall faulty products and
correct marketing failures.
Summary
At the heart of the debate on privatizing government audits of medical
devices is the issue of the role and function of government. This is not a novel
debate that began with New Public Management (NPM) or National Performance
Review (NPR), but one that swings back to the Jefferson-Madison era. What gives
the pendulum momentum is the imbalance created when citizens feel there are either
too many resources spent trying to regulate or provide services in areas where the
market proves itself to be more efficient or too little control and accountability over a
market-based system focused only on profit. When either situation occurs,
privatization offers a number of advantages and disadvantages as shown in Figure 3.
Evidence in the literature suggests that there are selective benefits associated with
privatizing regulatory audits. Shifting this responsibility to third parties may result
in a smaller more cost-effective government where resources are devoted to the core
business goals of an agency. Audits performed by more flexible, skilled, and better-
equipped private auditing entities provide incentives for innovation by the
manufacturing market. As the EPA study shows, privatization can offer the elements
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of cost efficiency that government seeks—reduced production costs, ability to focus
on more critical or specific tasks, and increased capability to audit more of the
industry that government cannot afford to cover. A focus solely on cost efficiency
compromises accountability and increases the uncertainty that the public is protected
from product risk. Seeking a solution to this dilemma may tilt the scale away from
any cost savings. This suggests that the efficiency problems faced by the
government may not be resolvable by shifting auditing functions to the private
sector. In “The Irony of Privatization,” Miller and Simmons (1998) state that the
unique constraints on government, such as public scrutiny, restrictive laws, problems
of scale, complex processes, and the service character of much of the public sector’s
output, complicate public sector productivity and result in cost increases that exceed
those in the private sector. Additional managerial time and resources in the form of
paperwork, oversight capacity, and other transaction costs are needed to preserve the
principles of democratic governance. If not preserved, the “disinvestment in the
capacity of government to perform the functions citizens expect of their government
is a high risk strategy where the negative consequences tend to be cumulative and
difficult to reverse” (Moe, 1996).
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ADVANTAG E DISADVANTAG E
Increases Decreases
Efficiency
P
R
1
Responsiveness
Cost Savings Equity
Increased coverage
V
Transparency
Improved innovation
A
Government
Incentives for quality
T
involvement
service
1=0
I
C =i
Protection against
Better service quality Z unknown
Specialized resources
A
Active intervention
Market regulation
T
I
A
Fairness
Impartiality
o
Public involvement
N
Morale/attitude
Capacity
Figure 3. Advantages and disadvantages of privatization.
The Role of Government
A brief historical perspective on the changing role of government in response
to the efficiency challenge provides a backdrop for the discussion of advantages and
disadvantages of privatizing the audit of medical device manufacturing practices.
The underlying tension when deciding on the most efficient mechanism for
minimizing risk to public safety from medical devices flows from the ongoing debate
about government as an efficient producer and provider of public service. German
sociologist Max Weber’s concept of an ideal bureaucracy—the impersonal and
unbiased obligations o f administrators, the hierarchy o f offices and roles, and the
defined sphere of competence—supplemented with expectations of accountability to
an elective body representative of the public resulted in the belief that government
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
provides the greatest degree of efficiency in the conduct of public safety. Initially,
bureaucracy’s protective responsibility generated little focus on the cost and
th
efficiency for regulatory audit functions. In fact, as early as the 19 century,
government inspections were thought to be most cost effective for a number of
reasons. First, government sanctioned inspections helped to minimize the cost to
consumers of learning about the quality of products from companies with which they
may potentially do business. Second, government had the capacity to inspect a large
number of goods and services in a uniform manner that certified product quality
thereby making multiple inspections by local entities unnecessary, particularly for
products that needed to travel through several hands within a short period of time
(Telser & Higinbotham, 1977). Third, federal government involvement in inspecting
product quality served to resolve conflicts among localities with different interests.
Finally, government inspection programs provided uniform costs and inspection
expertise to all producers wanting to enter the competitive market. The prevailing
concern was for protection against physical harm or danger against fraud and abuse
and for enhanced accountability and trust.
The monopolistic control of traditional bureaucracy and its aversion to risk
allows for the development of detailed rules and before-the-fact controls that dictate
the behavior of producers and manufacturers. Legislated quantity and quality
controls have been quite common throughout American history. Today, these
controls also make the settlement of disputes cheaper by either fixing certain
dimensions of quality and quantity or establishing the means of their measurement
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Olds, 1999). These uniform standards of quality set the boundaries for market
behavior. When there is a natural inclination to disrespect the law, disobey the rules,
violate regulations, particularly if the expected benefit from violation (i.e., the
avoided cost of compliance) is greater than the expected cost of violation
(Kunreuther et al., 2001), an appropriate role for traditional bureaucracy is to
monitor performance through audit, enforce compliance with rules, and apply
sanctions when performance does not meet set standards.
Proponents of traditional government control contend that government-
sanctioned auditing represents the people’s consent; and, lacking that representation,
the market-manufacturer relationship driven by business incentives creates distrust
and uncertainty. Government is not only in the best position to consider all of the
interests of society and, therefore, command the private sector to produce optimal
results, but traditional government is also best suited to identify and implement the
best public policies regardless of cost (Kennedy, 2001). Prior to some of the more
recent thinking on government roles and responsibilities, economists viewed public
goods and protection, such as national defense and stability, policy making,
regulation and enforcement (Savas, 1987), as a public sector responsibility because
the lack of incentive to pay for the services by those who cannot be excluded from
using those goods would not provide the necessary profit to entice market
involvement. Government functioning as the regulatory oversight body protected all
of the public—the citizens and the market.
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Uche (2001) contends that regulatory involvement or intervention, whether
explicit legal control or formal peer group control by government, has always been
important to economic and social prosperity because it provides an umbrella of order
for the general public. Government and the regulations that provide this “umbrella
of protection” are frequently viewed as paternalistic in nature because serving the
interest of the public at large results in a perception of unnecessary or overreaching
involvement and control. A classic example is the government-mandated seat belt
laws premised on the belief that avoidable automobile injuries that result from the
failure to wear a safety belt increase medical costs that are borne by the taxpayer
interested in reducing such costs. Opponents of traditional government contend that
paternalistic government is unnecessary and that private markets, even with their
share of uncertainty and imperfections, provide an incentive for continued
improvement and increased public welfare by offering greater flexibility and choice.
Any successful efforts to improve government performance and increase individual
welfare must make greater use of market forces.
The Rise of New Public Management
As the overall scope and size of government expanded dramatically during
th
the 20 century, disillusionment with the ability of an over-controlling government
to respond to the needs of the public it represented created a strong negative public
opinion. The enormous size of the public bureaucracy and its inefficient
performance and service quality had fallen short of public expectation. Public safety
above all cost gave way to criticism of the increasingly exorbitant cost that seemed
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to provide less security of safety. For example, concerns heightened when
responsible agencies, such as the EPA and the Nuclear Regulatory Commission
(NRC), were perceived as inefficient in protecting against hazardous waste removal
and the release of health-threatening contaminants. Citizens accused the EPA and
the NRC of infrequent and inadequate inspections of the industry, while the
inspected industry levied charges of overly intrusive, costly and burdensome
regulatory requirements and the lack of a technically sophisticated inspection force
to evaluate the industry (McNulty, 1999). Opponents to fixing the bureaucratic
inadequacies through increasing staff realized what Virginia, New York, and other
states learned in the mid-1800s—the escalating cost of the inspection program was
due to the payment required to maintain inspectors and inspection facilities, that
reduced inspector availability limited the inspections that could be conducted in rural
areas; and somewhat less obvious, the cost to innovation due to the inflexibility of
inspection standards was excessive (Olds, 1999). Rigid government standards
forbade useful modifications and made experimentation, upon which progress
depended, illegal. In other words, inspecting against outdated government standards
established a static level of quality that hampered the advancement of technological
change (Lewin Group, 2001).
Disappointment with an inefficient government and the very elements that
once were used to praise its virtues resulted in a profound shift in the opinion
regarding the role government plays in public management. The NPM movement
sought to reform government involvement in and control over public affairs.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Believers in NPM argued that government and public bureaucracies had grown too
big and too costly to maintain. Some argued that market mechanisms offered
rewards for efficiency and cost reductions that did not exist in a public sector
(Downs & Larkey, 1986) where a monopolistic or semi-monopolistic character made
it unlikely to require a high level of efficiency. Under the banner of getting
government to operate like a business, NPM sought to use private-sector and
business approaches in areas of the public sector once considered the responsibility
of government’s (Pollitt, 1990; Kaboolian, 1998; Denhardt & Denhardt, 2000)
reasoning that competitive alternatives to auditing services would reduce government
cost. Advocating for a cutback in management, downsizing, streamlining,
privatization, contracting-out, and deregulation (Gore, 1995), the NPM theory argued
for a minimalist role of government and a greater role for the private sector that they
believed to be more efficient in providing quality services and products because
market mechanisms force competition in order to survive. Management reform in
public agencies, spurred by the publications Reinventing Government (Osborne &
Gaebler, 1992) and Banishing Bureaucracy (Osborne & Plastrik, 1997), drove the
concepts of NPM beyond the themes of using market forces and competition to
increase efficiency and reduce costs to one of stripping agencies down to their core
tasks and eliminating peripheral functions. The NPM concepts would not propose to
eliminate the government’s role in the enforcement of the regulations, but could alter
that role by shifting the responsibility for developing standards of quality and
performance and the auditing of compliance to those standards to private entities.
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Believers in NPM reasoned that relieving government of this responsibility had
minimal impact on market accountability because the efficient operation of any
market requires a clear set of rules that guide expectations and encourage people to
reach mutually beneficial agreements. Many NPMers would agree that established
market consensus on the rules and guidelines was more critical than who established
the rules. Opponents would argue that the basis of those agreements are financial
and, because the market lacks the necessary impartiality, the government’s role of
defining the general rules that establish the boundaries within which the markets are
allowed to operate and that resolve disagreements about the application of these rules
is critical. Government development and implementation of safety, inspection, and
licensing regulations has usually been justified by pointing to the need to guarantee
that certain products, such as beef and implanted medical devices, meet minimum
standards of public safety.
Over the past twelve years, the application of NPM theories to regulatory
auditing has swept the nation and the world as public managers have concentrated on
privatizing auditing responsibilities by restructuring bureaucratic agencies,
establishing new processes for enhancing effectiveness, and decentralizing decision
making. Coined by Lane (1994) as DPM—deregulation, privatization, and
marketization—this framework challenges not only how device safety oversight is to
be judged, but also what audits should be produced by the public sector. This
efficiency approach of shrinking government participation and making
administration more efficient through the use of the private sector and private sector
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
performance-management techniques has been the driving force behind the market-
type reforms in auditing manufacturing quality for medical devices in Great Britain,
Canada, and Australia, and in the call for reforms in the FDA’s inspection oversight
of medical products.
After two decades in which the United States and the world economy has
been dominated by deregulation, privatization, and faith in the market-based system,
the debate about the “place of government” may be changing or “entering an
historical watershed” (Mallaby, 2003). The author points to evidence of a broad
disaffection with pro-market reformism that runs from disappointment at
privatization and marketization in Russia to large-scale scandals in the United States
(California energy crisis, WorldCom). Evidence of the change includes a return to
calls for increased government or regulatory involvement and control (Sarbanes-
Oxley Act of 2002) as a result of disappointment in corporate governance in such
areas as the transfer of airport security from the private sector to the government;
government surveillance of risks from cell phones (regulated by FDA), and
strengthened government control over dietary supplements. Even amongst those
who uphold the NPM belief that less government is best government, increased and
successive corporate scandals that demonstrate that weak oversight tempts market
control mechanisms to forsake professional ethics may prompt a retreat from the
push for strict market control. Mallaby contends that the type of regulation that
forces the firm to produce public goods like cleaner air or safer medical devices
without making provisions for government’s watchful eye is failing because of the
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tendency to exaggerate the private sector’s virtues. Left to their own devices,
corporations do not necessarily seek profits by virtuously inventing new products
and more efficient ways of making them; they also seek profits by employing low-
cost manufacturing techniques and processes that operate at the margins of safety,
colluding with one another to set elevated prices, defrauding customers, and
lobbying the government for favors. Despite assumptions that markets facilitate a
smaller, more efficient public, the prospect of compromising safety for profit may
reverse the shrinking of government control and production. The cause may not be a
matter of political inertia or the public’s incurable low tolerance risk, but a reflection
of the market’s limited ability to function without some form of government
involvement.
The New Public Service Alternative
The role of government is not merely to control the actions of the
manufacturing industry through regulation and decree (traditional bureaucracy) nor
is it simply to establish guidelines and incentives (NPM), but it includes providing
flexibility through which the market determines its own level of compliance.
Government is a substantial component in the process of building integrity and
responsiveness into public health protection, but it cannot accomplish this goal in
isolation. Proponents of the New Public Service theory (NPS) suggest a role for
government that focuses on greater responsiveness to public expectations through
alternative approaches to managing public regulatory processes in ways that are less
dominated by issues of authority and control.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Robert and Janet Denhardt (2000) suggest that government is no longer in
charge, as the policies that guide society today are the outcomes of a complex set of
interactions involving multiple groups and multiple interests that ultimately combine
in unpredictable ways. Government would, therefore, act in concert with private and
nonprofit third parties to seek solutions to the problems of medical device safety and
reduced risk. Where NPM moved away from safeguarding administrative discretion
under traditional bureaucracy and toward trusting the market to deliver more
efficient and cost-saving audits, NPS moved toward increasing non-government
involvement while placing government in a greater accountability role for the
conduct of institutions and the development of standards, a role that protects public
interest and government responsibility for enforcing statutory and constitutional law.
Government functioning under traditional bureaucracy responds to the need for
ensuring reduced manufacturing risks by saying “yes” to providing all auditing
services. In contrast, the NPM government response to the request would shift that
service provision role to market mechanisms or third parties. A government
operating under an NPS philosophy would broker its involvement in the process by
bringing together the proper players and establishing the appropriate control
mechanisms to facilitate, negotiate, and develop solutions to the challenges that cost
and accountability bring in terms of control and production. In an NPS role,
government acts to facilitate solutions, but is responsible for assuring that those
solutions are consistent with the public interest in substance and process.
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Summary
As summarized in Figure 4, government involvement and accountability
changes as the philosophy of bureaucracy adapts to the political environment. In
these roles, government can function along a continuum of greater or lesser control
and responsibility in providing audits of manufacturing processes. The level of
control and production exercised under these modes of government operation is also
related to efficiency costs and accountability. When government fully controls the
process, the associated costs and the resulting accountability are likely much higher
than minimal levels of control and production.
Philosophy Government
Involvement
Accountability
Bureaucracy Government is
best qualified to
determine needs
and methods to
protect the public
Government has total
responsibility for
protecting the public
and bears total costs
Public servants
accountable to
government,
oversight bodies;
information
publicly owned
and is accessible
New Public
Management
Market forces best
suited to deliver
services; citizen
and service users
play a bigger role
in determining
appropriate
interventions
Government
underwrites cost
through contracts;
users bear cost
through fees imposed
for service
Principle-agent
relationship;
contractor
accountable to
contract holder
(government) or
to purchaser of
services
New Public
Service
Shared
responsibility
between
government,
market, and
citizens
Government acts as
facilitator for services
delivered
Government
participates in
nongovernmental
organizations
with
accountability
Figure 4. Changing role and responsibility of government.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
While the three models discussed above focus on broad regulatory
responsibility, the following discussion provides additional detail on the benefits of
privatizing these auditing responsibilities and the accountability concerns that may
arise if auditing is privatized.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER III
MODELS OF THIRD-PARTY REGULATORY AUDITS
Introduction
For several years, regulated industry, consumers, and elected representatives
criticized the traditional command-and-control methods of minimizing public health
risk through the monitoring of compliance to regulatory requirements. Chided for
their dated and archaic processes, unlimited scope and coverage, limited capabilities
of inspectors, unbalanced treatment of compliant and recalcitrant firms, and general
ineffectiveness, government continues to shoulder growing criticism in the face of
new challenges of regulating cutting edge technology produced by global firms in
global settings.
A series of new nongovernmental systems for monitoring and enforcing
standards have emerged over the past 10 years. These systems are designed to
strengthen the capability of the manufacturer’s auditor as well as develop
international systems of auditing that operate outside the nation-state (O’Rourke,
2003). The International Organization for Standardization (ISO) family of quality
conformance standards is perhaps the most popular and well known example. In
most cases, these nongovernment systems employ third-party auditors to assess
conformance to safety standards. Belke (2001) described the Environmental
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Protection Agency (EPA) and the Occupational Safety and Health Administration’s
(OSHA) successful use of third-party organizations in auditing the implementation
of the 1990 amendments to the Clean Air Act. A number of states, including
Delaware and Pennsylvania, also used nongovernmental organizations in their
monitoring of the implementation of environmental risk management requirements
for chemical facilities (Banish & Antoff, 2000; Collins, 1997; Halpem & Splain,
2001). Studies describing the efficiency and effectiveness of nongovernmental
mechanisms for monitoring compliance with labor, financial and food processing
standards also describe the usefulness of these parties to complement existing state
regulatory systems. In essence, the literature provides general examples of
governance systems that monitor compliance to regulatory standards designed to
reduce risk.
There is little in the literature, however, describing or evaluating alternative
market-based mechanisms for minimizing risk to the public through the audit of
compliance to quality manufacturing standards and regulations for medical devices.
Given the pressure to develop alternatives to the Food and Drug Administration’s
(FDA) system of monitoring the manufacturing of quality medical products, the
purpose of the discussion in this chapter is to identify and analyze current models in
operation using the efficiency, accountability, control, and production parameters
discussed in Chapter 2, and to provide insight into the implications and potential
impact these transformative models may have on the U.S. system.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
There are a number of reasons why this discussion is focused on the
European model and the variations of that model adopted by other countries. First
the alliance of European nations is the largest block of medical device manufacturers
in the world. The impact of the European model on other countries is so great that it
is rapidly becoming the framework for regulatory auditing around the world.
Second, these market-based models represent a significant change in the traditional
role and function of government. They suggest new forms of collaboration with
firms and nongovernmental organizations, provide models that enable the FDA to
respond to Congressional concerns about cost savings, and bring the U.S. monitoring
and surveillance structure in line with other nations. Third, the mounting pressures
on U.S.-based multi-national companies to comply with the European marketing
system bring pressure to the FDA to alter their governance structure. Fourth, these
alternatives serve as a yardstick to measure cost efficiency and accountability
concerns.
Auditing processes in Europe, Canada, and Australia are specifically chosen
because each had a relatively large and established system prior to the change, but
chose differing ways of operating in response to the need for change. Most
important, their new governance structures closely parallel the three ways
government operates to become more cost efficient—traditional bureaucracy, New
Public Management (NPM), and the New Public Service—in response to the
dissatisfaction with efficiency and accountability of government services. At one
end of the spectrum is the FDA inspection program that functions primarily as a
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
traditional bureaucracy. At the opposing end, Europe made the most drastic change
with the implementation of a market-based system that has a true minimalist
approach (Durant, 1998). Both the Australian and the Canadian systems operate
between these regulatory audit bookends. Except for the FDA model, these new
forms of governance are very young. They were established and are functioning
without the benefit of experience or research that examined their effectiveness or
accountability. Because there is an insufficient amount of information on the
operation of these two systems, there is minimal focus on actual model cost in this
comparative analysis. This exploratory examination will focus on how each system
operates, the role of government in the manufacturing audit process, efficiencies
expected from each model, and the accountability mechanisms expected from the
public.
An International View on Regulatory Audits
Many nations around the world regulate medical device quality in some
manner, and no two regulatory systems are exactly alike. This is largely because
regulatory systems reflect national traditions of law, culture, and economic
resources. When questions concerning the efficiency of government structure and
the resources needed to maintain a perceived inefficient and inflexible government
swell, the traditions of law and culture are challenged, and governments change laws
to accommodate the manner in which they operate. The shift in philosophy toward
non-traditional ways of regulating manufacturing quality appears to align with
changes in concepts of the role of government and the cost that government incurs in
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the regulatory oversight process. A summary of the roles and responsibilities of
government based on three operating philosophies, as presented in Chapter 2, frames
the discussion of the approaches that the other governments use to regulate medical
device inspections. A traditional bureaucracy assumes that government carries the
full responsibility and cost for public health protection through the implementation
of regulations and policies for which it is directly accountable to the legislature. The
market-driven NPM allows the market to deliver goods and services based on public
need and demand. In many cases, governments perform a secondary role as
contractor. When the responsibility for delivery of public goods and services is
shared, government under the NPS philosophy facilitates the relationship between
market and citizenry. The changing relationship of government and market in the
provision of goods and services provides models for nations seeking to reduce the
cost and size of government.
Internationally, governments are reducing unilateral authority and breaking
their monopolistic hold on the delivery of services, such as auditing for regulatory
compliance. Problems presented by new technologies, staffing shortages, financial
stress, and the perception of government maladministration force many nations to
embrace many of the major tenets of the NPM theory. Nations such as the United
States and Canada have long operated according to the traditional bureaucratic model
where the federal government had sole authority and responsibility for protecting the
public from problems caused by or related to the use of medical devices. Major
device-producing nations in Europe, such as the United Kingdom and Germany,
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
followed a similar philosophy until increasing concerns about the negative impact of
financing a large government on the country’s economic viability led to a
fundamental shift in the philosophy of government’s role in securing the safety of
medical products through the inspection process.
Since early 1980 when Britain, France, Italy, and Japan experimented with
limited attempts at transferring functions to the market sector, privatization quickly
became a major tool for accomplishing government responsibilities. Privatizing
audits for quality manufacturing of devices became a key issue in the United
Kingdom during the 1980s with Margaret Thatcher’s Conservative government and,
as codified in their Medical Device Directives (laws), became a cornerstone of the
new policy that governed the review of quality manufacturing in the European block.
The expansion of this market-based concept to an entire European community where
regulatory systems are in various stages of development prompted some authors to
state that global markets are pushing changes in auditing goals and perspectives with
forces that sometimes surpass those of the regulatory bodies within or outside of
their borders (O’Rourke, 2003). For example, in Canada, Japan, and the European
Economic Community, responsibility for developing public policies and standards
that ensure compliance with manufacturing quality standards lie beyond the realm of
one federal agency. Whole spheres of decision-making now belong to broader
structures that include private and non-government organizations. An example is the
network of organizations that work together to develop global standards for use by
government and nongovernment bodies to audit and report on medical product
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
manufacturing quality. The nations represented in this network—Australia, Canada,
Japan, European Union (EU) and the United States—form the Global Harmonization
Task Force, a consortium of regulators, industry representatives, and academia from
those five countries. In addition to developing quality standards for all types of
industry, this representative body also addresses significant global issues, such as the
development and monitoring of plans to address the containment of mad cow disease
and potential injuries from breast implants. This transnational collaborative effort
can be attributed to the alternative rule and decision-making processes of the
European System and others modeling themselves after the EU.
European Auditing Model
The change from an individual state regulatory scheme to a community of
member states that rely on third parties to regulate the quality of medical device
manufacturing is one of the most unique models of privatization. Using this
approach, the EU transformed the bureaucratic model of regulatory auditing that
existed in those individual states into a market-based monitoring system by adopting
a regulatory framework that sharply diminished government resources and
involvement in most critical decisions related to product safety. Created by the
Maastricht Treaty of 1992, the EU system was strategically designed as a mechanism
to harness the resources and increase the economic standing in that fledgling area.
As such, the governing laws and regulations for monitoring medical product safety
are rooted in economic principles. This approach of removing bureaucratic barriers
to trade across the Member states is noteworthy in its call for the private sector to
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
carry out practically all of the functions that were once solely the domain of the more
established regulatory systems in such European countries as Germany, the U.K.,
and France. Redesigned in order to increase economic trade, the outcome of this
policy change was to remove market access decisions from the individual national
ministries (Member State government bodies), which were viewed by business as
nationalistic and nontransparent, and to place these decisions with private sector
bodies that were believed to be more objective, transparent, and accountable to
comm unity-wi de commercial interests (Horton, 1995).
In the EU model, government seeks to increase efficiency by eliminating cost
associated with participation in the audit process and by focusing resources on the
broad oversight of product safety. The role of the European Commission (EC), the
overarching governance structure for the EU, reflects the philosophy of NPM, as this
adoption of market-based oversight requires new forms of collaboration, new roles
for nongovernmental organizations, new responsibilities of manufacturers, and new
responses from local (Member State) and national (EC) authorities (O’Rourke,
2003). The primary role of the EC is to orchestrate the development of governance
structures, regulations, and policies across the 25 countries. As the principle
government executive agency of the EU, the EC functions as the foci for developing
and issuing audit policy and regulations that must be agreed to and adopted by each
Member State. Medical device market entry and post market audit requirements
described in the EU Medical Device Directives (regulations governing medical
devices in the EU) are risk-based—a tiered approach that dictates the level of safety
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
data required in market submissions and the surveillance monitoring required for
products that present varying degrees of risk to the public.
The structural framework for regulating product safety calls for the
assessment of compliance with requirements by private organizations that are
accredited by private institutions that the government recognizes as being qualified
to train and assess the skills and capabilities of auditors. This assessment by the
private auditing body is required only if the product falls within the category of
highest risk devices. For example, high-risk devices such as implanted pacemakers
and heart defibrillators, require a review of the clinical and patient safety studies
conducted to show that the device can be effectively and safely used in humans.
Additional requirements call for product testing and a full audit of the manufacturing
quality system by a third-party assessment body. Lower-risk device manufacturers
are permitted to self-certify compliance of their manufacturing processes to quality
standards but are not required to conduct studies nor have products tested. Devices
that fall between the upper and lower risk categories may be subject to a third-party
assessment of their manufacturing quality systems and some finished product testing
against established standards. Not only is the assessment of manufacturing quality
performed by private entities, the development of the standards used by third parties
to determine conformance to quality is also a nongovernment task. While
government participates in the process, these standards are developed by consensus
between industry, government, and independent experts. The standards development
and acceptance process is owned by an independent group and is funded by sales of
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
standards documents to users. Prior to the passage of the new EU requirements,
compliance with these standards was considered voluntary. The recent governing
regulations issued by the Commission incorporated mandatory requirements for
compliance with voluntary standards for manufacturing quality.
The cornerstone of the European model is the use of these consensus quality
standards to which a manufacturer’s certification of compliance may be verified by
an independent third-party certification body. The EC reasoned that well-developed
and widely-accepted standards that provide criteria for managing and directing the
assessment of product development, performance, and monitoring could be
administered by private groups and, therefore, reduce government responsibility for
these. A Wallis and North (1986) research study shows that standards used to certify
dimensions of quality and performance reduce measurement cost and, therefore,
transaction costs to business. This is the principle behind the ISO certification
process that is rapidly becoming the basis for regulating medical devices around the
world.
Founded in 1946 in Geneva, Switzerland, this international standards group
oversees the development of common sets of standards for the manufacturing, trade,
and communication industries. The ISO 9000 series of standards, published in 1987
and revised and reissued in 1994, embodies quality management ideas and external
quality assurances. Developed by a worldwide federation of national standards
bodies, they are by nature voluntary consensus standards that contain quality
principles widely accepted by experts around the world. Regulatory bodies exercise
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
little control over this development process. Agreement on technical and managerial
standards for manufacturing quality is made by a combination of industry and
government representatives from multiple nations with differing perspectives on the
definitions and processes for determining quality manufacturing. In a consensus
process such as this, government control over the content of these standards is
minimized because a single vote by a regulatory body on a multi-interest
international standards committee is diluted by the representation of the group.
These ISO standards do not specify how a company runs the business or how
business should implement these quality elements; they only define the critical
elements that must be considered in the production of a quality product. A
manufacturing system receives an ISO certification when the organization’s process
system conforms to the standards as determined by the third party or registrar’s
application of the standard. In the EU, obtaining an ISO 9000 certification is similar
to obtaining a regulatory seal of approval for marketing products. Investment in the
ISO 9000 certification process is not a trivial expense, but is a costly endeavor borne
solely by the manufacturer. The cost of obtaining certification ranges from $20,000
for firms with annual sales less than $11 million to over $500,000 for firms with
annual sales over $1 billion (Barnes, 1998). While the certification process may
seem like an unreasonable expense, particularly for small business, ISO certification
provides market value. In their study to determine why companies pay the cost of
pursuing ISO certification, Gupta and Zhender (1994) found a positive relationship
between a firm’s value and the attainment of a regulatory seal of approval. Elmuti
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(1996) showed that manufacturers found that certification improved communication
within the organization by providing a clear understanding of targeted requirements
and resulted in an increased market share, higher selling prices, reduced costs, and a
higher stock price. Docking and Dowen (1999) concluded that there are six top
external and internal benefits that motivate firms to obtain ISO 9000 certification:
• Higher perceived quality;
• Perceived competitive edge;
• Reduced customer quality audits;
• Improved customer satisfaction;
• Increase in market share; and
• Quicker access of products to market.
At the time of their study, they found no evidence that companies seek ISO
registration because of client demands or competitor rivalry. Perhaps the most
pressing reason that manufacturers pay for audits is the EU mandate that medical
devices receive the ISO certification or “CE” mark before being sold in Europe
(Anderson, Daly, & Johnson, 1999). Many other nations have adopted this system
and now require a CE mark for importation.
In addition to the minimalist role of government in setting standards for
behavior, management, and assessment of manufacturing quality, the EU model
provides for little government involvement in the audit itself. Figure 5 shows how
the process for auditing medical device manufacturing quality is accomplished in the
EU. The EC governing responsibility includes developing criteria and coordinating
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
European Community (EC)
negotiates requirements, policies and
processes with regulatory bodies
(DA) of Member States
I
DA recognizes or approves accreditation
bodies (RAB); removes product
presenting immediate harm to public
I
RAB accredits assessment bodies
(CAB); registers the quality system;
grants, suspends or withdraws the CE
mark
CAB hires, trains, oversees auditors to
performs third-party conformity assessment
Auditors assess manufacturing quality
systems’ conformance to ISO
requirements
Manufacturer covers cost o f audit;
responds to needed corrections
Figure 5. EU audit process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the policies used by those Member States that qualify private auditing bodies.
Regulatory authorities in each Member State, known as Designating Authorities
(DA), use the ISO-based criteria established by the EC and agreed to by Member
States to recognize the capability of an independent accrediting body or Registrar
Accrediting Body (RAB) to certify auditors based on ISO criteria. Certifying
organizations called Conformity Assessment Bodies (CAB) hire or contract with
independent auditors to conduct manufacturing audits. Those CABs accredited by an
EU participating nation’s RAB are considered equivalent in skill and capability
across Europe as is the CE mark granted by the RAB. The process for gaining
certification is straightforward, beginning with the common procedures for the CAB
audit process that are provided in the ISO documents. During the certifying
assessment, ISO certified auditors check for compliance with ISO standards. All
communication related to the audit—findings, responses to findings, and corrections
made—occurs between the CAB and the manufacturer. There are no mandatory
requirements that the CAB or the RAB report audit findings to the regulatory
authorities. A manufacturer may voluntarily notify the DA of manufacturing non
conformances and product failures found by auditors. There are a few countries in
the EU—those with more mature regulatory systems—that maintained or enacted
mandatory reporting for deaths caused by medical devices. In a subsequent visit, the
CAB assesses whether the manufacturer addressed any non-conformances found in
previous audits; and, if not corrected, the auditor evaluates the impact any
noncompliance may have on the viability of the CE mark. Failing an ISO audit is
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rare because the process itself is viewed as helping to bring products to market and
not scoring the manufacturers’ capabilities. Any non-conformances found are either
addressed as the audit progresses or addressed and verified at the next audit. For
products awaiting market entry, the CE mark is withheld until the auditor certifies
the process to be in conformance.
Recommendations for withdrawal of the CE mark for marketed products are
directed to the DA through the RAB. The CE mark is withdrawn only in rare
situations; and, to date, anecdotal information from regulatory authorities in the U.K.
and Germany suggest only a handful of withdrawals out of thousands of CE marks
issued. Responsibility for alternative enforcement mechanisms that address fraud,
improprieties, or consistent reports from the public of product failure belong to
individual Member States; as there is no overall accountability scheme devised for
this transnational system of governance. Each Member State also has the authority
to remove any marketed product that presents a public health risk, although the EC
criteria for removal are narrow and are subject to the interpretation of each individual
entity. Such was the case when a contaminated blood dialyzer used to filter
impurities within individuals suffering from kidney failure caused 15 deaths
collectively in Spain, France, and the U.K. The defective product was removed from
public distribution in each country only after extensive debate within the upper levels
of their governments.
Clearly, the minimalist approach to government involvement in the EU
model upholds the many claims of why cost efficiencies are realized when
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nongovernment organizations are involved in conducting manufacturing quality
audits: lower government program cost; quicker response to the manufacturing
community; availability of more specialized resources; and increased innovation
through the use of more flexible requirements. This European change in
government’s role for assuring safety of medical devices has significantly changed
the number of public servants available to oversee this responsibility. Member
States and the EC expend the majority of resources in the development of regulations
and policies, the recognition and accreditation of RABs and, to a limited degree, the
response to product performance reports (or complaints) from consumers and other
government bodies. Costs of regulating under this model are in staffing of oversight
functions because the Member States require few auditors and the need for travel to
facilitate the auditing processes is minimal. Neither Commission nor Member State
officials contribute to staff costs for auditing or program overhead because
government audits are not conducted. Larger countries with pre-existing auditing
programs like the U.K. and Germany reduced auditors and technical staff by
approximately one third to one half, with many of the lesser advanced systems
having as few as one staff person responsible for medical device issues in the DA.
Those remaining auditors are reprogrammed to standards development and other
functions. Countries with few device manufacturing facilities like Spain and Greece,
that may not have had government auditing programs prior to the formation of the
EU, have less sophisticated and immature systems of oversight and, therefore, fewer
staff with the skills and capabilities of carrying out those responsibilities. In a
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
discussion with an official representative from the EU Commission, the
representative estimated that the number of staff tasked with handling medical
devices is approximately three persons in the Commission, and 2 to 30 staff
collectively in the 25 Member States. Given this vast difference in experience
among the Member countries, it is reasonable to conclude that the government’s
accountability to the public for correcting or intervening in problems varies
significantly.
One could argue that the EU model provides the market with greater
economic benefit and the public with greater assurance of manufacturing safety
because every manufacturing facility that markets in Europe receives at least one
audit each year as a requirement of maintaining the CE mark. The principle-agent
agreement between the manufacturer and the private assessment body enhances the
timeliness of the audit and, likewise, facilitates the timely market access of products
because it provides manufacturers with access to auditors as needed and has the
potential to increase the availability of auditors specialized in the technology of the
product inspected. Private assessment bodies have the option of hiring or contracting
for expertise based on their frequency of need. With respect to the positive impact
this model has on innovation, two Lewin Group surveys found that the use of ISO
standards allowed for greater flexibility in the innovation of new products (Olds,
1999; Lewin Group, 2001). Their responses suggest that the convergence of industry
and academic experts in the development of standards and the requirement that
auditors maintain technical expertise in the products they are certified to audit
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
allowed for a more advanced movement of the technology than the rigidly
maintained requirements of the FDA and compensated for the lack of adequately
trained inspectors found in the FDA inspection system.
Those aspects of the EU system that create efficiencies also contribute to the
challenges of accountability discussed in Chapter 2—relationship between citizen
and government; legislative accountability; shift in public service function and
capacity; and increased uncertainty of where responsibilities lie. Government’s
move from active surveillance to passive monitoring changes the dynamics of
leadership and responsibility for ensuring safe products. There is little doubt that the
requirement for frequent auditing of the manufacturing process by independent
assessors keeps a more trained and expert set of eyes on the process than government
has the capacity to supply; however, greater vigilance from a private assessor may
not equate to having the appropriate authority present to address the consequences of
unsafe manufacturing performance. Government is accountable for making sure that
standards of behavior and technical competence are adhered to in the selection of
organizations that hire auditors; but, in this EU model, government is not directly
accountable for ensuring the regularity, efficiency, or quality of the audit.
Qualification and certification standards under ISO address requirements and
restrictions on conflict of interest for registrars and auditors; but it can be difficult to
protect against fraud and abuse, particularly when the livelihood of the auditor
depends upon employment by that manufacturer who has a stake in keeping the
product marketable. In numerous meetings with former FDA inspectors who have
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
left the Agency and are now independent third-party auditors, they frequently
disclose that as private contractors, they function as well paid advisors to
manufacturers who have the latitude of accepting and addressing all, part or none of
the auditing findings, particularly if they are significant enough to affect the
manufacturers’ market standings or profit margin. Many claim that negotiations
aimed at altering the audit report include reference to loss of contract and future
employment.
The interaction between the certifying or auditing body and the manufacturer
and the option for voluntary disclosure of information also raises concerns about the
openness of the process and the ability of the public to obtain adequate information
necessary to make informed decisions. The EU audit model does not prohibit public
interaction with government but it does sanction the closed loop interaction and
information sharing between the certifying or auditing body and the manufacturer.
This lack of available information provides little opportunity for the government to
have a proactive role in preventing mishaps or an interventionist role in resolving
problems of product risk to safety. It significantly reduces public access to
information that can be used to hold the DAs accountable for a manufacturer’s
actions. The combination of the transnational nature of this system, which is a
composite of historically strong, weak and nonexistent bodies that regulate auditing
of manufacturers and the absence of a formal accountability structure (Harlow,
2002), also creates a lack of uniformity in how issues are handled. This can have a
dramatic public health impact if significant safety problems reported to the DA of
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
one Member State are the result of manufacturing deficiencies that did not surface
from an assessment body accredited in a sister Member State. With no clear line of
responsibility, confusion exists in both the public and the state regarding who is
responsible for the action necessary to mitigate the risk. The public is uncertain
about who is responsible for the action necessary to remove the risk, and the
regulatory body has little information available to resolve the uncertainty.
Canadian Audit Model
Responding to similar concerns regarding the inability of government to
provide resources for an auditing program to assess quality manufacturing of medical
devices and as a nation aware of the economic implications of trade with the EC,
Canada sought ways to modify their monitoring system. Confronted with the high
cost of government programs, the complaints from their industry regarding overly
burdensome medical device regulations and the need to compete in the international
market, the Canadian government revised its regulations in 1998 to reflect many of
the principles of the EU system. On January 1,2003, Canada adopted new
requirements specific to quality manufacturing under the new Medical Devices
Regulations. Less radical in terms of government involvement than the EU model,
Canada opted for a model where the government had a strong oversight role—one
that maintained its ability to control devices introduced or removed from the market.
Canada licenses over 15,000 medical device manufacturers annually for marketing
within its borders, with over 8% of those manufacturing facilities located in Canada.
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Medical Device Bureau under Health Canada’s Therapeutic Products
Directorate (TPD) has the responsibility to assure that products manufactured within
Canada’s borders or products sold to Canadian citizens conform to the requirements
of their law. Manufacturers must meet the regulatory requirements as established in
the government’s medical device regulations; however, the responsibility for
determining the extent to which those government standards are met no longer falls
solely under Health Canada’s purview. Risk-based in accordance with where and
how the device is used in the body, these regulations require that moderate to high
risk devices meet manufacturing quality requirements found in the standards
developed by the National Standards of Canada (NSC), which, like ISO, is a
representative voluntary standards development body.
Prior to the new legislation, NSC carried the majority of the responsibility for
setting standards of manufacturing performance in Canada. As a resource saving
under the new requirements, Heath Canada deemed NSC standards equivalent to ISO
standards for quality manufacturing in order to reduce the overlap in auditing
requirements. Manufacturers of higher risk medical products, classified as Levels II,
III and IV, must be licensed according to these standards before marketing. Similar
to the EU system, manufacturers of devices classified in the lowest risk category,
Class I, are not subject to any mandated quality audit; and there is no requirement for
verification of quality manufacturing performance by independent third-party
auditors. The licensing process does require manufacturers of medium risk Class II
devices to attest or self-certify that the quality system under which their devices are
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
manufactured satisfies the basic manufacturing requirements under the ISO standard
and to self-certify that higher class products are manufactured according to a more
thorough ISO quality system requirement. Independent third parties must also verify
these attestations.
Prior to the implementation of Canada’s revised regulatory audit policy,
assessing manufacturing quality was a government function conducted by a very
small staff of inspectors nationwide. Canada realized during the revision process
that 22 inspectors were insufficient to cover the 8600+ manufacturers licensed in
Canada that would require an audit. The government attempted to compensate for
relinquishing its auditing and monitoring role by developing partnerships with other
device-producing nations in order to increase the information on the safety of
products used by Canadians but manufactured in other countries such as the U.S., the
U.K., and Germany. In a move to devolve the government’s role in the regulatory
process and reduce the cost of government oversight yet maintain its competitive
place in the global market, Canada embraced a quasi-NPM philosophy in the new
regulation. Canadian officials wanted a government more involved than as a
bystander as perceived in the EU model, and focused their attention on the New
Public Service “builder of the boat” (Denhardt & Denhardt, 2000) philosophy.
Constrained by a national budget that prevented any resource increase for the
national regulatory inspection program, TPD decided that it would not conduct
inspections but would use a third-party registration and accrediting system similar to
Europe’s. Health Canada sought to build a coalition or network between
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
government, industry, and private interest (a strong contingent in Canada) to develop
a system that used market-based mechanisms for conducting many aspects of
government responsibility, but retained an appropriate level of government oversight
and control. Canada sought to balance efficiency concerns with accountability
(Harlow, 2002).
The Canadian model is designed to have government develop regulations;
provide expertise for the NSC and ISO standards development processes; institute
mandatory mechanisms for disclosure of information on manufacturing performance
and product quality; establish an oversight role in the accreditation and registration
process; and maintain its enforcement responsibilities. Canada achieves these goals
through their Canadian Medical Devices Conformity Assessment System
(CMDCAS) governance structure. This system describes the roles and
responsibilities of government, the Standards Council of Canada (SCC), and the
process by which third parties accredit and register device manufacturers and assess
their manufacturing quality. By building a coalition with the country’s leading
standards accreditation body (SCC), government serves as a key participant on the
umbrella board that oversees the accrediting of the CMDCAS medical device risk
management system, but does not totally fund or control this effort. Once they are
certified by criteria developed and monitored by government, private organizations
share responsibility for regulatory implementation by using the standards developed
through this joint venture to audit manufacturing practices.
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Canadian process of auditing for quality in manufacturing facilities and
the role of government expanded as a result of implementing CMDCAS. As shown
in Figure 6, communication between government and the manufacturer now includes
Registrars and third-party conformity assessment bodies. Registrars assume the
responsibility and cost for training and qualifying auditors. Audit costs are fee-based
and paid to third-party auditors by the contracting manufacturers. Health Canada
recognizes registrars (and auditors) accredited in the EU system; however, they must
also be accredited under the Canadian device requirements. Canada no longer uses
its auditors for routine determination of manufacturer conformance to the
regulations, but does retain a small staff of expert auditors who have the authority to
spot check the auditing capabilities of the third-party auditor or to investigate
medical device problems at a manufacturing site. Audit results are reported to TPD,
where manufacturing problems are trended across product types and government
actively communicates with the public on these issues. Health Canada has an active
recall and enforcement program so that problems can be addressed.
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Health Canada (HC) establishes
requirements, selects, trains, and audits
accreditation body (SCC) and Registrars;
take regulatory action
SCC accredits Registrars; corrects,
suspends, or withdraws accreditation
Registrar performs third-party
conformity assessment of manufacturer
quality system according to HC
requirements; registers the quality
system
♦
Registrar certifies system; if appropriate,
issues report to manufacturer and TPP
Figure 6. Canadian audit process.
It appears that key efficiencies of this model are realized by the reduced cost
of funding a standards development program and maintaining a large audit staff.
Prior to introducing market-based participation into the process, the appropriated
budget for TPD barely covered the cost of maintaining their small cadre of
experienced staff necessary to conduct the sporadic oversight audits that arose.
According to the Director of the Medical Devices Bureau of Canada, including the
use of third-party auditors increases the opportunity to achieve the goals of
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
improving product quality and reducing risk, because the presence of additional
auditors increases the number of facilities assessed at a lower cost to government.
Although Canada lacks the data to support the claim, the TPD has anecdotal
information suggesting that manufacturers also realize a shorter time to bring
products to market because they are not constrained by TPD’s inability to conduct
timely audits or their ability to maintain appropriately trained and specialized
auditors. Resources saved under the CMDCAS system allowed the program to shift
focus to new oversight responsibilities. Canada did not hire additional investigators
but used highly expert staff to serve with standards development groups, although it
would seem that the program cannot sustain the new responsibilities of evaluating
and monitoring registrars, periodically evaluating manufacturers, and trending
products where licenses or certifications are pulled due to manufacturing non
conformances without additional resources.
The primary concern of accountability for quality manufacturing under this
model rests not so much with the diminution of government’s regulatory
responsibility and control, but with the additional layer of governance between the
TPD and the manufacturer and the problems that fee-for-service present. Some
reduction in the public service objective of guarding citizens’ rights may occur as
Canada shifts from engaging in the basic service of direct monitoring of
manufacturing quality to focusing their energies on oversight of the accrediting
bodies. Canada instituted legislative requirements and processes, such as direct
reporting to government of audit results, to minimize the loss of information needed
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to take risk-reducing actions prior to or as a result of manufacturing
nonconformance. While these measures cast light on the transparency of actions
between the audited and the auditor, they do not eliminate the need for careful
monitoring and additional safeguards. Strengthening the oversight may not provide
the same protective shield as having direct and immediate connection with the
manufacturer, particularly in the intervention to correct real-time hazards, but it does
maintain a presence for Canada on the citizen-administrator-political-representative
path.
Based on news accounts and confirmation from Canadian government
officials, the citizens of Canada appear to be very similar to those in the United
States in their demand for government protection and intervention when product
problems arise. Political pressure for public accountability can be equally
demanding. In order to respond directly to citizens and their representatives in
government, Canada collects and maintains significant information on the audits and
their findings from which safety reports and press releases are issued when needed.
When problems arise, this system does allow for some public access to information
through government disclosure that would not be possible without the reporting
requirement.
Australian Audit Model
Recognizing the worldwide changes in landscape with respect to regulating
medical products, on October 4,2002, the Australian government implemented the
amended Therapeutic Goods Act of 1989. A number of significant events prompted
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
changes in what was then their first comprehensive national system that replaced a
mixture of federal and state regulatory responsibilities. Those changes included the
active participation of the country in an international effort to harmonize the
regulation of medical devices, the recognition of the EC approach to medical device
regulation as a de facto world standard, and an internal desire to address
shortcomings identified with the existing system. Timing of the Australian legislative
changes in light of the EU’s implementation of a strong market driven system also
provided an opportunity for the country to observe the potential problems with that
model. What seemed unique about Australia’s proposed changes was the decision to
move in an opposite direction of the one established in the EC. The minimalist
theories of lesser government involvement and more privatization of government
functions that formed the basis of the practices and polices of Great Britain did not
hold for Australia, who moved toward giving government increased oversight of
medical devices. Their new approach to regulating medical devices set out
additional requirements for market entry and provided broad power to implement the
requirements through administrative arrangements with penalty provisions as a last
resort.
Like the other countries discussed, the Therapeutic Goods Administration
(TGA) elected to use a risk-based classification scheme to stratify the levels of
requirements for assessing manufacturing quality. Unlike other countries, the
requirements are applicable across all product risk levels. For example, expressing
the belief that Australia had in general fewer and/or less stringent regulations for
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
many types of devices and, therefore, lacked adequate government oversight, their
legislature revised the 1989 Act to include the institution of minimal requirements
for lower risk products, a more comprehensive assessment of new technologies
before allowing them to be marketed, the extension of manufacturing quality system
requirements to more devices, a boosting of reporting requirements by manufacturers
when their products fail, and the adoption of ISO standards in assessing the quality
of manufacturing operations. Almost all of these new requirements would be
conducted under new powers given to the government. Where Canada and the EC
provided no audit of quality for low risk devices, Australia gave that segment of the
industry a five-year transition period to come into compliance with their new
regulations, including those that had been previously exempt from compliance with
quality standards. According to the National Manager for the TGA, the equivalent of
the Commissioner of FDA, Australia’s observation of the EU system raised a
number of concerns about the control over and accountability for the quality of
medical device manufacturing. Reliance on private intermediary bodies for the
evaluation and enforcement of safety, quality and performance of the $130+ million
domestically consumed and $590+ million exported products the Australian industry
produced in 2001 was discomforting to the Australian government.
While many countries focused on reducing staff costs, Australia’s relatively
small device auditing program of 16 staff would increase to accommodate the
additional oversight responsibilities. To offset these increased costs, the new
legislation also included a provision for assessing fees for audits that were payable to
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the government for this service. Not only did the new system increase the number of
devices to be audited, it called for an audit program totally financed by user fees.
Staffing for audits is not dictated by the national government budget, but by the
industry need for and use of auditors. Although the regulation established a user fee
program to cover the audit costs, it did not provide full funding for administering the
program. As demands for audits increase, associated transaction costs will also rise.
Australian officials are uncertain about the impact that increased monitoring and
other oversight programs that need to be established will have on the budget, but
there is some anticipation that the existing staff is insufficient to cover all of the
additional responsibilities not covered by audit fees.
As the EC moved further in the direction of giving third parties complete
control in the assessment of manufacturing quality, an approach that is more
independent of government than any other major nation, Australia moved in the
opposite direction. Taking a firmer stand on what the government’s role should be in
the protection of public health and safety, the TGA proposed that medical device
conformity assessment and auditing functions in Australia be performed by
government bodies. The new process, shown in Figure 7, illustrates the direct
responsibility TGA assumes for auditing and conducting all interactions with their
domestic and foreign regulated industry. Auditors are hired, trained, and accredited
by TGA to audit; and the information from those audits is transmitted directly to the
government.
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TGA hires and trains
investigators
Investigators inspect and submit
report to manufacturer
t
Manufacturer responds to TGA
and submits predetermined
inspection fee
Figure 7. Australia audit process.
In response to suggestions by the Australian private sector that the
government adopt the EC system that uses accredited public and private bodies to
undertake medical device conformity assessment to Australian requirements, TGA
disagreed, contending that the protection from any potential risk these products
present to the citizenry is a public health responsibility that the government could not
transfer. The TGA publicly stated that the potential risk posed by medical products
required that the government act in the interest of public health to provide the
independent scrutiny necessary to ensure the quality, safety, and performance of
medical products available for public use and consumption. The TGA did offer to
reassess their position on third party audits after having an opportunity to monitor
and assess the progress o f the EC third party system on such issues as safeguards
against improper influence and appropriate areas for eligible third-party
involvement.
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
On the surface, it appears that efficiency in terms of cost reduction was not
the primary concern of this model. This model seems to support the concerns
expressed in the literature that cost reduction efficiencies should not overshadow
concerns of effectiveness and representativeness. Under this new structure, change
in accountability increases rather than decreases the citizen-govemment relationship
and the public service protection responsibility, as TGA is more accountable for
broader oversight and is not hampered by the potential accountability gap that results
from having nongovernmental auditors involved in the process. The level of
uncertainty for responsibility is reduced because the direct line between the
manufacturer and the enforcement group is not hampered by the inability to disclose
information, the adequacy and correctness of information on which to take
enforcement actions, or any improprieties caused by conflicts of interest.
The model does face many of the challenges to efficiency that traditional
bureaucracies face, except the production costs associated with auditing will not be
borne by government. Since Australia participates in the development of standards,
the problems of technical inflexibility and the high cost of a rigid government-
controlled, standard-setting process is eliminated. Australia’s flexibility to increase
staff and expertise should have little impact on government costs; however,
increased government responsibility for and control of transactions not covered by
user fees could cost considerably more than the existing program costs or more than
the collected fees would cover. The fee-based nature of this program allows
Australia to hire the needed experts; but, if hiring is permanent, retaining these
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
experts may incur cost increases if the user fees do not address this shortcoming.
Because of its inexperience with this system, Australia was unable to provide
estimates on the cost of increased staffing or overhead.
FDA’s Regulatory Inspection Process
In order to compare the previously discussed alternative audit approaches to
the FDA inspection process, this section presents a discussion of the current FDA
model. The FDA uses the term “inspection” to describe its process of evaluating
manufacturing quality. Therefore, in this section, the term “inspection” will be used.
Functioning in a customary bureaucratic mode, the FDA carries the nation’s
weight of protecting the public from medical product manufacturing mishaps. Large
segments of the regulated industry also recognize government leadership in this area.
In a 2001 report, entitled The Impact o f Regulation and Market Dynamics on
Innovation, published by The Lewin Group, the authors state:
The single most important gatekeeper for technological
innovation in health care is the U.S. Food and Drug
Administration.. . . The FDA has considerable responsibility
for regulating new and existing marketed technologies subject
to its purview in a manner that not only protects and promotes
the health of the public, but that facilitates ongoing
technological advancement and timely patient access to
proven new technologies. FDA establishes and implements
evidence requirements and other aspects of medical
technology regulation that affect, directly and indirectly, the
frill technology lifecycle, from product concept through
marketing approval and post market surveillance.
A central part of the FDA’s responsibility to protect the public health is the
authority given in the Food, Drug and Cosmetic Act of 1976 (Act) to assure that
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
medical devices and radiation-emitting products—both domestic and imported—
meet those safety and quality manufacturing standards that make devices acceptable
for use in the United States. In addition to reviewing submissions that contain
technical and scientific evidence justifying the safety and effectiveness of these
medical devices, this government body satisfies its responsibility through inspecting
and monitoring device manufacturing so that potential hazards are identified and
corrected in time to prevent or minimize public exposure.
Congressional support for this role is evident by the legislation it continues to
pass. Figure 8 shows that there has been no legislation passed since the original
authority given to the FDA in the Medical Device Act of 1976 that challenges or
reduces the Agency’s responsibility in investigative oversight. In fact, a number of
Congressional hearings took the Agency to task for missed opportunities to perform
its inspection functions adequately. Concerned that the FDA was not aggressive
enough in its proactive investigation of problems with heart valves and respirators,
Congressman John Dingell sponsored the Safe Medical Devices Act of 1990 that
added responsibility for investigating product recalls and reporting medical device
mishaps in user facilities. Calls for the reduction of big government and excessive
regulations during the Clinton Administration resulted in the FDA developing a more
concise inspection procedure and adopting a “kinder” inspection approach.
Influenced by a few prominent manufacturers, Congress initiated debates on
privatizing inspections, but there was little support to make this change in the Act of
1997. Over the past five years, the medical device industry’s interest in changing the
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
involvement of the FDA in quality manufacturing inspections to mirror the
international process intensified the pressure on Congressional representatives to
pass legislation mandating changes in the Agency.
Congressional Legislation Areas of Responsibility
1938 Food, Drug and
Cosmetic Act
Regulates the sale of foods, therapeutic
devices (pharmaceuticals) and cosmetics
1976 Medical Device
Amendments to the FD&C
Act
Adds the regulation of medical devices to the
FD&C Act
1978 Good Manufacturing
Practices
Establishes standards for quality
manufacturing; biennial inspections required
for moderate- to high-risk devices
1984 Medical Device
Reporting
Requires inspection oversight to assure
investigation and reporting by manufacturers
to FDA of deaths, serious injuries and
malfunctions related to devices they
manufacture
1990 Safe Medical Devices
Act
Additional authority to recall devices
1995 Medical Device User
Facility Reporting
Requires user facilities to report deaths,
serious injuries, and malfunction of devices to
FDA
1997 Food & Drug
Administration
Modernization Act
Added inspection responsibility to cover
diminished premarket data requirements
Figure 8. Laws that increase FDA responsibility.
The size of the medical device industry and the innovate technology over
which the FDA exercises its regulatory authority has seen phenomenal growth over
the years. As the favorable economic climate results in the U.S. consumption of
internationally produced devices, the FDA’s sphere of responsibility expands.
Figure 9 shows the growth in registered manufacturing establishments over the past
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
six years. The U.S. responsibility for regulatory inspection oversight of a market
that in 1998 produced more than $68.8 billion worth of medical devices and
diagnostics products, employed more that 300,000 Americans and realized a $7.6
billion trade surplus in medical devices, while the overall U.S. trade deficit was $360
billion (Lewin Report 2001), is exclusively a government function. A large
component of this responsibility is oversight for quality manufacturing. The FDA
uses internal staff to conduct all inspections and develop reports on all
communications that occur between the FDA and the manufacturer. With the
exception of confidential and trade secret information, inspection reports are
accessible to the public under FOIA. All hiring, training, and certification as well as
overhead costs associated with the inspection process are provided through the
Agency’s budget.
Registered Manufacturing Establishments
(U.S. and Foreign)
21000
20000
19000
18000
17000
16000
15000
1998 1999 2000 2001 2002 2003
Figure 9. Growth of registered manufacturers.
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 10 explains the FDA investigative process. During an inspection,
FDA investigators are instructed to respond to questions and discuss deficiencies in
light of current policies so that the manufacturer can formulate a response. This
interaction can be positive for the industry. In a 1998 survey conducted by the
Medical Device Grassroots Task Force—an organization of device industry
executives who worked with the Agency to improve the relationship between the
regulated industry and the regulators—large and small manufacturers commented
that the value of an FDA inspection is in the education they receive from the
investigator during the inspection. Industry found the exchange more informative
than reading the requirements. The law requires that FDA assess biennially the
quality of the manufacturing process for moderate- to high-risk devices, designated
as Class II and Class III, respectively, that are surgically implanted or intended to
support or sustain human life and whose failure could reasonably be expected to
result in a significant adverse health event. Although lower risk product (Class I) is
not subject to the biennial requirement, manufacturers are required to comply with
the quality manufacturing requirements developed by the FDA and have documented
records available at the time of inspection. As in the more recent requirements
passed for the EU and Canadian markets, Congress assumed that lower risk product
required minimal oversight because the potential for injury was small. The FDA’s
experience with manufacturing problems of low risk, or Class I, devices shows how
important this monitoring requirement is regardless of the product risk class. For
example, inadequate manufacturing processes that allowed the used of tap water to
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
wash tubing used in the production of a kidney dialysis device resulted in the growth
of bacteria that caused a number of illnesses and deaths in 1995. Additionally, in
2000, an inappropriate sterilization and sealing procedure for an iodine swab, a
seemingly incidental component in a surgical suite kit, resulted in the recall of
hundreds of thousands of product costing millions of dollars in lost revenue, a costly
mistake for a number of manufacturers that include this inexpensive product in their
marketed kits.
FDA hires, trains, Investigators Manufacturer
and certifies
------ fe.
inspect and issue
.4—
corresponds
investigators observations to
manufacturer
with FDA
A
I
t
Manufacturer contracts
with consultant for
assistance with quality
system; assessment
given to manufacturer
Figure 10. FDA inspection process.
Total responsibility for the medical device inspection program also means
that the FDA carries the weight of writing regulations, guidance documents, and
standards that instruct its investigative staff and the regulated industry on compliance
expectations. Two government-developed documents, the 1978 Current Good
Manufacturing Practices (cGMPs) and the revised 1998 Quality System Regulation,
serve as the basis for all inspections conducted by FDA. The FDA participates
heavily in the development of ISO standards and allows the use of some ISO
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
standards in the development of documents that attest to product safety before the
device is marketed. ISO standards that are used internationally to audit for
manufacturing quality are not accepted in lieu of the FDA’s inspection requirements.
While the FDA procedures cover many of the same concepts as the ISO standards
for regulating the manufacture of medical devices, FDA inspection of these facilities
is more proscriptive in that they test the manufacturer’s application of the quality
system regulation and not merely the documents indicating the existence of a system.
This more in-depth approach to inspections raises concerns from manufacturers
about the time required to conduct FDA inspections and, because those inspections
can require more time, the FDA’s capability of providing adequate coverage of the
industry. A more in-depth inspection process would also seem to be more costly
because of the need for greater resources. As with the other models examined, the
search for data to conduct an analysis of the efficiency of this process in terms of
cost to inspect manufacturing quality systems proved unsuccessful. The FDA views
program costs in terms of the total budget allocated to provide product market access
or monitor post marketing performance and does not collect cost data based on a
particular program activity or a specific element of a program, such as the inspection
component of a post market monitoring program. It is difficult to determine the
actual cost of an individual inspection or for the program without those available
measures.
An examination of data retrieved through inspection describing FDA’s ability
to meet their inspection responsibility is provided in Figure 11, which shows that
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
between 1996 and 2003, the Agency covered on average 25% of the manufacturing
facilities requiring audits. The rate of coverage ranged from a low of 15% in 2000 to
a high of 38% in 1996. These numbers are well below the biennial requirement that
the FDA inspect 50% of the market in any given year. This coverage does not
include approximately 10,000 low-risk manufacturers and distributors that are not
required by law to receive an FDA inspection. More illustrative of the FDA’s ability
FDA Inspection Trends
3500 -----------------------------------------------------------------------------------------------------------------
3000 ----------- — ----------—.......................................— - - - - - - - - - -— ——------------
2500 - - - - - - - --------------------- -- - -- -—— - - - - - - - - - - - - -- - -—------------------
2000 -----------------------------------------------------------------------------------------------------------------
1500
□ Foreign
1000
FY96 FY97 FY98 FY99 FY00 FY01 FY02 FY03
Figure 11. Accomplishments of domestic and foreign medical device
inspections.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to provide adequate protection from risk is 2003 data showing that 1,200 moderate-
risk Class II device manufacturers and approximately 66 high risk Class III
manufacturers never received an FDA inspection. Attempts to increase the number
of inspections conducted by improving the availability of staff investigators have
been unsuccessful. The number of investigators dedicated to this task remains
around 150.
Under this model of full government involvement and control, inspections
may also be used to support other government responsibilities. Investigators may
obtain evidence to effect immediate correction or support legal actions when
violations are discovered. During an inspection, if an investigator observes a
violation that involves a failure by the manufacturer or sponsor to obtain market
approval, a manufacturing practice that is in serious violation of applicable laws and
regulations, or a product that poses a serious risk of harm to health, the inspector is
authorized to order one or more of the following actions: shutdown of the facility;
detention of medical device(s); physical destruction of medical device(s); or
termination of all shipments of medical devices until the FDA reauthorizes
distribution. These stringent measures are taken only under very serious conditions
and after being granted immediate but limited permission by the courts.
Complaints about the bureaucratic and “inflexible” application of these
standards, particularly when the manufacturer also operates in the less stringent
international environment, occasionally result in a manufacturer claiming an
inspection to be an invasion of his Constitutional rights. A manufacturer unfamiliar
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with his responsibility under the medical device regulations may refuse entry to the
investigator. Accountability to public health protection is so strong in this society
that the Supreme Court, in recognizing that inspections conducted in the interest of
public health protection do violate protection granted under the Fourth Amendment
of the Constitution against invasion of the privacy of businesses, allowed inspections
in three types of circumstances: permission by the owner or occupier of the
manufacturing site; emergencies that pose an immediate threat to health or safety;
and in large-scale protection situations such as when open eating areas of restaurants
might be subject to emitted pollutants (Gostin, 2000). These circumstances highlight
examples of how substantial public interest in minimizing risk justifies the routine
oversight of regulatory officials to achieve the objective of compliance with health
and safety standards. In this situation, if the FDA is refused entry to inspect, the
courts can issue an inspection warrant. Warrants are typically issued for criminal
investigations, however, courts will issue warrants for health and safety inspections
if there is specific evidence of an existing violation of health and safety standards
(e.g., existing history of inability or unwillingness to correct past violations), or if
there is a reasonable plan supported by valid public interest (Gostin, 2000).
Legislative changes to existing regulations bolstered the FDA’s
accountability for addressing failed product in the marketplace. Since 1997,
additional reporting requirements and investigations of failed devices allowed the
Agency to trend failures and recalled product to correct market failures. What is
most significant about the changes to the new requirements is the leverage the FDA
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
exercises with the legislature because of its ability to collect and analyze inspection
information from previous years on product problems and failures and
manufacturing problems across different devices—information that would be
difficult to compile under a voluntary reporting system. According to a 1990
Inspector General study examining device recalls over a six-year period,
approximately 44% of the recalls issued for moderate- and high-risk devices were
due to faulty design and the lack of design controls. The study commented that the
FDA was unable to resolve these fundamental quality problems through the
inspection of manufacturing processes alone. The report further stated that because
the market did not have the ability to bring about change to such a large and diverse
manufacturing group, even if they had a collective picture of the status of
manufacturing quality, the force of law was necessary to institute revisions
mandating better controls in the development of safer products. When government is
given these kinds of additional capabilities, it is able to trend problems across
manufacturers and product types and prohibit or correct products that fail
performance and safety requirements.
The FDA collects both general and specific information on faulty products
that results in recall situations requiring removal of products from the market or, in
cases where the product cannot be recalled, such as an implanted device, notification
advising patients to seek further medical advice. Recall statistics on the total number
of medical device recalls that occurred between fiscal years 1997 and 2001 across
the three risk categories is presented in Figure 12. Specific information on the top
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
six devices recalled between 1995 and 2001 is listed in Figure 13, which provides an
example of the ability given FDA to trend problems across manufacturers and
prevent or correct products that fail performance and safety requirements when
information is accessible. Many of these products are essential to patient care.
Ventilators, one of the top three recalled devices for five of the seven years, were
involved in 26 recall actions involving 10 different medical device manufacturers.
Other examples of multiple recalls involving multiple manufacturers include:
infusion pumps that resulted in 36 actions involving 19 manufacturers; blood
chemistry analyzers used in the diagnosis or monitoring of blood components, with
44 actions involving 12 manufacturers; and cardiac defibrillators used to
automatically respond and correct malfunctions of heart rhythms, with 19 recalls
involving 19 manufacturers. These examples are only a subset of devices recalled by
multiple manufacturers for significant problems.
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Med. Dev. Five Year Recall Statistics
1200
1000
600
400
200
19S7 1998 1999 2000
□ C lass! B Class II □ Class I
Figure 12. Five-year recall statistics.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
F Y 95 F Y 96 F Y 97 F Y 98 F Y 99 FYOO FY01
Linear
Accelerator
Tomography
System
Dialysate
Concentrate
Infusion
Pump
Chemistry
Analyzer
IV Starter
Kit
Ventilator
IV Catheter Ventilator Infusion
Pump
Ventilator Infusion
Pump
Ventilator Automatic
Implantable
Cardioverter
Defibrillator
Ventilator Fixation Rod Electrosurgery
devices
Dialysate
Concentrate
Fluoroscopic
X-ray
Chemistry
Analyzer
Infusion
Pump
IV
Administration
Set
ECG
Electrode
Automatic
Implantable
Cardioverter
Defibrillator
Surgical
Laser
Ventilator Infusion
Pump
Catheter
C-P Bypass
Infusion Pump Surgical Laser Laparoscope IV Catheter Tomography
System
IV Catheter Implantable
Pacer
DC
Defibrillator
Electrosurgery Breathing
Circuit
Protective
Apron
Automatic
Implantable
Cardioverter
Defibrillator
Automatic
Implantable
Cardioverter
Defibrillator
Chemistry
Analyzer
Figure 13. Medical devices most frequently recalled.
Restricting availability of problematic devices and functioning as a key
knowledge manager for safety information on medical devices are but two ways the
U.S. government promotes public health and safety. In other countries government
functions as a partner with nongovernmental organizations in these two areas;
however, in the United States, government alone is authorized to enforce
conformance with publicly established standards of conduct and quality or take
punitive action to correct behavior when the lack of conformance threatens public
safety. As part of that enforcement responsibility, the FDA monitors the prevalence
of violative manufacturers across the industry, as well as the types of violations
committed. The data in Figures 14 and 15 show the percentage of domestic and
international manufacturers where significant violations were found during
inspections. This information provides an overview for the FDA and the market of
the industry’s ability to conform to the regulations and an opportunity for both
parties to adjust their strategies of intervention. Those government strategies may
include technical advice, such as the development of specific standards and guidance
documents to assist industry in producing safe products; procedural advice oriented
toward preparing for a successful inspection; or they may call for the development of
new regulations intended to improve the process. Although the FDA invests
considerable resources in the time-intensive national and international (ISO)
standards development process, the Agency believes that the guidance they develop
as a result of direct knowledge of the industry provides more timely and responsive
information to the public and manufacturers.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
QS Inspections for US Medical Device
Manufacturers
2000
g 1500
_ o
§ I 1000
1 “ a
| 500
0
95 96 97 98 99 00 01 02 03
Fiscal Y ear
B ill
15 5 C
10^2
-■ —Total Inspections Violation Rates (%)
Figure 14. Violation rates for domestic inspections.
QS Inspections for Foreign Medical Device M anufacturers
600
«
■ 5 I 400
° I 200
w
c
- Q ■
95 96 97 98 99 00 01 02 03
Fiscal Year
* — Total Inspections Violation Rates (%)
10
Figure 15. Violation rates for foreign inspections.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The FDA model can be considered the most inefficient of the four discussed
in terms of cost to government. Because all costs for overseeing product risk
reduction from inadequate manufacturing quality stems from budget appropriated
through Congressional legislation, this model is in jeopardy of continuing the pattern
of providing fewer inspections at an increased cost without the timeliness that
business desires in order to remain competitive or the quality that the public expects.
Additionally, unless significant resources from dwindling budgets are diverted to
train or hire required specialists, the ability of the FDA to supply competent
inspectors capable of evaluating changing technology will decrease. Both factors
challenge this model’s ability to be accountable for minimizing public health risks.
Like many U.S. government bodies, the FDA is accountable to multiple
stakeholders and interest groups. Seeking quick market access, manufacturers
expect to receive fair and equitable inspections, reports, and follow-up activities
consistent with their competitors. Users or health care professionals play a critical
role in providing the public with service and information on the use, risks, and
benefits of a device and are often the conduit of information among the patient, the
manufacturer, and the government. They demand the availability of safe
technologies and must rely on the manufacturer and the government to ensure safety
and effectiveness of these products. Patients expect access to information on product
safety and protection against unnecessary risk because they have limited control over
the risks and benefits to which they are exposed. Finally, as an indirect stakeholder,
Congress expects responsible and accountable execution of responsibilities for public
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
health protection. It is this need to satisfy all stakeholders that defines the
responsibilities of the federal government and creates the tension. For example,
while the manufacturing community contends that problems with products and the
solutions that address the problems are proprietary, potentially detrimental to the
continued availability of “medically necessary” products and should, therefore, not
be made public, the health care professional and patient requires notification of
existing and potential problems to allay serious health consequences from
malfunctioning products. The FDA is expected to take actions that notify the public
about recalled and banned products; actions to correct nonconforming products; and
to require changes in the product, its manufacturing methods, labels, advertising, or
use. Health professionals, patients, and patient advocacy groups have come to
expect government issuance of warnings and safety alerts.
The ability to collect and analyze information on the manufacturing practices
for existing and new technologies, trend device performance in the marketplace,
manage the knowledge gained to develop or modify standards of quality and
performance, and guard the entrusted accountability through regulations authorizing
enforcement allows the FDA to have one of the most, if not the most, comprehensive
device surveillance systems in the world. A number of collaborative relationships
between the FDA and other national governments are fostered so that FDA
information is shared with those governments that do not have the authority to
collect this type of information for use in protecting their citizens. Although costly,
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
this is one of the many advantages of having the government serve as information
manager for public health protection.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTERIV
AN FDA THIRD-PARTY INSPECTION PROGRAM
Introduction
This chapter presents an examination of the only example of an FDA third-
party program for inspecting medical device facility compliance with the FDA’s
good manufacturing practice regulations. The third parties in this case are state
government inspectors who are not independent or private auditors; therefore,
implications drawn about private audits will present a challenge. The purpose for
looking at this case is to provide useful information from an actual experience on
production (direct) and transaction (hidden) costs and where it may be more or less
efficient to use a non-FDA inspector.
States as a Third-Party Auditor
In 1994, the FDA initiated a program that allowed a non-federal entity to
inspect medical device manufacturers for compliance with FDA regulations for
quality manufacturing practices. The FDA was most interested in finding out if
inspectors trained to conduct medical device inspections using different standards or
approaches could produce the same inspection results in an equal or more efficient
time frame. The FDA allowed this practice because of the increasing possibility that
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the medical device inspection program could not keep pace with the growing number
of manufacturers marketing products, especially with respect to the expanded
regulatory responsibilities given to the Agency by Congress. The FDA believed that
contracting with those state governments that had some knowledge of and experience
with device manufacturing would provide an opportunity to explore the possibility of
using this approach on a larger basis (McNulty, 2001). In order to gain the greatest
benefit while expending the least start-up cost in terms of training and oversight, the
FDA selection criteria included states whose investigators had a modest (minimum
of 1 year) level of experience in conducting device inspections. The FDA most
preferred those state inspectors experienced with the ISO auditing standards, as they
could also provide information on the possibility of combining inspections with other
types of audits. One reason the FDA chose this approach was to test the “one-stop
shopping” hypothesis that industry claimed was the primary reason for pushing
privatization. The market believed that the benefits of having FDA inspections
conducted by private ISO auditors with knowledge of quality system assessment was
a potential cost savings realized by eliminating multiple inspections. It is not
uncommon, particularly among the larger international companies, to be subjected to
multiple audits from the FDA, State, ISO, and component suppliers in one month.
No inspectors involved in this program had experience with ISO auditing.
California, Colorado and Texas met the criteria and in 1994 entered into a
contractual agreement with the FDA for the purpose of conducting inspections of
manufacturers that marketed moderate- to high- risk products according to FDA
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
guidance and regulations. Under contract, the states provided the required personnel,
materials, services, facilities, and performed all other tasks necessary for or incident
to the performance of inspections. Additionally, these third-party inspectors were
instructed to issue a listing of significant observations using the standard Form FDA
483 to the manufacturing facility and submit copies of the FDA 483 and a quarterly
report summarizing all contractual activities to the responsible FDA local or district
office. In turn, the FDA committed to providing the contractor with a list and
background information on designated moderate- to high-risk manufacturers
requiring inspections, conducting initial and follow-up training in FDA law and
inspection procedures, providing continuous updates on changes to FDA policies and
guidance documents, and providing payment for services as agreed to in the contract.
As a quality control measure, the FDA would conduct joint inspections with the
contractor, review all audit reports submitted, and conduct independent re
inspections of selected facilities evaluated by the contract inspector. Because the
legal staff sought to minimize the possibility of losing a court case due to challenges
to a third-party inspector’s qualifications or experience with the FDA procedures,
FDA retained the responsibility of re-inspecting problematic manufacturers to
develop enforcement actions. The FDA did expect contract inspectors to indicate
violative findings significant enough to warrant official action because the ultimate
goal was to test the third parties’ abilities to remove the most risky devices.
The contracts covered inspections conducted during fiscal years 1994,1995,
and 1996. The number of contract inspections was negotiated yearly; however, both
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
parties anticipated a relatively consistent number— 100,50 and 25 from California,
Texas and Colorado, respectively. The actual number of inspections negotiated and
completed is shown in Table 1. Inspection data for 1994 is not included because the
greater part of that initial year was spent conducting training and certification
inspections for each state participant. Additionally, a number of contract inspections
were classified as “washouts,” meaning manufacturers were out of business or no
longer manufacturing medical devices subject to FDA regulation. Washouts were
not counted as official inspections, and the FDA supplied additional locations as
substitutes.
Table 1
Third-Party Inspection Accomplishments
California Colorado Texas
Contracted
FY 1995 80 30 50
FY 1996 80 30 65
Total 160 60 115
Conducted
FY 1995 97 30 48
FY 1996 23 19 32
Total 120 49 80
% Completed (’95/’96) 75 82 70
Washouts (out of
business, non
manufacturers, etc.)
FY 1995 128 25 8
FY 1996 60 16 5
111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A key argument in favor of this privatization effort was the anticipated cost
savings to FDA. Leaders in the organization reasoned that the use of external
inspectors would reduce the need for a full-time inspection force and the associated
direct and overhead costs. The literature describes two components of service
delivery costs, production or actual cost of delivering the inspection and transaction
costs or the cost of oversight to ensure accountability of the third party (Ferris &
Graddy, 1991). Elements of the two cost components associated with the FDA
inspection program are listed in Figure 16. Contrary to Niskanen’s contention that
bureaucrats in FDA would know the actual cost of delivering inspections, attempts to
calculate true program (production and transaction) costs proved to be a challenge.
The available data suggest that the FDA bases the average cost of an inspection on
production costs and what is considered overhead—training, leave, some equipment
cost, and administrative costs. Although there was no documentation that transaction
costs were considered in the calculation of inspection costs, the FDA indicated the
approximate cost of a routine inspection during the same time period of the state
contact to be $4,550 (see Table 2).
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Production Costs
Pre-inspection preparation
On-site inspection
Report preparation
Travel
Preparatory training
Transaction Costs
Training development, delivery, and update
On-site training by FDA staff: time and travel cost
Program planning and implementation
Consultations with third-party inspectors
Evaluating third-party reports
QA audits of third-party
Follow-up o f questionable inspections
Expansion o f monitoring/tracking system
Program monitoring and evaluation
Figure 16. Costs associated with an FDA inspection.
Table 2
Comparison of FDA and Third-Party Inspection Costs
Average cost of FDA inspection $4,550.00
Average cost of third-party inspection $4,236.00
FDA oversight cost
4.5 staff @ $70,000
Classroom training
$315,000.00
$ 84,000.00
Additional cost of third-party inspection
FDA oversight cost/number of inspections
$399,000.00/175 $2,280.00
Cost of third-party inspection
Average third-party cost + additional FDA cost $6,516.00
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The basis for inspection costs under the state contract program mirrored that
of the FDA’s production and overhead costs, although it is not obvious that the
components of overhead for both programs were the same. The actual cost of the
contract program could not be verified because cost data were not calculated or
maintained; however, use of the available information on direct inspection cost and
program overhead or transaction cost is useful in illustrating how the comparative
costs might change. Using the total contract cost (for the three states) and the
number of inspections to be conducted, the average cost of the contract inspection
was $4,236. The average contract cost for inspecting a facility remained constant,
although total contract costs increased as inspections were added to accommodate for
the washouts. Additional data on the cost for FDA oversight, including report
review and inspection evaluation, and the cost of classroom training is listed. Merely
adding these two transaction costs to the average cost of a third party inspection
suggests that the true cost of the third-party inspection is greater than anticipated, as
shown in Table 2. A review of the costs associated with this third party inspection
program also suggests that the final cost of a non-FDA inspection is greater than the
average cost of an FDA inspection. Much of this increase is due to the transaction
costs associated with establishing the program. Costs are expected to be higher in
the first contract year and decrease in subsequent years, as familiarity with inspection
processes and government policies increases. Likewise, expenses associated with
program development and implementation should decrease. Costs associated with
monitoring may not decrease; and, according to FDA managers, it takes three to five
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
years to develop a fully skilled and independent investigator. Therefore, all costs
associated with a third-party program are likely to be higher than FDA inspection
costs for some time.
Actual performance for the first full year of the contract was positive. The
initial 1995 rate of accomplishment for inspections, ranging from 96% to 121% for
the three states, declined in 1996, particularly for California. Responses from the
states and the FDA program manager suggest a number of reasons, including the
extra inspections conducted in 1995; the increase in inspection duration due to the
level of difficulty for some of the inspections conducted in 1996; a loss of FDA
trained investigators in the states; the FDA lag time in training new investigators;
and the beginning of a series of Congressional challenges to what industry perceived
as FDA’s aggressive inspection techniques. Although the states were not directly
affected by Congressional oversight, scrutiny of the FDA had a calming effect on the
program. Additional feedback from the states indicated that when evaluating the
training session, over half of the 17 trained inspectors commented on the detailed
nature of the FDA inspection, questioning whether the level of detail provided any
greater assurance of product safety than other standards they may have used.
In addition to performance results, the data collected also provided
information on the timeliness of inspections conducted by third-party contract
investigators. According to the contract, moderate- and high-risk inspections were
projected to take 44 and 33 hours respectively. On moderate risk inspections,
California averaged 53 hours; Denver and Texas averaged 42 hours; and the FDA
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
averaged 40 hours. For high risk and new products coming to market, or when the
inspection history suggested problems with the manufacturers’ adherence to the
quality standards, third-party inspection time ranged from 80 to 85 hours for all
third-party inspections, as compared to 72 hours for equivalent inspections
conducted by FDA during the same time period.
The third-party project also faced human resource challenges. All contractors
lost FDA trained inspectors during the three-year contract period. This reduced the
number of commissioned inspectors and, therefore, jeopardized the number of
inspections that could be conducted. When states were able to hire competent
inspectors for this contract, the FDA was unable to provide the staff resources
necessary to conduct new or additional training courses. Comments from contract
inspectors reflected the need and, in some cases, the frustration of not having
adequate FDA staff available to respond to questions or to provide the additional
training opportunities promised in the contract. The lack of adequate staffing for
interacting with the contractors also increased communication problems. Third-party
inspectors found communication with the FDA on changes to programs or policies
inconsistent, resulting in differences in the interpretation of violations of the
regulation. Experienced FDA staff skilled in the legalistic and protective culture of
the FDA also found it difficult to disclose or share information on regulatory
inspection techniques and strategies, a substantial portion of which is necessary to
develop compliance justifications.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Quarterly reports submitted to FDA from the third parties and internal
correspondence about the program also reveal that FDA staff spent considerable
time, some stating as many as 10 hours per inspection, unofficially “retraining”
contract inspectors prior to the actual inspection because of the time lapse between
initial training and the first inspection. The need to review preliminary inspection
findings required additional, unplanned government time. The actual time spent
reviewing reports from third parties averaged 5l A hours, exceeding the 3 hours
planned, with considerable unaccounted time spent clarifying the reports with the
contract inspectors. If these inspections were on new products coming to market or
where significant violations were suspected, the number of consultations with the
FDA field office and headquarters staff during and after the inspection increased.
Comparison of Inspection Results
The Investigators Operations Manual (IOM) states that the Form FDA 483
Inspectional Observations sheet should be used to “report all significant
objectionable conditions noted during the inspection.. . . ” Form FDA 483 is written
notification issued for objectionable conditions or deviations from the manufacturing
quality standard; for failure to report deaths, injuries, or malfunctions associated with
the manufacture of a device; or for failure to implement a tracking system for certain
life supporting, life sustaining medical devices that would result in serious adverse
health consequences if they failed. Under FOIA, this document can be released to
the public after removing proprietary information. Third-party inspectors were
instructed to issue FDA 483s to the inspected establishment’s most responsible
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
manager at the close of the inspection. The data in Table 3 indicate that third parties
issued a total of 139 FDA 483s. Based on an examination of those findings,
California clearly found more significant and objectionable conditions than either
Colorado or Texas, with findings 54% and 63% greater than either state. Except for
those issued by Texas in 1995 and 1996, the percentage of FDA 483s issued by the
third party almost doubled those issued by the FDA during the same time period.
This seems to indicate that third-party inspections found significant problems in over
half of the facilities they entered, as opposed to only a third of the facilities that FDA
entered.
One possibility could be that as newly trained in the FDA inspection
techniques, the third parties follow the procedures very closely and may exercise
more conservative judgment in their determinations. As mentioned previously, the
lack of clarity on the policies and procedures also may have prompted a more
conservative judgment call by the third-party investigator. A more experienced
investigator might exercise more discretion in the application of the regulations or
may have better selectivity when identifying those requirements that are significant
and overlook the lesser significant ones. Interestingly, the number of FDA 483s
issued by California and Colorado indicating significant observations increased in
the second year as the third-party inspectors gained more experience with the
technique. While California and Colorado increased their issuance, Texas issued
only half as many as in the first year. During the second year, third parties also
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inspected a larger number of higher risk device manufacturing facilities than those
inspected in 1996.
Table 3
Comparison of Form FDA 483 Observation Findings
California Colorado Texas FDA
Total Inspections
Conducted
FY 1995 97 30 48 1662
FY 1996 23 19 32 1702
Total 120 49 80 3364
Total FDA 483s
Issued
FY 1995 58 20 21 515
FY 1996 18 15 7 562
Total 76 35 28 1077
% FDA 483s Issued
FY 1995 60 67 44 31
FY 1996 78 79 22 33
Total 63 71 35 32
An examination of third-party inspection results also provided valuable data
on the compliance status of the manufacturers visited. The FDA classifies an
inspection based on the level of deficiencies found and the significance of the
deficiencies with respect to their impact on the safety of the product. The FDA uses
a three tiered classification to determine the appropriate level of follow-up needed
after an inspection. If deviations from the manufacturing quality regulations are
significant enough to warrant an official warning notification, which could be a
prelude to an enforcement action, the inspection is considered Official Action
Indicated (OAI). Problems of concern that require attention, but less than immediate
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
corrective action, are classified as Voluntary Action Indicated (VAI). Those
inspections with the least significant FDA 483 observations, such as sloppiness in
recording available data in the appropriate areas, are considered No Action Indicated
(NAI).
An evaluation of the final classifications based on the third-party assessment
was performed to compare how the third party and the FDA viewed significant
violations—an indication of the ability of third parties to find the same types of
problems as FDA. This comparison found in Table 4 and Table 5 shows a
significant difference between the two groups for the most violative types of
inspections (OAI) conducted in 1995. Colorado found no inspections with problems
significant enough to warrant OAI ratings in 1995 or 1996. In 1996, the percentage
of inspections initially classified as OAI by California and Texas was similar to
inspections conducted by FDA. There was considerable difference between FDA
and the third-party inspectors in the percentage of classifications rated VAI and NAI.
This difference might suggest that classification determinations made by third parties
are more conservative than those made by the FDA. Because the FDA makes all
final classifications, the data was examined to identify differences between initial
classifications made by third parties and the final classification made by the FDA.
This review found that senior FDA officials recommended changes in two
classifications made by both California and Colorado. In both cases, the FDA would
have given a lower VAI classification. Noted in a number of comments from the
FDA staff was the lack of supporting documentation supplied by the third-party
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inspector. This documentation frequently provides additional or explanatory facts
that may alter the classification outcome in either direction. Staff notes also indicate
that some inspections did not cover certain critical elements, such as whether there
was recalled product not reported to the FDA or if manufacturers failed to report a
death associated with a device. These are two of the most important components of
an inspection because both may signal a direct link between product performance
and patient injury. The consequence of misidentifying manufacturing problems or
not inspecting critical components mandated by the regulations could be costly if
those problems result in devices that cause serious adverse health consequences or
death.
Table 4
Third-Party Inspection Classifications
California Colorado Texas
Official Action Indicated
Inspections
FY 1995 2 0 4
FY 1996 3 0 4
Total 5 0 8
Voluntary Action
Indicated Inspections
FY 1995 56 20 10
FY 1996 15 15 1
Total 71 35 11
No Action Indicated
Inspections
FY 1995 39 10 7
FY 1996 5 4 2
Total 43 14 9
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5
Percentage of Third-Party Inspection Classifications
California Colorado Texas FDA
% Official Action
Indicated Inspections
FY 1995 2 0 8 20
FY 1996 13 0 13 12
% Voluntary Action
Indicated Inspections
FY 1995 56 67 21 44
FY 1996 65 79 3 36
% No Action
Indicated Inspections
FY 1995 40 33 15 35
FY 1996 22 21 6 52
The final assessment compared the types of deficiencies found during an
inspection. Data from third-party inspections was compared to findings by FDA
inspectors to determine if both groups found similar types of problems. This
information is valuable to the FDA because it provides a broad overview of the
problematic manufacturing areas within and across medical products; provides focus
for the strategic and risk-based inspection planning; signals potential training needs
for industry and FDA; and identifies areas where additional guidance for assisting
manufacturers needs to be developed.
The market benefits from the feedback provided to regulatory affairs
representatives on the types of problems they are encountering with products and the
weaknesses in their internal auditing process. In addition to preparing a list of
inspection observations, the investigator prepares a narrative Establishment
Inspection Report (EIR) for each inspection. Data from 57 (42, 9 and 6 from
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
California, Colorado and Texas respectively) EIRs prepared by third-party
investigators were analyzed to determine the types of problems noted. These 57
EIRs represent all that were available from the inspections where FDA 483s were
issued. This information was compared to the types of observations the FDA noted
on FDA 483s issued during the same time period. Using the data from the 1996
inspections, Table 6 shows the top eleven significant violations indicated on the FDA
483s for both groups. The table further compares the percentage of violations
observed by the third party to the observations of FDA inspectors. For example, the
third-party investigators found violations related to conducting or correcting
deficiencies in internal audits in 61% of the inspections conducted; FDA found
similar violations in 53% of the inspections conducted. These findings suggest that
significant problems found by third-party inspectors compared favorably to those
found by FDA inspectors and at the same rate of occurrences.
Table 6
Comparison of Top Ten Deficiencies Found
States FDA
% %
Inspection Test Procedure 71 76
Internal Audits 61 53
Complaint Handling 55 39
Device History Records 52 55
Component/device Specs 48 50
Process Validation 35 39
Processing Procedures 29 32
Process Change Controls 29 26
Equipment Calibration 29 24
Equipment Specifications 29 21
Change Control Procedures 23 26
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER V
SUMMARY AND CONCLUSIONS
The role of government in auditing manufacturing quality for the four
systems examined (European Union [EU], Canada, Australia, and the United States)
can be described by the following four modes of operation: (1) market-based,
(2) govemment-market participatory-based, (3) government-controlled fee-based,
and (4) govemment-fmanced-based. When selecting an operating mode for
managing an audit process, the level of control and production government desires or
is required to exercise has a direct impact on the degree of cost savings realized and
the government’s responsiveness to accountability concerns. Figures 17 and 18 are
provided as illustrations of this concept. Market-based auditing systems rely heavily
on non-government and third-party entities to evaluate manufacturing quality. When
government adopts this New Public Management (NPM) mode, government
produces less and has less control over the audit process. In the govemment-market
participatory mode, the functions necessary to assure manufacturing quality are
shared or distributed between government and the market, with costs being covered
by each group. Government exercises considerably more involvement in audit
production than the market-based system at less cost when fees are collected and
used by the government to fund the audit process. Government has more control
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
over the process, but does not shoulder most of the production cost of implementing
the program. Finally, when a government-financed auditing program is
implemented, the government assumes all of the responsibility and the market incurs
no direct cost but government maximizes its involvement and control over product
safety.
■ S
o
cj
o
c
Government Controlled
Fee-Based
Government Financed
Market-Based
Government-Market
Participatory
Government Cost
Figure 17. Relationship of four system types to accountability
and efficiency costs.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
d
( L >
s
< U
_ >
' o
j a
c i
< u
ID
>
o
O
Government-Market
Participatory
Government Financed
Market-Based
Government Controlled
Fee-Based
Government Control
Figure 18. Relationship of four system types to government
production and control.
Except for the FDA model, changes in how audits are conducted and who
conducts them reflect a tradeoff between efficiency and accountability, supporting
opinions that there is a strong connection between the two. For example, if
government exercises the maximum amount of control (i.e., full development of
laws, standards, reporting requirements and exercises enforcement) and is the sole
producer in the process (i.e., actively engaged in and fully responsible for auditing
and all other interactions with the manufacturer), then it is highly likely that the
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
governing body assumes the full cost of the audit process—development of
regulations and standards, staffing, and auditing—and full accountability for
addressing risks and communicating with the public and legislature. When the
market is assumed to be self-correcting or a third party assesses compliance to
manufacturing quality requirements, government produces less, and layers of
nongovernmental bodies dilute control over the process. Cost to government is less,
and the chain of accountability becomes fuzzier. When government is the highest
producer in the audit process but exercises minimal control or produces little but has
significant control over the audit process, it has the greatest flexibility in terms of
cost and accountability, if cost reduction is not the driving force. An expanded
version of this framework is used to summarize the models and present the study
conclusions.
Characteristics or responsibilities related to involvement and control are
included in Figure 19 to better display the factors that impact on cost and
accountability. Each quadrant reflects the general responsibilities of government in
that particular category. Within each category, government exercises its role through
the development of risk-based regulations and its involvement in the development of
standards used to determine quality and performance as an operational basis for
assessing the device manufacturing safety and audit effectiveness.
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
High Production-Low Control High Production-High Control
Develops regulations/policies
Develops/active participant in standards
development
Audits for high risk with staff specialists
and experts
Audits nongovernmental auditors
Enforces regulations
Shares cost o f monitoring program
Develops regulations / policies
Develops standards
Staffs specialists and experts
Assesses compliance with
requirements
Monitors trends for safety
Enforces regulations
Assumes full cost o f monitoring
program
Low Production-Low Control Low Production-High Control
Participates in development o f regulations/
policies
Participant in standards development
Assumes little cost o f program
Market noncompliance is self correcting
Market responsibility for auditing
Develops regulations/policies
Responsible for nongovernmental
auditing program
Monitors auditors
Monitors manufacturing performance
Enforces regulations
Figure 19. Characteristics of government production versus control.
The EU market-based model is considered lowest in control and production
and, therefore, less costly; while Australia’s government controlled, fee-based model
and FDA’s government-financed models are most costly because of their level of
control and production in the process. The governing structure of the EU employs
fewer resources to conduct or oversee the quality auditing process, shifting the cost
to the market. This approach allows for a reduction of costs to those Member States
that operated under a more mature system. Restructuring the Canadian system to a
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
govemment-market participatory system transferred the cost of auditing to the
regulated industry. The government reduced its production of audits, thereby saving
the cost of those resources and allowing for a redirection to other strategically critical
areas such as monitoring of third parties. Requirement for increased production in
the Australian government-controlled, fee-based model may increase the cost for
program oversight, although the cost of auditing should not increase. Where concern
exists about the efficiencies (adequate resources and expertise at lowest cost) of
timeliness and on-site presence in facilities, the market-controlled and fee-based
models provide greater resolution resulting in quicker access to technologies,
increased knowledge (reduced risk) of the quality of manufacturing and, possibly, a
reduced audit cost because of competition. Reducing the need for auditors reduces
the cost of staffing, overhead, and training required to maintain a technically
competent staff to address the changing technology. The appearance of a significant
cost reduction could be elusive because of the need to expend resources to maintain a
skilled and equally competent workforce capable of monitoring private auditors.
Significant reductions in expertise, competency, and staff capable of
monitoring nongovernmental auditors and the regulated industry challenge the
government’s ability to exercise responsible accountability. Accountability concerns
are greatest in the EU market-based model, where government involvement is
focused on developing high-level policy. The lack of a relationship between the
regulated and the regulator interferes with the control government has over the audit
process and the information gained, thus weakening the regulator’s ability to give a
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
timely response to public concerns. The closed process created by the contractual
arrangement between the third party and the manufacturer coupled with the
voluntary disclosure of audit results shifts the accountability for taking correcting
audit concerns to the manufacturer.
A greater balance between efficiency and accountability occurs in the
Australian government-controlled, fee-based model and the participatory market-
based model of Canada. These systems mirror the high product-low control and the
high control-low production models. Both models provide a greater degree of
accountability than the EU market-based model but at a lower cost to government
than the FDA model. Time may prove the high control fee-based model to be more
costly than anticipated because of unexpected transaction cost. The participatory
market-based system may experience similar cost increases if the role of the
Canadian government in changing from auditor to auditor monitor incurs new
structures and responsibilities, but the expectation for high-level accountability
remains uncompromised. Australian and Canadian governance models also provide
the best balance between cost and accountability, when the responsibility for
minimizing risk through the inspection of medical devices shifts from government to
nongovernment groups. Increasing efficiency by merely reducing the cost of
resources dedicated to auditing compromises the government’s ability to address risk
prevention and resolution; however, reinvesting in selective oversight activities
restores some of that capability. Each model contains positive and negative elements
that must be weighed when considering privatizing inspections for manufacturing
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
quality. Discussing the challenges each model presents in light of the FDA third-
party inspection program provides insight into the benefits from cost savings and the
accountability concerns that the FDA may be forced to balance if directed to
implement an alternative system.
The FDA might look favorably at changing from a govemment-financed-
based system to one that uses third parties if true cost reductions are realized without
sacrificing substantial involvement and oversight. Third-party program
attractiveness hinges on FDA maintaining knowledge about devices on the market,
preserving quality and integrity in the inspection process, locating alternatives that
provide adequate technological specialization, and upholding Agency involvement in
standards development activities that are flexible enough to expand with innovation
while protecting the public health and safety. Analysis of the FDA’s limited use of
third parties suggests that a number of these efficiencies can be realized. First,
trained third-party inspectors may be capable of conducting an FDA equivalent
inspection that finds significant violations or indications of increased risk,
particularly when experience is gained, at a production cost that may not be
significantly different from the cost to FDA. Additionally, while third-party
inspection times for conducting moderate to high-risk inspections initially took
longer, inspection times for two of the third parties improved with experience. The
analysis also suggests that over time, third parties are capable of detecting similar
audit deficiencies as the FDA, indicating that third parties may also maintain the
quality of the inspection.
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Using a third party may also aggravate the budget dilemma for the FDA. If
third parties visit a greater number of facilities and uncover more noncompliance, the
resources needed to determine the extent of the risk involved and any potential
enforcement action escalates rapidly. These types of inspections and follow-up
activities incur much greater resource cost than routine surveillance monitoring.
Limited feedback also suggests the need to develop more comprehensive guidance
for third parties on the FDA regulatory standards for inspections to supplement the
initial training. Comments from FDA staff suggest that the overhead and transaction
costs of this effort alone may equate to an additional one quarter to a half million
dollars yearly. These factors have real implications when considering the three
alternative models. If cost reduction is the highest priority, the greatest difference or
savings in cost stems from the level of control over the process and involvement in
conducting audits that the FDA exercises. Based on the third-party study, the goal of
lowering cost by reducing government inspectors without compromising the quality
of the inspection is attainable if the FDA chooses to oversee in a high control-low
production manner. All of the alternative models transfer the cost of inspections to
the industry, whether that cost is paid through a contractual arrangement or as fee to
government. Current law requires the FDA to conduct biennial inspections of those
manufacturing facilities that generate products used by the public. The discussion in
Chapter 2 highlights the FDA’s inability to meet that legal obligation. If new
inspection requirements mandate a biennial inspection and allow for surrogate FDA
inspectors, the financial burden on industry increases significantly. If industry pays
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the FDA for inspections, some of the increased cost will be offset. Fee-based models
may cover all cost or only the cost of the inspection, depending upon how industry
influences the legislation. In all cases except the fee assessment to industry for
covering FDA inspections, the accountability is affected because a new monitoring
role is substituted for a direct inspection role. The flow of information may be
hampered and the trustworthiness of the information may be questioned, not so much
because of the inability to conduct an adequate inspection, if trained, but because of
the FDA’s inability to fully monitor the privatized process in order to eliminate
conflicts of interest.
A high level of control translates into an increased cost of monitoring the
program, which means adding on-site auditing of the third parties and more closely
scrutinizing the reports submitted. The cost of additional training and
communication and the need for restructuring or adding functions to cover many of
the additional transaction costs may be offset by cost savings from a reduction of
inspections. Data on the cost of establishing a yearly monitoring function is needed
to adequately evaluate any savings. Maintaining control over the standards used to
assess manufacturing quality may require the cost of clarifying the standards or
providing additional training as requirements change. If Congressional pressure
changes the level of control government can exercise over developing standards for
evaluating quality and FDA is mandated to use ISO standards, the Agency faces a
one-time cost of changing the regulation and an ongoing cost of training staff
inspectors and industry on the expectations when using these standards. If
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
government takes an active role in setting and enforcing standards, then the need for
government to conduct or subsidize inspections can be minimized. If a medical
device producer pays the cost of obtaining a certification stating that its devices meet
the relevant standards, the government’s resources can be directed toward making
sure that the standards are clear and that violations are punished through either
government enforcement or private lawsuits. Government involvement in the audit
process would not be eliminated, but regulatory interaction will be reduced.
If the FDA selects a high production-low control model, costs increase to
cover third-party monitoring and expanded participation in the standards process. As
suggested in the pilot, if third-party capability for finding significant risk concerns in
moderate- to high-risk products is comparable to the FDA’s inspection findings,
under the low control-low production model, FDA could save by developing a
smaller monitoring and oversight program.
Analysis of the third-party program also provided insight into other indirect
cost savings. During the first year of the contract, third parties encountered a large
number of manufacturers that were out of business or no longer in the medical
device manufacturing business. The considerable number of visits that resulted in no
viable inspections suggests that the FDA itself may be spending considerable
resources visiting facilities with no payoff and expending funds maintaining a
database that contains outdated and unusable information. Transferring inspection
responsibility to a third-party system where existing manufacturers would contract
when an inspection is needed could save the cost of washout visits, particularly in the
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
low production-low control model and lesser cost for the other models. Cost savings
would be less under the high control-low production model if manufacturers are
required to register their marketing and manufacturing facilities with the
government.
The nature of the FDA third-party program revealed less about accountability
under privatization. The data suggest that with appropriate training, third-party
inspectors are capable of producing quality audits, indicating that public safety and
risk reduction can be maintained. How third-party inspectors handled the findings
and transfer of information, both important to maintaining a high level of
accountability, raised chain of accountability questions in each of the three non-FDA
models examined. The third-party program highlighted difficulties in documenting
critical information, and the evaluation of the three alternative models showed
significant gaps in information transfer of accountability questions in any of the three
alternative models. The vocal and participatory citizenry in the United States
expects continued risk communication from the FDA. To maintain this
accountability link, the FDA needs to remain connected to the inspection process,
either through strong monitoring systems or government fee-based inspections.
Likewise, the information needs required to maintain a strong enforcement presence
necessitates a relatively strong control and audit production oversight function. It is
unlikely that information expectations and requirements under the FOLA will
decrease or that manufacturers will reduce their expectation that the FDA protect
“good businesses” from those that intentionally skirt the regulations. Alternative
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mechanisms that allow for a continuous information flow without on-site presence,
such as Information Technology (IT) systems, are expensive to establish and
maintain.
The FDA should, therefore, seek to implement, or at least seriously consider,
models that provide a high level of control with a reduced level of audit production.
The greatest benefit in adopting this model is the potential cost savings from reduced
human resources needed for inspections. Whether those savings are actually
realized, and the above discussion could certainly argue against any true savings,
models that provide the assurance of some control and involvement in the audit
process provides a strong government presence and responsibility in the areas that
most protect the public from risk.
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SELECTED BIBLIOGRAPHY
Anderson, S., Daly, D., & Johnson, M. (1999). Why firms seek ISO 9000
certification: Regulatory compliance or competitive advantage? Production
and Operations, 8, 28-44.
Arens, A., & Loebbecke, J. (1997). Auditing: An integrated approach (7th ed.).
Upper Saddle River, NJ: Prentice Hall.
Banish, R., & Antoff R. (2000, June 6). Third party audit pilot project in the state o f
Delaware: Final report. New Castle: State of Delaware, Department of
Environmental Resources and Environmental Control Division, Division of
Air and Waste Management.
Barnes, F. C. (1998). ISO 9000. Myth and reality: A reasonable approach to ISO
9000. S. A. M. Advanced Management Journal, 63, 23-31.
Belke, J. (2001, April 23). The case for voluntary third party risk management
program audits. Washington, DC: U.S. EPA, Chemical Emergency
Preparedness and Prevention Office.
Blanchard, L., & Hinnant, C. (1998). Market-based reforms in government: Toward
a social subcontract. Administration & Society, 30,483-512.
Borcherding, T. (1988). Some revisionist thoughts on the theory of public
bureaucracy. European Journal o f Political Economy, 4 ,47-64.
Boylan, S. (2000). Using experimental asset markets to illustrate the value of
auditing. Issues in Accounting Education, 15,43-65.
Brown, M., & Brudney, J. (1998). A smarter, better, faster, and cheaper
government: Contracting and geographic information systems. Public
Administration Review, 58, 335-345.
Bunker, M., & Davis, C. (1998). Privatized government functions and freedom of
information: Public accountability in an age of private governance.
Journalism and Mass Communication Quarterly, 75, 464-477.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Caiden, G. (1994). Administrative reform—American style—Deregulating the
public service. Public Administration Review, 54, 123-129.
Chen, Y., & Shin, R. (2000). Seizing global opportunities for accomplishing
agencies missions: The case of ISO 14000. Public Administration Quarterly,
24, 69-94.
Collins, L. (1997). A proposal for piloting a new beginning for industrial
environmental compliance. Minutes o f a roundtable meeting on the use o f
third party auditors to ensure safe handling o f chemicals. Philadelphia:
Wharton School, University of Pennsylvania.
Collins, L., Belke, J., Halpem, M., Katz, R., Kunreuther, H., & McNulty, P. (2002).
The insurance industry as a qualified third party auditor. Professional Safety,
47^ 31-38.
Daft, R. (1991). Organizational theory and design (4th ed.). St. Paul, MN: West
Publishing Co.
Davenport, T. (1995). Why reengineering failed—-And what comes next? Fast
Company, 1, 69-74.
Dean, Ed. (1996). Outsourcing. Available from
http://mijuno.lare.nasa.gov/gov/dfc/os.html.
Deloitte & Touche. (1993). Management consulting and quality standards update:
ISO 9000 survey. Fairfax: C. E. E. M. Information Services.
Deloitte & Touche. (1999). DOD streamlines purchasing practices, replaces
milspecs. Deloitte and Touche Review, 11, 23.
Denhardt, R., & Denhardt, J. (2000). The new public service: Serving rather than
steering. Public Administration Review, 60^ 549-559.
Dilger, R., Moffett, R., & Struyk, L. (1997). Privatization of municipal services in
America’s largest cities. Public Administration Review, 57, 21-26.
Docking, D., & Dowen, R. (1999). Market interpretation of ISO 9000 registration.
The Journal o f Financial Research, 2 2 ,147-161.
Donahue, J. (1989). The privatization decision: Public ends, private means. New
York: Basic Books.
Downs, A. (1994). Inside Bureaucracy. Prospect Heights, IL: Waveland.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Downs, G., & Larkey, P. (1986). The search for government efficiency. New York:
Random House.
Durant, R. (1998). Agenda setting, the “third wave,” and the administrative state.
Administration & Society, 30, 211-248.
Durant, R., & Legge, J. (2002). Politics, public opinion, and privatization in France:
Assessing the calculus of consent for market reforms. Public Administration
Review, 62^ 307-323.
Edington, B. (1997). Health incentives: Canadian health reform in an international
context. Canadian Public Policy, 25,341-343.
Elmuti, D. (1996, September/October). World class standards for global
competitiveness: An overview of ISO 9000. Industrial Managementa 5-9.
Er, J. (1996). A third party approach to environmental regulation and possible roles
fo r insurance companies. Unpublished doctoral dissertation, University of
Pennsylvania.
Ferris, J., & Graddy, E. (1986). Contracting out: For what? With whom? Public
Administration Review, 46a 332-344.
Ferris, J., & Graddy, E. (1991). Production costs, transaction costs, and local
government contractor choice. Economic Inquiry, 29, 541-554.
Fiorito, L. (2003). Ronald Coase and the institutionalist once again: A note on a
neglected essay by Lawrence Kelso Frank. Journal o f Economic Issues, 37,
797-811.
Food and Drug Administration (FDA), Office of Enforcement. (2001). The FDA
Enforcement Study. Washington, DC: Department of Health and Human
Services.
Food and Drug Administration (FDA) (2003). Investigations operations manual.
Washington, DC: Department of Health and Human Services.
Frederickson, G. (1997). The spirit o f public administration. San Francisco, CA:
Jossey-Bass.
Frederickson, G., & Frederickson, D. (1995). Public perceptions of ethics in
government. Annals o f the American Academy o f Political and Social
Science, 537 163-172.
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fulton, J. (1993). A Canadian take on American health care. Financial Executive, 9,
10-13.
Gargan, J. (1981). Consideration of local government capacity. Public
Administration Review, 41, 649-658.
Gendron, Y., Cooper, D., & Townley, B. (2001). In the name of accountability:
State auditing, independence and new public management. Accounting,
Auditing & Accountability Journal, 14, 278-290.
Gore, A. (1993). The Gore Commission Report on Reinventing Government. New
York: Times Books.
Gore, A. (1995). Common sense government: Works better and costs less. New
York: Random House.
Gostin, L. (2000). Public health law. Berkley, CA: University of California Press.
Gray, G. (1998). Access to medial care under strain: New pressures in Canada and
Australia. Journal o f Health Politics, Policy and Law, 23, 905-947.
Gupta, M., & Zhender, D. (1994). Outsourcing and its impact on operations
strategy. Production and Inventory Management, 35, 70-75.
Halpem, M. & Splain, J. (2001, January). Financial considerations associated with
third party audits in the RMP program. Presented at the Roundtable Meeting
on Third Party Auditors in the RMP process, Wharton School, University of
Pennsylvania Risk Management and Decision Process Center.
Haque, S. (2001). The diminishing publicness of public service under the current
mode of governance. Public Administration Review, 61,65-82.
Harlow, C. (2002). Accountability in the European Union. Oxford, UK: Oxford
University Press.
Horton, L. (1995). Medical device regulation in the European Union. Food and
Drug Law Journal, 50,461-476.
Iannucci, A. (2001). Medical device inspection evaluation. Final report to the
Medical Device Industry Initiatives Grassroots Task Force. Irvine, CA: The
UC Irvine Center for Statistical Consulting. Unpublished study.
ICMA (1989). Service delivery in the 1990s. Alternative Approaches for Local
Governments. Washington, DC: International City Management Association.
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Johnston, J., & Romzek, B. (1999). Contracting and accountability in state Medicaid
reform: Rhetoric, theories, and reality. Public Administration Review, 59,
383-399.
Kaboolian, L. (1998). The new public management: Challenging the boundaries of
the management vs. administration debate. Public Administration Review,
58, 189-194.
Kachelmeier, S. (1991). A laboratory market investigation of the demand for
strategic auditing. Auditing: A Journal o f Practice & Theory, 10, 25-48.
Keane, C., Marx, J., & Ricci, E. (2001a). Privatization and the scope of public
health. American Journal o f Public Health, 91, 611-617.
Keane, C., Marx, J., & Ricci, E. (2001b). Perceived outcomes of privatization in
local health departments. Milbank Quarterly, 79, 115-137.
Keane, C., Marx, J., & Ricci, E. (2003). Managerial and professional beliefs
influencing public health privatization: Results of a national survey of local
health department directors. Journal o f Health and Social Behavior, 44, 97-
111.
Keane, C., Marx, J., Ricci, E., & Barron, G. (2002). The perceived impact of
privatization on local health departments. American Journal o f Public
Health, 92, 1178-1180.
Kearney R., & Hays, S. (1998). Reinventing government, the new public
management and civil service systems in international perspective: The
danger of throwing the baby out with the bathwater. Review o f Public
Personnel Administration, 18, 38-54.
Kennedy, J. (2001). A better way to regulate. Policy Review, 109, 57-75.
Kettl, D. (1993). Sharing power: Public governance and private markets.
Washington, DC: The Brookings Institution.
Kettl, D. (2000). The transformation of governance: Globalization, devolution, and
the role of government. Public Administration Review, 60, 488-497.
Kettl, D., & Dilulio, J., Jr. (1995). Cutting government. The Brookings Review, 13,
40-44.
Kettl, D., & Dilulio, J. (Eds.). (1995). Inside the reinvention machine. Washington,
DC: The Brookings Institution.
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kniss, C. (1999). Public administration reform. Journal o f Public Administration
Research and Theory, 9, 493-502.
Kunreuther, H., McNulty, P., & Kang, Y. (2001a). Third party inspection as an
alternative to command and control regulation. In K. Deketelaere & E. W.
Orsteds (Eds.), Environmental Contracts. Cambridge, MA: Kluwer Law
International.
Kunreuther, H,, McNulty, P., & Kang, Y. (2001b). Improving environmental safety
through third party inspection. Paper presented at the 2001 Environmental
Contracts: Comparative Approaches to Regulatory Innovation in the United
States and Europe Conference at the University of Pennsylvania,
Philadelphia.
Landry, J. (1993). Administrative reform and political control in Canada.
International Political Science Review, 14, 335-349.
Lane, J. (1994). Will public management drive out public administration? Asian
Journal o f Public Administration, 1 6 ,139-151.
Lavelle, J., Krumwiede, D., & Sheu, C. (2000). A privatization model for
government outsourcing. Production and Inventory Management Journal,
41, 45-52.
The Lewin Group for AdvaMed. (2001). The impact o f regulation and market
dynamics on innovation (Issue No. 4). Washington, DC: The Lewin Group.
Madhok, A. (2002). Reassessing the fundamentals and beyond: The transaction cost
and resource-based theories of the firm and the institutional structure of
production. Strategic Management Journal, 23, 535-550.
Mallaby, S. (2003). The place of government. Policy Review, 117, 21-29.
Mazmanian, D., & Sabatier, P. (1989). Implementation and public policy. Lanham,
MD: University Press of America.
McNally, R., & Griffin, A. (2004). Firm and individual choice drivers in make-or-
buy decisions: A diminishing role for transaction cost economics? Journal o f
Supply Chain Management, 40,4-18.
McNulty, E. (1999). Evaluating the use of third parties to measure process safety
management in small firms. Presentation at 1999 Annual Symposium at
Texas A&M University, College Station, Texas.
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
McNulty, P. (2001, February 27). Minutes o f the Roundtable Meeting held January
26, 2001, on the use o f third party auditors to ensure the safe handling o f
hazardous chemicals. Philadelphia: Wharton School, University of
Pennsylvania.
Miller, H., & Simmons, J. (1998). The irony of privatization. Administration &
Society, 3(^513-532.
Milward, B. (1994). Implications of contracting out: new roles for the hollow state.
In Patricia Ingraham & Barbara S. Romzek (Eds.), New paradigms for
government: Issues for the changing public service (pp. 41-62). San
Francisco: Jossey-Bass.
Milward, B. (1996). Symposium on the hollow state: capacity, control and
performance in interorganizational settings. Journal o f Public Administration
Research and Theory, 6, 193-195.
Mintzberg, H. (1996). Managing government, governing management. Harvard
Business Review, 74, 75-83.
Moe, R. (1994). The reinventing government exercise: Misinterpreting the problem,
misjudging the consequences. Public Administration Review, 54, 111-122.
Moe, R. (1996). Managing privatization: A new challenge to public administration.
In B. Guy Peters & Bert A. Rockman (Eds.), Agenda for excellence 2:
Administering the state. Chatham, NJ: Chatham House.
Normanton, E. (1996). The accountability and audit o f government. Manchester,
UK: Manchester University Press.
Olds, K. (1999). Public export inspections in the United States and their
privatization. Cato Journal, 19, 17-37.
O’Rourke, D. (2003). Outsourcing regulation: Analyzing nongovernmental systems
of labor standards and monitoring. Policy Studies Journal, 31, 1-29.
Osborne, D., & Gaebler, T. (1992). Reinventing government. Reading, MA:
Addison-Wesley.
Osborne, D., & Plastrik, P. (1997). Banishing bureaucracy: The five strategies for
reinventing government. New York: Addison-Wesley Publishing Co., Inc.
Peters, B., & Pierre, J. (1998). Governance without government? Rethinking public
administration. Journal o f Public Administration Research and Theory, 8,
223-243.
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pollitt, C. (1990). Managerialism and the public service. Oxford, UK: Blackwell.
Prager, J. (1994). Contracting out government services: Lessons from the private
sector. Public Administration Review, 54, 176-184.
Ramanadham, V. V. (1994). Privatization and after: monitoring and regulation.
New York: Routledge.
Rehfuss, J. (1989). Contracting out in government. San Francisco: Jossey-Bass.
Romzek, B., & Dubnick, M. (1987). Accountability in the public sector: Lessons
from the challenger tragedy. Public Administration Review, 47, 227-238.
Saltman, R., & Von Otter, C. (1992). Planned markets and public competition:
Strategic reform in northern European health systems. Buckingham, UK:
Open University Press.
Savas, E. (1987). Privatization: the key to better government. Chatham, NJ:
Chatham House Publishers.
Sclar, E. (2000). You don’ t always get what you pay for: The economics o f
privatization. Ithaca, NY: Cornell University Press.
Shin, R., & Chen, Y. (2000). Seizing global opportunities for accomplishing
agencies’ missions: the case of ISO 14000. Public Administration Quarterly,
24, 69-94.
Shleifer, A. (1998). State versus private ownership. Journal o f Economic
Perspectives, 12, 133-150.
Sparrow, M. (1994). Imposing duties: government’ s changing approach to
compliance. Westport, CT: Praeger.
Sparrow, M. (2000). The regulatory craft: Controlling risks, solving problems, and
managing compliance. Washington, DC: The Brookings Institution.
Telser, L., & Higinbotham, H. (1977). Organized futures markets: Costs and
benefits. Journal o f Political Economy, 85, 969-1000.
Uche, C. (2001). The theory of regulation: A review article. Journal o f Financial
Regulation and Compliance, 9, 67-80.
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
U.S. Environmental Protection Agency (EPA). (2001). Third party audit pilot
project in the Commonwealth o f Pennsylvania. Final Report. Philadelphia:
U.S. EPA, Region III.
U.S. General Accounting Office. (1986). Medical devices: Early warning o f
problems is hampered by severe underreporting (GAO Publication No. B-
223710). Washington, DC: U.S. Government Printing Office.
U.S. General Accounting Office. (1997). Privatization: Lessons learned by state and
local governments (GAO/GGD97-48). Report to the Chairman, House
Republican Task Force on Privatization. Washington, DC: U.S. Government
Printing Office.
U.S. General Accounting Office. (1998). Social service privatization: expansion
poses challenges in ensuring accountability for program results
(GAO/HEHS98-6). Report to the Chairman, House Subcommittee on
Human Resources, Committee on Government Reform and Oversight.
Washington, DC: U.S. Government Printing Office.
Van Slyke, D. (2003). The mythology of privatization in contracting for social
services. Public Administration Review, 63, 296-316.
Viscusi, W., & Gayer, T. (2002). Safety at any price? Regulation, 25154-63.
Wallin, B. (1997). The need for a privatization process: Lessons from development
and implementation. Public Administration Review, 46, 285-291.
Wallis, J., & North, D. (1986). Measuring the transaction sector in the American
economy 1870-1970. In S. L. Engerman & R. E. Gallman (Eds.), Long-term
factors in American economic growth. Chicago: University of Chicago
Press.
Watking, S. (2002, December). Plans for job competition. Federal Times.
Williamson, O. (1981). The economics of organization. American Journal o f
Sociology, 87, 548-577.
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
LIST OF STUDY INTERVIEWEES AND DISCUSSANTS
1. Director, Office of Devices
Blood and Tissues
Australian Therapeutic Goods Administration, 2001 and 2002
2. National Manager,
Therapeutic Goods Administration, Australia, 2000-2002
3. Director, Medical Devices Bureau
Therapeutic Products Directorate,
Health Canada, 2000-2002
4. Unit Head for Medical Products, Directorate - General Enterprise, European
Commission, 2002
5. Deputy Executive, Medical Devices Agency
United Kingdom 2001-2003
146
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
INTERVIEW/DISCUSSION AREAS AND QUESTIONS
The following questions and topics will be used to discuss the framework of your
audit/inspection program, experiences with your system for auditing/inspecting and
the accountability government, and the market exercise under your audit/inspection
system.
Any written or background materials that you wish to share before, during or after
the meeting will be greatly appreciated.
• Describe the process that is used in your country to audit/inspect medical
devices for manufacturing quality (e.g., authorities, responsibilities at
various levels in the process, how the process addresses noncompliant
manufacturers).
• Discuss the difficulties in implementing your program (e.g., resource,
authority).
• Discuss any advantages or disadvantages of using your system.
• Describe and discuss your organization’s responsibility/role in this audit
process (size, staffing, direct and indirect costs of auditing/inspecting).
• Discuss the manufacturing market in your country (e.g., size, types of
product produced internally/externally, relationship with government).
• Discuss your experiences with contracting and/or third parties.
• Describe your organization’s involvement in the standards development
program.
• Describe the accountability mechanisms in your country’s system of
regulating medical device manufacturing quality.
• Discuss your stakeholder interaction and communication.
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
A search for theory: Performance management to improve transportation safety
PDF
Federal mentor -protege programs: A pilot study on protege survival and growth
PDF
An analysis of health risk selection and quality of care under Medicare fee -for -service and Medicare managed care health care systems
PDF
Defending against bioterrorism: Lessons from the 2001 anthrax attacks
PDF
A preliminary assessment of the influence of certain variables on the implementation of recommendations by scientific -technical advisory committees
PDF
Care management for the uninsured: A force field analysis of the business case
PDF
A financial audit model for entrepreneurial governments
PDF
Assessing ombudsman performance: Two case studies in North America
PDF
Determining risk propensity of government program managers for high risk /high payoff projects
PDF
A configuration study of multiagency partnerships as practiced in Taipei City government
PDF
Education reform: Variables associated with establishing policies and authorizing funding for elementary and secondary education
PDF
Educational reform for private state -approved schools in California
PDF
Executive development and effectiveness: A study of mobility assignment experiences of career executives in federal research and development agencies
PDF
"Are we really in charge?": An analysis of the principal -agent relationship
PDF
Interoperability among complex defense computer -based systems: The Navy case
PDF
Banking on the commons: An institutional analysis of groundwater banking programs in California's Central Valley
PDF
Getting 'how' and 'why' straight: A critical discourse analysis of the National Partnership for Reinventing Government's ideological discourse on information and communication technologies
PDF
Adapting and applying a mission-focused strategic framework for emergency management
PDF
Executive spending power: Flexibility in obligation and outlay timing as a measure of federal budgetary and policy control
PDF
"Focused free thinking" in military intelligence analysis: Lessons from best practices
Asset Metadata
Creator
Gill, Lillian J.
(author)
Core Title
An analysis of third -party regulatory audit systems for medical device safety
School
School of Policy, Planning and Development
Degree
Doctor of Public Administration
Degree Program
Public Administration
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
health sciences, public health,OAI-PMH Harvest,Political Science, public administration
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Graddy, Elizabeth (
committee chair
), Daniel, Beverly (
committee member
), Wholey, Joseph (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-569112
Unique identifier
UC11335767
Identifier
3155410.pdf (filename),usctheses-c16-569112 (legacy record id)
Legacy Identifier
3155410.pdf
Dmrecord
569112
Document Type
Dissertation
Rights
Gill, Lillian J.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
health sciences, public health