Biomedical Research: HHS Direction Needed to Address Financial	 
Conflicts of Interest (26-NOV-01, GAO-02-89).			 
								 
Recent allegations that conflicts of interest may have affected  
biomedical research and led to harming human research subjects	 
have heightened concerns about the financial relationships	 
between individual research investigators or their research	 
institutions and private industry. These financial partnerships  
have yielded significant achievements, notably treatments for	 
diseases and conditions such as acquired immunodeficiency	 
syndrome (AIDS) and strokes. Their benefits notwithstanding, some
collaborations have raised concerns that research investigators  
or institutions with significant financial interests in the	 
research may focus attention on the financial rewards,		 
compromising their integrity and the safety of human subjects. In
such situations, the significant financial interest presents a	 
conflict of interest. GAO reviewed five universities which have  
developed broad policies and procedures regarding individual	 
investigators' financial conflicts of interest. All five had	 
difficulty providing basic data on investigators' financial	 
conflicts of interest in clinical research involving human	 
subjects. The universities acknowledged a need for better	 
coordination of information about investigators' financial	 
relationships, and several of the universities were developing	 
mechanisms to do so. Policies and procedures at the five	 
universities addressed aspects of institutional financial	 
conflicts of interest, such as technology transfer activities and
financial relationships with small start-up companies that market
university-developed products. However, Department of Health and 
Human Services (HHS) regulations and its oversight of financial  
conflicts of interest in biomedical research have limitations for
promoting the integrity of research and protecting human	 
subjects. Recently, HHS has taken steps to improve its oversight 
and monitoring. Further, HHS has drafted guidance on financial	 
conflicts of interest. However, this guidance does not provide	 
detailed advice on managing institutional financial conflicts of 
interest.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-02-89						        
    ACCNO:   A02485						        
  TITLE:     Biomedical Research: HHS Direction Needed to Address     
Financial Conflicts of Interest 				 
     DATE:   11/26/2001 
  SUBJECT:   Biomedical research				 
	     Colleges and universities				 
	     Conflict of interests				 
	     Federal regulations				 
	     Safety						 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-02-89
     
Report to the Ranking Minority Member, Subcommittee on Public Health,
Committee on Health, Education, Labor, and Pensions, U. S. Senate

United States General Accounting Office

GAO

November 2001 BIOMEDICAL RESEARCH

HHS Direction Needed to Address Financial Conflicts of Interest

GAO- 02- 89

Page i GAO- 02- 89 Financial Conflicts in Biomedical Research Letter 1

Results in Brief 3 Background 5 Universities Developed Broad Conflict- of-
Interest Policies for

Investigators That Varied in Implementation 11 Although Not Required,
Universities Had Policies and Procedures

That Addressed Aspects of Institutional Financial Conflicts of Interest 18
Current Regulations and Oversight Have Limitations for Promoting

Research Integrity and Human Subjects Protection 22 Conclusions 29
Recommendations for Executive Action 29 Agency Comments 30

Appendix I Federal Regulations Pertaining to Financial Interests in Research
32

Appendix II Scope and Methodology 33 Sample Selection 33 Site Visits 34

Appendix III Comments From the Department of Health and Human Services 35

Appendix IV GAO Contact and Staff Acknowledgments 40 GAO Contact 40 Staff
Acknowledgments 40

Related GAO Products 41 Contents

Page ii GAO- 02- 89 Financial Conflicts in Biomedical Research

Abbreviations AAMC Association of American Medical Colleges AAU Association
of American Universities AIDS acquired immunodeficiency syndrome AUTM
Association of University Technology Managers FDA Food and Drug
Administration HHS Department of Health and Human Services IRB institutional
review board NHRPAC National Human Research Protections Advisory Committee
NIH National Institutes of Health OER Office of Extramural Research OHRP
Office for Human Research Protections PHS Public Health Service

Page 1 GAO- 02- 89 Financial Conflicts in Biomedical Research

November 26, 2001 The Honorable Bill Frist Ranking Minority Member
Subcommittee on Public Health Committee on Health, Education, Labor, and
Pensions United States Senate

Dear Senator Frist: Recent allegations that conflicts of interest may have
affected the integrity of biomedical research and led to harming human
research subjects have heightened concerns about the financial relationships
between individual research investigators or their research institutions and
private industry. The Bayh- Dole Act, enacted in 1980, facilitated research
collaborations between private industry and research institutions such as
universities, and since then, financial relationships between them have
grown. 1 In biomedical research, these relationships often pair academic
research expertise and facilities with industry resources for technology
transfer in order to bring innovations from the laboratory into practical
medical application. The increase in financial collaborations has paralleled
2 decades of rapid growth in federal and private biomedical research
spending, now reaching into the billions of dollars each year. 2 These
financial partnerships and research funding have yielded significant
achievements, notably treatments for diseases and conditions such as
acquired immunodeficiency syndrome (AIDS) and strokes.

Their benefits notwithstanding, some collaborations have raised concerns
that research investigators or institutions that have significant financial
interests in the research may focus attention on the financial rewards of
the research, compromising its integrity and the safety of human subjects.
In such situations, the significant financial interest presents a conflict
of interest.

1 Amendments to the Patent and Trademark Act, P. L. 96- 517 sect. 6( a), Dec.
12, 1980, 35 U. S. C. sect.sect. 200- 212. Section 6( a) is commonly referred to as
?Bayh- Dole? after its two main sponsors, former Senators Birch Bayh and
Robert Dole.

2 The federal government funds biomedical research primarily through the
Department of Health and Human Services? (HHS) Public Health Service with
grants from the National Institutes of Health. Nongovernment (private)
funding comes from sources such as drug or biotechnology companies.

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 02- 89 Financial Conflicts in Biomedical Research

HHS regulations pertaining to financial interests in biomedical research are
divided into rules covering federally funded research and rules governing
privately funded, federally regulated research. For HHS- funded research,
institutions receiving grants from the National Institutes of Health (NIH)
must abide by a Public Health Service (PHS) regulation governing individual
investigators? financial interests. 3 Sponsors of publicly or privately
funded research on drugs, medical devices, or biological products, such as
vaccines, regulated by the Food and Drug Administration (FDA) must abide by
FDA?s regulation concerning investigators? financial interests when filing a
marketing application. 4 (See app. I for a description of federal
regulations pertaining to the financial interests of research
investigators.) While these regulations govern the financial interests of
individual investigators, no similar federal regulations apply to the
financial interests of an institution.

HHS? Office for Human Research Protections (OHRP) plays a role in enforcing
the HHS financial interest regulations through its oversight of
institutional review boards (IRB), which review research involving human
research subjects. 5 OHRP, which oversees all research conducted or funded
by HHS that involves human subjects, is responsible for enforcing the HHS
human subjects protection regulations. 6

Because of your concerns about the risks that financial conflicts of
interest pose to the integrity of biomedical research and the well- being of
human research subjects, you asked us to examine (1) how academic research
institutions are implementing HHS? regulations governing

3 PHS, an operating division within HHS, includes agencies such as NIH, the
Centers for Disease Control and Prevention, and the Food and Drug
Administration; and program offices under the Office of Public Health and
Science such as the Surgeon General, Emergency Preparedness, and Minority
Health.

4 A sponsor is defined as the party that supports a particular study at the
time it was carried out. (21 C. F. R. sect. 54.2 (h)). Marketing applications
are often filed by sponsors, but in some cases applicants may contract out
for covered studies.

5 Institutions conducting research using human subjects must have their IRBs
review and approve research proposals. Usually, the majority of IRB members
are affiliated with the institution. Some institutions rely on commercial or
independent IRBs.

6 The human subjects protection regulations (45 C. F. R. pt. 46) are a set
of requirements embodied in regulation for the protection of human research
subjects; they apply only to HHS funded or HHS sponsored research. FDA has
comparable regulations for FDA regulated research (21 C. F. R. pts. 50, 56).
Other federal agencies involved in human subjects research have similar
regulations.

Page 3 GAO- 02- 89 Financial Conflicts in Biomedical Research

individual investigators? financial interests, (2) the types of policies and
procedures these institutions have to minimize and manage institutional
financial conflicts of interest, and (3) the HHS regulations and oversight
intended to ensure that financial conflicts of interest do not affect the
integrity of research and the safety of human research subjects.

To address these objectives, we reviewed the HHS regulations on research
investigators? financial interests and the protection of human research
subjects; interviewed officials at NIH, FDA, OHRP, and experts at academic
associations (including the Association of American Medical Colleges and the
Association of American Universities) and other relevant organizations
(including the National Bioethics Advisory Commission); and visited five
universities to study in- depth their financial conflict- ofinterest
policies and processes. 7 We selected these universities from among the 20
institutions that receive the largest funding from NIH for biomedical
research and that have a high degree of technology transfer activity (that
is, the patenting and licensing of new technologies). Although our focus was
financial conflicts of interest in biomedical research, at each university,
we reviewed universitywide policies and procedures on financial conflicts of
interest and related matters and interviewed administrative officials. We
also interviewed biomedical research investigators, and we reviewed a sample
of financial disclosure files of biomedical research investigators. Because
we looked at a relatively small sample of the 483 institutions of higher
education receiving research funding from NIH in fiscal year 1999, our
findings are not generalizable to other universities or nonacademic research
institutions. We performed our work from February through September 2001 in
accordance with generally accepted government auditing standards. (App. II
describes our scope and methodology in greater detail.)

The five universities in our study had developed broad policies and
procedures regarding individual investigators? financial conflicts of
interest as required by the PHS regulation that, for the most part, applied
to both publicly and privately funded research. Federal requirements on
disclosure of financial interests and management of conflicts of interest
are flexible to allow institutions to implement them in ways that meet their

7 The five universities are the University of California, Los Angeles;
University of North Carolina, Chapel Hill; University of Washington,
Seattle; Washington University, St. Louis; and Yale University, New Haven.
Results in Brief

Page 4 GAO- 02- 89 Financial Conflicts in Biomedical Research

individual circumstances. Consequently, the universities? policies differed
in their content, such as the kinds of financial relationships they
considered to be manageable conflicts of interest, and in their
implementation. In addition, although they all used similar management
strategies for conflicts, they differed in how they employed them. The data
for overseeing various aspects of investigators? research activities and
financial interests were kept in multiple offices, files, and formats within
each of the five universities, making it a challenge to ensure that
conflicts of interest were appropriately managed and not overlooked. All
five universities had difficulty providing basic data on investigators?
financial conflicts of interest in clinical research involving human
subjects. The universities generally acknowledged a need for better
coordination of information about investigators? financial relationships,
and several of the universities told us they were developing mechanisms to
do so. University officials said they would like to have access to
information from HHS and other institutions about institutions? policies and
implementation practices, which could help them improve their practices.

Policies and procedures at the five universities addressed aspects of
institutional financial conflicts of interest, such as technology transfer
activities and financial relationships with small start- up companies that
market university- developed products. For example, they established a

?firewall? between the overall management of the institutions? investments
and their academic affairs, including research activities, thereby
encouraging independent decisionmaking in each area. Policies and procedures
in the areas of technology transfer and university- related startup
companies varied considerably. Two of the five universities, for example,
separated their technology transfer offices organizationally from other
research activities, which officials said helped keep ongoing research from
affecting revenue- generating decisions. In addition, the universities
placed limits on the amount of their equity holdings- ranging from 2 percent
of a company?s equity at one university to 49 percent at another- and on
their roles in the management of university- related startup companies.

The HHS regulations and its oversight of financial conflicts of interest in
biomedical research have limitations for promoting the integrity of research
and protecting human subjects. First, the HHS financial interest
regulations- both the PHS and FDA regulations- are not directly linked to
the regulations on human subjects protection, which means that financial
interest information may not necessarily be conveyed to IRBs for
consideration when they review research proposals for risks to human
subjects. Second, the PHS and FDA financial interest regulations differ in

Page 5 GAO- 02- 89 Financial Conflicts in Biomedical Research

terms of when they require review of investigators? financial disclosures,
and in the amounts of their disclosure thresholds. Institutions are required
to report investigators? financial conflicts of interest to PHS before
research funds are spent; FDA is not involved in reviewing financial
disclosures until a sponsor submits that information as part of a marketing
application, which is after the research has been completed. Third, although
the PHS regulation does not require institutions to report the specific
details of a financial conflict of interest, some of the universities we
visited were still confused about the conditions under which they needed to
report and what they needed to report. One university, for example,
mistakenly assumed it needed to report only the financial conflicts of
interest that could not be managed; therefore, if it had eliminated a
conflict of interest, it did not report it. Recently, HHS has taken steps to
improve its oversight and monitoring. NIH has included a review of financial
conflict- of- interest issues in its site visits to institutions, and FDA
now lists the review of financial disclosures in its guidance to reviewers
of drug marketing applications. In addition, HHS has drafted guidance on
financial conflicts of interest that is promising. However, this guidance
does not provide detailed advice on managing institutional financial
conflicts of interest.

We are recommending that HHS undertake efforts to highlight and communicate
best practices for institutions to identify and manage investigator and
institutional financial conflicts of interest. We are also recommending that
HHS develop specific guidance or regulations to address institutional
financial conflicts of interest. In its comments on a draft of this report,
HHS concurred with our recommendations.

Enormous growth in government and private biomedical research funding and in
financial relationships between government- funded investigators and private
industry has increased the potential for financial conflicts of interest to
occur that could compromise research integrity and the safety of
participants. HHS has regulations on individual investigator financial
interests in federally funded or regulated research. The academic and
professional communities also have developed policies and guidelines on
conflicts of interest and have recently devoted resources to study this
issue in more depth. Background

Page 6 GAO- 02- 89 Financial Conflicts in Biomedical Research

The budget of NIH, the principal federal agency that funds biomedical
research, grew from a little over $3 billion in fiscal year 1980 to more
than $20 billion in fiscal year 2001. Most NIH grants and contracts are
awarded through universities and medical centers to investigators conducting
research at these institutions. Private industry funding grew even more
rapidly- funding by drug companies alone rose from $1.5 billion in 1980 to
$22.4 billion in 2000. 8 Industry sponsors of biomedical research either
conduct the research themselves or provide the funding to university
investigators, other research institutions, contract research organizations,
or private medical practices. Collaborations between government- funded
research investigators and private industry also have increased, in part
because of the Bayh- Dole Act. The act gave universities, nonprofit
corporations, and small businesses the ability to retain patents and
commercialize their federally funded inventions in order to facilitate the
commercialization of new technologies. University- generated patents rose
from about 250 per year before 1980 to more than 4,800 in 1998.

As the boundary between academia and industry has become less distinct,
concerns have been raised about the potential for financial conflicts of
interest in investigators? as well as institutions? relationships with
private industry. Investigators? financial relationships with outside
interests can include working, contracting, or consulting for a company;
holding a management position or board membership or having other fiduciary
relationships; or owning stock or other securities. A conflict of interest
occurs when these relationships compromise, or appear to compromise, an
investigator?s professional judgment and independence in the design,
conduct, or publication of his or her research. For example, financial
conflicts of interest may affect the recruitment of human research subjects
such that inappropriate participants are enrolled. These conflicts also may
influence the informed consent process- by which the risks and benefits of a
study are communicated to the participants- resulting in participants who
are not fully informed about a study?s potential harm to them. Furthermore,
an investigator?s financial stake in a product may bias the development and
reporting of research results or make the investigator reluctant to share
information with other investigators in order to maintain his or her
competitive edge. Financial conflicts of interest could bias the publication
of research findings. For example, a corporate sponsor of research with a
vested financial interest in the study outcome

8 PhRMA, Pharmaceutical Industry Profile 2000. Growth in Biomedical

Research Funding and Collaborative Relationships

Investigator and Institutional Financial Conflicts of Interest

Page 7 GAO- 02- 89 Financial Conflicts in Biomedical Research

may try to ensure that only findings favorable to the sponsor?s product are
published.

Institutional financial conflicts of interest may arise because of an
institution?s desire to participate in technology transfer activities and
its need to remain financially sound. While companies may invest in
universities by supporting positions such as endowed chairs or facilities
such as research laboratories, universities also may invest financial
resources in companies that sponsor research at the institution. Such
investments would include owning stock in a pharmaceutical company or
investing in a small start- up company formed by entrepreneurial faculty who
have invented products and want to market them commercially. Startup
companies are generally nonpublicly traded enterprises. An investor?s
financial stake in a start- up may result in future financial gain.
Sometimes, however, an institution?s economic goals may conflict with its
goals of fostering objective, unbiased research. Financial interests may
color its review, approval, or monitoring of research conducted under its
auspices or its allocation of equipment, facilities, and staff for research.
For example, in a case that came to light in the late 1980s, the president
of one large university provided venture capital equal to one- fifth of the
university?s endowment (funds that support the university) to invest in a
biotechnology start- up company that used technologies the university
developed, with the university consequently holding more than 70 percent of
the company?s equity. The company also had university officials on its board
of directors and conducted research through the university. Because of these
ties, university decisions about research were inappropriately commingled
with financial decisions about the start- up company.

Within HHS, responsibility for the oversight of federally funded or
regulated biomedical research rests primarily with three entities: NIH, FDA,
and OHRP. NIH is charged with ensuring that the research it funds complies
with applicable HHS regulations, including a PHS regulation on individual
investigators? financial interests. 9 This regulation, promulgated in 1995,
requires PHS- funded organizations or institutions (which include all NIH-
funded organizations) to maintain and enforce written policies on financial
conflicts of interest; inform their investigators of these policies; and
require investigators to disclose any ?significant financial interests? in

9 42 C. F. R. pt. 50 Subpart F. Federal Oversight and

Regulations

Page 8 GAO- 02- 89 Financial Conflicts in Biomedical Research

entities whose financial interests may be affected by the research. While
the PHS regulation uses the phrase ?conflict of interest? without defining
it, the regulation defines a ?significant financial interest? as including
income of an investigator or investigator?s spouse or dependent child
expected to exceed $10,000 over 12 months, or equity interests exceeding
$10,000 or 5 percent ownership of a company. It is left to institutional
officials to determine which significant financial interests constitute
conflicts of interest. Institutions must report a financial conflict of
interest to the PHS awarding component and explain whether the conflict has
been

?managed, reduced, or eliminated.? The PHS regulation does not define these
terms but provides several examples of strategies to be used. In practice,
the management of a financial conflict of interest includes strategies to
monitor any effects as well as to reduce or eliminate the financial
interest.

FDA is responsible for ensuring that the financial interests and
arrangements of clinical investigators do not interfere with the reliability
of data submitted to FDA in support of marketing applications for drugs,
biological products, or medical devices. Under FDA?s financial interest
regulation, 10 effective in 1999, sponsors submitting marketing applications
must certify that investigators did not have certain financial interests and
arrangements, or must disclose them. FDA uses this information in
conjunction with information submitted on the design and purpose of the
study, and information obtained through on- site inspections, to assess data
reliability. In contrast to PHS, FDA?s thresholds for financial interests
requiring disclosure include payments made by the sponsor of a study to the
investigator or his or her institution exceeding $25,000 (beyond the costs
incurred in conducting the study) or any equity interest an investigator has
in a publicly held company sponsoring the research that exceeds $50,000.

OHRP oversees all research conducted or funded by HHS that involves human
research subjects 11 and enforces the HHS regulations regarding the
protection of human subjects. 12 HHS? human subjects protection

10 Principally, 21 C. F. R. pt. 54, and applicable portions of parts 312,
314, 320, 330, 601, 807, 812, 814, and 860. 11 OHRP is in the Office of the
Assistant Secretary for Health. HHS established OHRP in June 2000 to assume
the human subject protection functions of the former Office for Protection
From Research Risks, which was part of NIH.

12 45 C. F. R. pt. 46.

Page 9 GAO- 02- 89 Financial Conflicts in Biomedical Research

regulations do not address directly the disclosure and management of
investigators? financial conflicts of interest. However, the regulations do
require a university?s IRB, which reviews research proposals involving human
research subjects, to weigh a study?s risks and benefits to participants,
and review the study?s participant consent form, as part of its review of
the research. Because financial conflicts of interest may affect the risk-
benefit analysis, the purpose of the IRB review implies consideration of
them. While the actual IRB review of a research proposal may not explicitly
consider financial conflicts of interest, IRBs have the right to request and
review information about investigators? financial interests that might pose
risks to subjects, and they may require an investigator to disclose
significant financial interests to the research subjects in the consent
form. The human subjects protection regulations also state that an IRB
member may not participate in the initial or continuing review of any
project in which he or she has a conflicting interest, except to provide
information requested by the IRB. 13

Unless biomedical research is federally funded or involves research or
products that need federal approval, it is not necessarily subject to the
HHS regulations and oversight pertaining to financial interests and human
subjects protection. A significant and growing amount of privately funded
biomedical research exists that is not under the purview of HHS regulations
and oversight.

The academic community and professional associations have demonstrated
concern about financial conflicts of interest in biomedical research for a
number of years and have taken steps to address this issue. In 1990, the
Association of American Medical Colleges (AAMC) issued a document that in
part defined institutional and individual responsibilities for dealing with
conflicts of interest in research and provided guidance to institutions in
developing policies and procedures to meet their unique situations and local
requirements. 14 In 1993, the Association of American Universities (AAU)
developed a framework for managing investigators? financial conflicts of
interest. 15 Also in 1993, the Association of Academic

13 45 C. F. R. sect. 46.107 (e), 21 C. F. R. sect. 56. 107 (e). 14 See Guidelines
for Dealing with Faculty Conflicts of Commitment and Conflicts of Interest
in Research (Washington, D. C.: AAMC, 1990). 15 See Association of American
Universities Framework Document on Managing Financial Conflicts of Interest
(http:// www. aau. edu/ reports/ FrwkCOI. html). Policy and Guidance From

Associations and Medical Journals

Page 10 GAO- 02- 89 Financial Conflicts in Biomedical Research

Health Centers convened a task force to study institutional financial
conflicts of interest and their management. Although this task force
produced a report, it did not develop specific guidelines on institutional
financial conflicts of interest. 16

More activity has occurred recently, partly because of concerns about
reports that financial conflicts of interest were associated with harm to
research participants. In April 2000, the American Society of Gene Therapy
adopted a policy strongly encouraging that its members have no equity, stock
options, or comparable arrangements in companies sponsoring a clinical
trial. Also in 2000, AAU formed the Task Force on Research Accountability,
which issued a report in June on improving the management of human subjects
protection systems. 17 In October 2001, the Task Force issued a report on
the management of individual and institutional financial conflicts of
interest, with specific guidelines and recommendations. 18 In 2001, AAMC
convened a task force of clinical investigators; patient representatives;
medical school, teaching hospital, and university leaders; and
representatives from industry, the legal community, and the media to study
the issue of conflicts of interest, update AAMC?s 1990 guidelines, and
develop new principles for addressing institutional financial conflicts of
interest.

Editors of the major medical journals also have expressed concern about the
competitive economic environment in which some clinical research is
conceived, study subjects are recruited, and data are analyzed and reported.
In response to these concerns, the International Committee of Medical
Journal Editors has revised and strengthened the section on publication
ethics in Uniform Requirements for Manuscripts Submitted to Biomedical
Journals: Writing and Editing for Biomedical Publication, which is a
reference widely used by individual journals as a basis for their editorial
policies. 19 As part of the document?s revised reporting requirements,
authors will need to disclose details of their own and the

16 See Conflicts of Interest in Institutional Decision- Making (Washington,
D. C.: Association of Academic Health Centers Task Force on Science Policy,
1994). 17 See Report on University Protections of Human Beings Who Are the
Subjects of Research (Washington, D. C.: AAU, 2000). 18 See Report on
Individual and Institutional Financial Conflict of Interest (Washington, D.
C.: AAU, 2001). 19 See Frank Davidoff and others, ?Sponsorship, Authorship,
and Accountability,? New England Journal of Medicine, Vol. 345, No. 11
(2001), pp. 825- 6.

Page 11 GAO- 02- 89 Financial Conflicts in Biomedical Research

sponsor?s role in a study. Some journals also may require the primary
authors to sign statements that they accept full responsibility for the
conduct of the trial, had access to the data, and controlled the decision to
publish.

The five universities we visited developed written financial conflict-
ofinterest policies for individual investigators that, for the most part,
extended to all publicly and privately funded research but varied in their
content and in how they were implemented. For example, the universities
differed in the kinds of financial relationships- such as paid consulting
and holding equity in a company- they considered to be manageable conflicts
of interest. In addition, some universities used formal monitoring
committees to manage conflicts, while one university allowed investigators
to develop self- management plans. The universities generally allowed
investigators to self- certify compliance with financial conflict-
ofinterest policies. Administrative data used to oversee investigators?
research activities and financial relationships at all five universities
were kept in various offices and in different databases. The universities
generally acknowledged a need for better coordination, and several of the
universities told us they were taking steps to develop these linkages.
Officials at some of the universities told us that they would like to have
access to information from HHS and other institutions that could help them
improve their practices.

The written financial conflict- of- interest policies at four of the
universities we visited extended beyond the requirements of the PHS
financial interest regulation to apply to all research conducted at the
universities, whether it was funded publicly or privately. The fifth
university?s written policy covered all publicly and privately funded
research except research sponsored by certain foundations and other
nonprofit organizations. Concern about actual, potential, or even perceived
conflicts of interest has led many other research institutions to develop
financial conflict- ofinterest policies that are broader than what the
federal regulation covers. A recently published survey of the top 100 NIH-
funded research institutions reported that more than 70 percent of the 89
respondents had written policies that were more extensive than the federal
regulation. 20

20 Mildred Cho and others, ?Policies on Faculty Conflicts of Interest at U.
S. Universities,?

Journal of the American Medical Association, Vol. 284 (2000), pp. 2203- 8.
Universities

Developed Broad Conflict- of- Interest Policies for Investigators That
Varied in Implementation

Universities? Conflict- ofInterest Policies Predated and Extended More
Broadly Than Federal Requirements

Page 12 GAO- 02- 89 Financial Conflicts in Biomedical Research

Four universities we visited had policies that predated the PHS regulation,
and they revised these policies following the regulation?s implementation in
1995. The fifth university developed its policy the year that PHS published
its regulation. In part because of the recent focus on conflict- ofinterest
issues, four of the five universities were in the process of reviewing and
revising their policies and procedures. These four universities had formed
task forces or working groups to assess their policies and procedures and
adapt them to the changing research environment.

The PHS regulation is flexible to allow institutions to implement it in ways
that meet their individual circumstances. The five universities had
differences in threshold amounts, timetables for disclosure, and processes
for disclosure. And, although they all used similar strategies to manage
financial conflicts of interest, they differed in how they employed them.
The extent of IRB involvement in the review of financial conflicts of
interest also varied, ranging from reviewing investigators? financial
disclosure documents to obtaining verbal information from investigators and
relying on informal exchanges between its members and the conflictof-
interest committee. All five universities, however, generally relied on
investigators to monitor their own compliance with the schools? financial
conflict- of- interest rules.

In addition to being shaped by the federal requirements, institutions?
policies and procedures also may reflect state laws, court cases, the
institution?s experiences with financial conflicts, its organizational
structure, and its technology transfer activity. For example, state ethics
laws influenced the policies at two universities we visited, and a court
case also influenced one of these universities? policies. 21 Four of the
universities had written policies with categories and classifications of
financial conflicts of interest. However, the fifth university?s written
policy did not have fixed rules about potential financial conflicts of
interest but instead listed 13 specific examples of activities that
represented actual, possible, or no conflict of interest.

21 As a result of this court case, the university was required to disclose
investigators? possible financial conflicts of interest in the consent form
when study participants were informed about the risks and benefits of a
study. Universities Varied in

Financial Conflict- ofInterest Policies and Management but Generally Allowed
Investigators to Self- Certify Compliance

Policies and Procedures

Page 13 GAO- 02- 89 Financial Conflicts in Biomedical Research

Policies at the five universities required research investigators to
disclose to the institution any significant financial interests. Three
universities set the threshold for disclosure at the same level as the PHS
requirement. Another set the threshold for publicly sponsored research at
the PHS level, while, for privately sponsored research, it set a separate
threshold of $250 in income or holdings. The remaining university set the
overall threshold for disclosure at the PHS level but had a more stringent
disclosure policy for investigators involved in clinical trials. To help
protect the interests of human research subjects, this university required
an investigator doing clinical research who has any financial interest in
the study to disclose it to the institution. Officials at one of the other
universities told us they also are considering whether to lower their
threshold from the PHS level for disclosure of financial interests in
clinical research. At four of the five universities, the overall proportion
of clinical researchers who disclosed a significant financial relationship
averaged 5 percent. At one university, these data were not readily
available.

The five universities differed in their timelines and processes for
disclosure of significant financial interests. Three of the universities
required an annual disclosure by research investigators, and two required
disclosure when a research proposal was submitted. All required updates
whenever there was a change in the investigator?s financial interests. Their
disclosure forms also varied, ranging from simply asking whether a
significant financial interest exists and what type of interest it is to
asking detailed questions about the nature and amount of the financial
interest. Several disclosure forms required supporting information to be
provided as an attachment or to be submitted later. All of the universities
took steps to preserve the confidentiality of personal information, with
some taking stronger measures than others. For example, while all five
limited review of disclosure forms to university officials or a designated
committee, one university redacted the names of investigators in the
disclosure forms before giving them to the conflict- of- interest committee
for review.

All five of the universities in our study had conflict- of- interest
committees that were responsible for the development and implementation of
financial conflict- of- interest policies and procedures. The configuration
of these committees and the extent of their involvement in the review of
disclosures varied. All five universities had universitywide committees that
handled the review of financial conflicts of interest. Three of these
universities had additional medical school conflict- of- interest
committees. At two of the five universities, either the chairperson of, or
staff to, the committee reviewed all disclosure forms and determined whether
the financial interest was a conflict, which would then need to be managed,

Page 14 GAO- 02- 89 Financial Conflicts in Biomedical Research

reduced, or eliminated; they referred complex cases to the full committee
for discussion and action. At another university, each department
chairperson reviewed the department?s investigator disclosures and forwarded
disclosures of activities that may be allowable or are presumptively not
allowable to the committee for further review. At the other two
universities, the committee members reviewed each financial disclosure.

We found some variation among the five universities in how their conflictof-
interest committees evaluated significant financial relationships. The
committees make these determinations in response to the PHS regulation,
which requires universities to decide whether a disclosed relationship
constitutes a financial conflict of interest that needs to be managed,
reduced, or eliminated. For example, one school?s policy stated that an
investigator conducting clinical research on a product he or she developed
that was licensed to an external organization in which the investigator had
equity or other direct relations might be permitted to continue with the
research after disclosure, with appropriate safeguards in place. But another
university?s policy stated that such a relationship would present serious
problems and that it would consider the relationship inappropriate unless it
could be managed very closely. In addition, while one university typically
allowed investigators who received grant funding to hold equity or receive
consulting fees from a company for which they were conducting clinical
research, another university strongly discouraged or limited this practice.

IRB involvement in the review of financial conflicts of interest also varied
at the five universities we visited. University officials told us that IRB
members, following federal regulations, recused themselves from reviewing
research protocols when they had a conflicting interest. At some of the
universities, the IRBs were more aware of investigators? financial interests
than at others. The IRB members at one university reviewed faculty financial
disclosure forms in detail as part of their review of the research protocol,
checked to make sure that all investigators associated with the grant had
filed disclosure forms, and, when appropriate, required disclosure to human
research subjects. At three other universities, the conflict- of- interest
committee was supposed to send the IRB a memo or report that summarized the
financial conflict and recommended a management strategy. At two of these
three universities, the IRB could overrule the management strategy the
conflict- of- interest committee recommended. At the third university, the
IRB did not have the authority to overrule a management strategy. The IRB at
the remaining university had no formal communication with the conflict- of-
interest committee;

Page 15 GAO- 02- 89 Financial Conflicts in Biomedical Research

instead, IRB members obtained verbal information about financial interests
from investigators. Officials at this university told us they also relied on
the overlapping membership between the conflict- of- interest committee and
the IRB to surface any issues regarding investigators? financial conflicts
of interest.

The universities we visited did one or more of the following to manage
financial conflicts of interest: (1) required disclosure, (2) monitored the
research, and (3) required divestiture of the financial interest. The
application of these strategies differed, however. Some universities had
fairly formal guidelines about when each strategy should be used, while
others applied the strategies on a case- by- case basis. For example,
officials at one university told us that the strategy used was sometimes
determined through negotiation and cooperation between the investigator and
the conflict- of- interest committee.

Disclosure of the financial interest can take different forms, depending on
the institution. One of the five universities we visited required all
investigators who reported financial interests to the institution to
disclose them in publications. The four remaining universities did not have
an across- the- board policy to require investigators to disclose financial
interests in publications, and some of the four decided on a case- by- case
basis. At two universities, if human research subjects were involved,
investigators had to disclose the interests to their study subjects. One of
these two universities required investigators to use specific language in
the consent form that described the investigator?s financial relationship
with the study sponsor. The other three universities decided on a case- by-
case basis whether investigators would be required to disclose financial
interests in the consent form.

Monitoring the research can involve establishing a formal monitoring
committee consisting of several faculty members who meet with the
investigator periodically to make sure that the significant financial
relationship is being handled appropriately and is not harming the integrity
of the research. For example, at one university, a subcommittee of the
medical school conflict- of- interest committee develops a ?monitoring

plan? for each case, outlining the composition, appointment, and
responsibilities of the monitoring committee. The plans are contingent upon
approval from the universitywide committee, and the subcommittee ensures
that the plan is carried out. Conversely, monitoring can be informal and
involve, for instance, an investigator designing a personal

?self- management? monitoring plan that satisfies the university?s
requirement for managing the financial conflict of interest. For example, at
Management Strategies

Page 16 GAO- 02- 89 Financial Conflicts in Biomedical Research

one university, an investigator with a significant financial interest in a
company designed a self- management monitoring plan that included limiting
the time spent with the company, keeping track of the time spent with this
company, and not allowing the company to be involved with the research
laboratory.

Divestiture of the financial interest is also an option, but several
universities told us that this strategy is infrequently imposed on
investigators and not often chosen by them. Of 111 investigators at four of
the universities we visited who had significant financial relationships with
industry in 2000, only 3 voluntarily divested their interests; none were
told to divest by their universities. 22 Some investigators with significant
financial interests may decide not to be involved in conducting the study,
but if they are the only ones with a key skill or knowledge for a particular
study, they may still want to play a role. For instance, an investigator in
a privately funded study at one university we visited was willing to
relinquish her rights as the head investigator on the project involving a
new surgical procedure but insisted that she be present in the operating
room during the surgery because of her expertise and understanding of the
procedure. Subsequently, the informed consent form was altered to reveal the
investigator?s financial interests; other investigator- initiated
safeguards, such as disclosure in publications of the investigator?s
financial interest, were put into place; and the investigator was permitted
to be present during the surgery.

Each of the five universities? written conflict- of- interest policies
stated that an investigator?s failure to comply with the policy, such as not
disclosing a significant financial interest or not following the required
management strategy, is cause for disciplinary action, ranging from fines to
termination of employment. University officials told us that they rarely
determined that sanctions were warranted. None of the five universities had
formal processes for verifying that individuals fully disclosed their
financial interests. Instead, some universities used informal methods for
identifying apparent inconsistencies, such as comparing disclosure forms
with those of prior years. They said they relied on investigators to comply
voluntarily with conflict- of- interest policies because they believed it
was important to have faculty support and maintain collegiality with the
investigators. Furthermore, some of the universities emphasized informing
faculty about

22 The fifth university was not able to report any information on individual
financial relationships. Compliance Enforcement

Page 17 GAO- 02- 89 Financial Conflicts in Biomedical Research

their financial conflict- of- interest policies, a requirement established
in the PHS regulation. To this end, for example, two of the universities had
incorporated financial conflict- of- interest education modules into their
investigator training.

The data for overseeing various aspects of investigators? research
activities and financial interests were kept in multiple offices, files, and
formats within each of the five universities, making it a challenge to
ensure that conflicts of interest were appropriately managed and not
overlooked. As part of our study, we asked the five universities to provide
some basic data on investigators? financial conflicts of interest in
clinical research involving human research subjects. All five of the
universities had difficulty providing the information requested, and one was
not able to provide any of the data. University officials told us it was
difficult to respond to our request because information on who received
funding to conduct clinical research, their financial disclosures and any
management strategies used in the event of a conflict, and the IRB?s review
of the research protocol was collected in different formats and maintained
in separate databases and files in various offices. In general, at these
universities, the conflict- of- interest committee or staff to the committee
maintained faculty disclosure forms; the grants and contracts office
maintained information about who receives funding from government and
nongovernment sources, and received reports when there was a financial
conflict of interest related to a grant; and the technology transfer office
had information about faculty relationships with industry because of its
role in helping faculty patent their inventions and license them. While
these entities serve distinct purposes, they have information that,
collectively, is important to managing investigators? financial conflicts of
interest.

Officials at the universities we visited generally acknowledged a need for
better coordination among their internal offices that have information about
and responsibility for investigators? financial relationships. They also
said that a centralized reporting system and integrated database for
financial interest information could help ensure that potential conflicts
are not overlooked and are monitored. Officials at several of the
universities reported they were beginning to develop these linkages.

Because the universities varied in their implementation of the federal
financial interest regulation, we observed different practices for reviewing
and managing financial conflicts of interest and saw that the universities
used different mechanisms for internal coordination and communication.
Research Administration

and Financial Relationship Information Kept in Multiple Locations and
Formats

Page 18 GAO- 02- 89 Financial Conflicts in Biomedical Research

Officials at some of the universities we visited expressed interest in
learning about best practices from HHS and other institutions for
identifying and managing financial conflicts of interest in biomedical
research, especially as they review and revise their policies.

While there are no federal regulations or guidelines on institutional
financial conflicts of interest or how to manage them, the universities we
visited had policies and procedures that addressed aspects of these issues,
such as the management of investment funds, technology transfer activities,
use of licensing income, and the acceptance of equity in start- up
companies. The five universities established a ?firewall? between the
overall management of university investments and academic affairs, including
research activities, by using professional investment managers. University
investments in small start- up companies, however, which sometimes occurred
as part of technology transfer activities, were more closely tied to
research activities. The universities had or were developing policies and
practices to mitigate or manage potential institutional financial conflicts
of interest in this area, but they varied considerably. One approach was to
separate organizationally the technology transfer office from other research
activities or to use other internal controls such as special advisory
committees to make decisions that otherwise could be influenced by ties to
either technology transfer or research activities. Another practice, which
all five universities used to varying degrees, was to limit the amount of
equity they accepted and the extent of their involvement in managing
university- related start- up companies.

The universities we visited established ?firewalls? to keep the management
of institutional investments separate from academic affairs, including
research activities. One university official told us that the organizational
barrier this created in large part prevented financial and academic
decisions from influencing one another. The five universities used
investment managers- either employees or contractors- who were responsible
for the university?s portfolio and day- to- day investment decisions. The
investment managers reported to an investment committee or directly to the
university?s board of directors. Generally, each university?s board of
directors had separate committees for investment and for academic affairs
that established policies and provided oversight. In addition, these
universities, in general, did not devote the university?s endowment to
investments in university- related start- up companies. At four of the five
universities, officials said that most investigators were not aware of
institutional investments, suggesting that decisions about these Although
Not

Required, Universities Had Policies and Procedures That Addressed Aspects of
Institutional Financial Conflicts of Interest

Universities Managed Institutional Investments Separately From Academic
Affairs

Page 19 GAO- 02- 89 Financial Conflicts in Biomedical Research

major university investments were distinct from day- to- day research
activities and academic affairs. However, at two of the five universities,
general information on how funds are invested, without specific amounts, is
available on the Internet. We were unable to readily locate such information
at the remaining universities.

In order to reduce opportunities for institutional financial conflicts of
interest, two of the universities organizationally separated the technology
transfer office from the research office, locating technology transfer
directly under the provost or vice provost, the chief academic officer of
the university. 23 Officials at one of these universities said that this
arrangement made it easier to manage institutional financial conflicts of
interest and that they believed the office of research should not be
influenced by technology transfer activities. The other universities located
their technology transfer offices under the vice provost for research or
vice chancellor for research. One university?s justification for locating
offices together was that communication was better when these offices were
organizationally aligned and that good communication would help prevent
financial conflicts of interest from occurring. Officials at another one of
these universities gave us an example of an internal control mechanism-
establishing an interdisciplinary committee to make an impartial decision
about which company is selected to license a product developed by a faculty
member- in order to avoid an institutional financial conflict of interest.

23 ?Technology transfer office? is a generic title; the universities we
visited gave this office various names, such as Office of Cooperative
Research and Center for Technology Management. This office encourages the
patenting and licensing of discoveries developed by faculty, students, and
staff, may help faculty obtain research support from corporate sponsors, and
also may be responsible for accepting equity in start- up companies or may
work with venture capital companies. At four universities, the technology
transfer office?s mission was to disseminate research knowledge and provide
services to the public. The remaining university stated that its office?s
mission was enhancing the university?s interaction with the private sector
and promoting new economic activities within the community. Some
Universities Tried to

Minimize Potential for Institutional Conflicts by Separating the Technology
Transfer Office From Academic Affairs

Page 20 GAO- 02- 89 Financial Conflicts in Biomedical Research

The five universities we visited had or were developing policies on
accepting equity in university- related start- up companies, such as
biotechnology companies. 24 During the technology transfer phase,
universities often accept equity from these companies in return for paying
patent and licensing fees. 25 The policies at the five universities varied
in their stringency. Four restricted the amount of equity they would accept
to a fixed proportion ranging from 2 percent to 20 percent. The remaining
university specified only that its equity position should not be greater
than 49 percent. One university?s policy stated that the university
generally requires having an equity position in a company when a faculty or
staff member develops technology in the course of university employment and
assists a business venture in the commercialization of that idea.

Four of the five universities reported that in fiscal year 1999 they spent
more in legal fees for technology transfer activities than they were
reimbursed through licensing agreements. Their technology transfer offices
provide a service to university faculty members and staff in facilitating
the transfer of technology to the private sector. 26 As one university
official said, faculty members should be able to pursue developing products
from their research even if they generate little or no profit. Consequently,
the universities said that they do not target opportunities for generating
profit and that most of their patents and licenses do not yield substantial
income. 27 The universities do not patent or license all inventions of their
faculty and staff, but they do assess whether the technology is worth the
investment and assign the rights to the researcher for those they decline to
patent or license.

24 All five universities have written intellectual property policies
specifying the allocation of the derived income for the university, the
inventor?s school or department, and the inventor. The proportion of the
licensing income allocated to the inventor ranged from 25 to 50 percent.

25 The cost of patenting and licensing a product averages $25,000 to
$50,000, and these companies may not have the cash available. 26 Written
policies on accepting equity in university- related start- ups state that
the purpose of accepting equity is to facilitate technology transfer and
that this purpose is secondary to the university?s primary missions of
education, research, and public service.

27 Licensing income constitutes a small portion of a university?s total
research income. Although the five universities ranked among the top
universities in licensing income, their licensing incomes ranged from less
than 1 percent to nearly 4 percent of their total research income. See The
Association of University Technology Managers, Inc., AUTM Licensing Survey,
FY 1999: Full Report (2000). Universities Placed Limits

on Their Equity Holdings and Roles in Start- Up Companies

Page 21 GAO- 02- 89 Financial Conflicts in Biomedical Research

Various parties are involved in the decision to accept equity holdings in a
university- related start- up. The universities we visited encouraged
faculty members and staff to disclose inventions to the technology transfer
office. The technology transfer staff review the disclosure to determine
both its commercial potential and its ownership. Most universities own
intellectual property, such as a patent, if significant university resources
were used or if it was developed through research conducted at the
university. The technology transfer office then attempts to find a private
company to license and underwrite the cost of developing and licensing the
product. At the early stage of product development, however, the commercial
potential of an invention is often uncertain. If no private company is found
to assume the financial risk for developing the product, the university may
consider taking an invention through the patenting and licensing process
itself and accept equity in payment from the company that will hold the
license. At four of the universities, the vice provost or vice chancellor
makes the decision to accept equity. At the remaining university, the
provost makes the decision. The school or department of the university that
employs the inventor also is often involved because, according to all five
universities? policies, it receives a portion of the licensing income. It
also may provide funds to license and develop the product.

After the decision to take equity, the university?s investment managers, who
are responsible for the university endowments and investments, then manage
the equity shares. University officials told us that once the equity is
transferred to these managers, they have virtually no other contact or
responsibilities for the equity. However, universities transfer the shares
to the investment managers at different times. The technology transfer
office at one university holds the equity until the company becomes public,
then transfers the equity to the university investment office. Another
university has guidelines for placing both individual and university
equities in escrow. Other universities transfer the equity after the
licensing agreement has been signed. In these cases, university officials
said that they are not sure what investment managers do with these holdings-
in particular, whether these proprietary holdings are managed differently
from other equity holdings.

The universities also restricted their involvement in the management of
university- related start- ups because of potential institutional financial
conflicts of interest in these ventures. Two universities we visited had
written policies that specified the university would not accept
representation on a start- up company?s board of directors, nor would it
exercise voting rights. Another university, however, reserved the right to
elect a member to the start- up?s board of directors. The member, in this

Page 22 GAO- 02- 89 Financial Conflicts in Biomedical Research

case, would be required to resign if the company registered with the
Securities and Exchange Commission for an initial public offering. The
remaining two universities had unofficial policies and are now reexamining
the appropriate roles and responsibilities of the university, such as using
nonpublic information to manage equity of a universityrelated start- up and
the role of the faculty member who established the start- up in the
university?s management of the equity.

In our review, we identified limitations with the HHS regulations and
oversight of financial conflicts of interest in biomedical research that
have implications for promoting the integrity of research and protecting
human research subjects. First, no direct link exists between the HHS
financial interest regulations and the human subjects protection regulations
with regard to the risks to human research subjects posed by investigators?
financial conflicts of interest. Second, although the PHS and FDA
regulations both address investigators? financial interests, PHS and FDA
conduct their reviews of this information at different points in the
research process and have different disclosure thresholds for what
constitutes a significant financial interest. Third, the universities we
visited indicated some confusion about what the PHS regulation specifically
required them to report to NIH. NIH and FDA have recently taken steps to
improve oversight and monitoring, such as conducting site visits, taking an
inventory of institutions? financial conflict- of- interest policies, and
providing guidance to reviewers of financial conflict- ofinterest
information. In addition, HHS has developed draft guidance on financial
relationships in clinical research, which is promising. However, this
guidance does not provide detailed advice on managing institutional
financial conflicts of interest.

No direct link exists between the HHS financial interest regulations and the
human subjects protection regulations. Such a link would help ensure that
IRBs are aware of financial conflicts of interest that might pose risks to
study subjects and would help minimize those risks. The PHS and FDA
financial interest regulations require disclosure to institutional officials
and to sponsors, but there is no mechanism to ensure that the disclosed
information reaches IRBs. And although the HHS human subjects protection
regulations require IRBs to evaluate research proposals for any foreseeable
risks the study might pose to human research subjects, they contain no
explicit provision that investigators disclose to IRBs their financial
interests. In our review of the five universities, we found that IRBs
learned about investigators? financial interests in various ways, Current
Regulations

and Oversight Have Limitations for Promoting Research Integrity and Human
Subjects Protection

No Direct Link Between HHS Financial Interest and Human Subjects Protection
Regulations

Page 23 GAO- 02- 89 Financial Conflicts in Biomedical Research

ranging from reviewing financial disclosures directly or receiving reports
from the conflict- of- interest committee to informally following up with
investigators. Without a direct link between the HHS financial interest and
human subjects protection regulations, either institutions are left to
develop their own ways to ensure that IRBs have information about financial
conflicts of interest or IRBs must seek out this information.

The timing of the disclosure of financial interests differs between the PHS
and FDA regulations. The PHS regulation requires institutions to report to
PHS the existence of any financial conflicts of interest before expenditures
are made, while FDA reviews investigators? financial interests only when the
sponsor submits a marketing application. The PHS regulation requires that
investigators receiving NIH funding must disclose to their institutions any
?significant financial interests? related to the research. The institution
then must determine whether a financial interest constitutes a conflict and,
if so, notify NIH that it exists and that it has been managed, reduced, or
eliminated. Through the PHS regulation, therefore, institutions and funding
agencies have an opportunity before research begins to protect human
research subjects from potential harm from investigator conflicts of
interest. But while the FDA regulation requires a clinical investigator to
disclose financial interests to the sponsor of a trial before beginning to
participate, FDA itself is not notified of financial interests that could
present a potential conflict of interest until this information is submitted
as part of a marketing application, which occurs after the research has been
conducted and research subjects have already participated. Although the IRB
is responsible for reviewing and minimizing risks to study subjects, the
timing of the disclosure of financial interests in the FDA regulation may
limit FDA?s ability to provide oversight of the process.

The timing of reports to FDA regarding financial interests is geared toward
the integrity of research findings. Since the objective of the FDA
regulation is ensuring data integrity for the purposes of product review,
the regulation focuses on payment arrangements and other financial interests
of clinical investigators that could introduce bias into studies. FDA told
us that it should be aware of such interests and arrangements as part of its
evaluation of marketing applications. An FDA official told us that FDA
expected the requirements for disclosure to help deter sponsors from hiring
or working with clinical investigators who have significant financial
interests that pose a conflict.

PHS and FDA also differ in their threshold amounts for disclosure of
financial interests. The PHS threshold- more than $10, 000 in expected PHS
and FDA Financial

Interest Regulations Are Not Uniform in Their Timing or Disclosure
Thresholds

Page 24 GAO- 02- 89 Financial Conflicts in Biomedical Research

income over 12 months or more than $10,000 in equity or 5 percent ownership
in a company- has not been updated for inflation since the regulation came
into effect in 1995. Some have expressed concern that the PHS threshold was
too low. For instance, in 1999, members of the NIH Regulatory Burden
workgroup stated that the PHS disclosure threshold was too low and could
trigger an excessive number of disclosures where there was no conflict that
needed to be managed. FDA?s thresholds- more than $25,000 in payments from
the sponsor of a clinical study to an investigator or an investigator
holding more than $50, 000 in equity in a publicly held company sponsoring
the research- are significantly higher than the PHS threshold.

The PHS regulation requires an institution to report that it has identified
a financial conflict of interest related to PHS- funded research and that it
has taken steps to manage, reduce, or eliminate it. Nevertheless, we found
that officials from the five universities were confused about the conditions
under which they needed to report to NIH and what they needed to report. At
the universities we visited, we found very few reports to NIH about
financial conflicts of interest. This could be because there were few
occurrences of significant financial interests involving NIH grants that
were deemed conflicts or because we could not determine from the reports
whether the universities had followed the reporting requirements. One
university operated under the mistaken assumption that it needed to report
only financial conflicts of interest that could not be managed; therefore,
it did not report them if they had been managed, minimized, or eliminated.
At another university, we found a case of clinical research involving human
subjects during our file review in which the university established a
management strategy for a financial conflict of interest but did not report
it to NIH. The university officials told us they had only reported two cases
to NIH since the regulation went into effect in 1995, and neither case
involved human research subjects.

In some instances, confusion about the requirements and concerns about
overreporting may lead to underreporting. Officials from two of the
universities told us they were confused about what they needed to report to
NIH. One university in our sample did not know whether it was Some Confusion
Exists

About PHS Reporting Requirements

Page 25 GAO- 02- 89 Financial Conflicts in Biomedical Research

responsible for reporting a conflict of interest if an investigator had an
NIH grant and the conflict was not related to that grant. 28

Confusion about reporting requirements also stems from the regulatory
silence regarding when financial interests should be viewed as posing a
potential conflict. Although the PHS regulation defines a significant
financial interest, it allows university officials to determine whether such
interests pose conflicts for investigators. Only those financial interests
meeting the minimum thresholds that are deemed to be conflicts of interest
must be reported. Thus, for example, at one of the universities, a
department head deemed that a financial relationship was not a material
conflict, even though it was considered a significant financial interest
under the PHS regulations.

NIH has taken steps recently to improve compliance with the financial
conflict- of- interest regulation by centralizing institutions? reports of
conflicts of interest at the Office of Extramural Research (OER), having OER
conduct site visits, and taking an inventory of institutions? financial
conflict- of- interest policies. NIH is responsible for ensuring that
institutions comply with the PHS regulation on financial conflicts of
interest. It may do this by reviewing an institution?s policies and
procedures on financial conflicts of interest, monitoring reports of
conflicts, conducting site visits, examining institutions? files, and
reviewing actions taken by institutions to manage financial conflicts of
interest. Institutions? reports of conflicts are sent to the funding
institutes and centers of NIH and are kept with the grant files. Because
these reports contain no details about the conflict and its management, NIH
program officials have little information to follow up on. NIH is authorized
to request more information about conflicts of interest from institutions,
but an official at NIH told us that NIH rarely seeks further information. In
late 2000, NIH?s institutes and centers began providing a copy of grantee
institutions? reports of financial conflicts of interest to OER, which
maintains summary data on conflicts of interest.

In fiscal year 2000, OER visited 10 institutions receiving NIH funding to
assess institutional understanding of NIH policies and requirements, and in

28 The PHS regulation requires a significant financial interest to be
reported only if it is related to the funding because it ?would reasonably
appear to be affected by the research for which PHS funding is sought? or if
the interest is ?in entities whose financial interests would reasonably
appear to be affected by the research.? NIH and FDA Have Taken

Steps to Improve Oversight and Monitoring

Page 26 GAO- 02- 89 Financial Conflicts in Biomedical Research

fiscal year 2001, OER visited 8 more institutions. 29 Financial conflict of
interest was one of many topics addressed. During the visits, the
institutions? officials discussed with NIH staff information in financial
conflict- of- interest files, including meeting minutes, documents, and
correspondence concerning how financial conflicts of interest had been
managed, reduced, or eliminated. In its findings and observations on the
site visits, NIH noted some of the concerns we have identified. For example,
NIH found that some institutions were confused about the definition of a
significant financial interest. In addition, some faculty expressed fear
that full disclosure of financial interests might result in limiting their
institutional salary or adversely affect NIH funding. NIH officials told us
that if they discovered a weakness during the visit, they provided guidance
and information to help the institution make appropriate improvements.

In January 2001, NIH asked 300 institutions with the largest amount of NIH
funding to send it copies of their financial conflict- of- interest policies
after officials learned that not all research institutions have an
investigator financial conflict- of- interest policy in place. A survey
published in 2000 of the 250 medical schools and other research institutions
with the highest NIH funding had found that 5 medical schools and 10 other
research institutions reported they did not have such a policy. 30 As of
September 2001, NIH had received policies from 293 of 300 grantee
institutions, and all of the top 100 funded institutions had a conflict of
interest policy in place. Officials at NIH said they plan to review the
policies they have collected to see if they contain all the required
elements.

FDA also recently has taken action to improve compliance with its financial
interest regulation by providing guidance for FDA reviewers of drug
marketing applications. FDA?s regulatory role allows it to review the
information in investigator financial disclosure reports in marketing
applications. If FDA determines that a financial interest of any clinical
investigator raises questions about the integrity of the data, FDA may audit
the data, ask the applicant to submit further analyses of the data or
conduct additional independent studies, or refuse to use the data from that

29 OER coordinates this work but supplements its staff with staff from other
NIH offices to carry out the work. 30 S. Van McCrary and others, ?A National
Survey of Policies on Disclosure of Conflict of Interest in Biomedical
Research,? New England Journal of Medicine, Vol. 343, No. 22 (2000), pp.
1621- 6.

Page 27 GAO- 02- 89 Financial Conflicts in Biomedical Research

study in support of the product application. Each of FDA?s centers
responsible for human drugs, biological products, and medical devices
determines how it will implement the financial interest regulation. 31 Until
recently, FDA did not provide systematic guidance to its reviewers about
evaluating investigator financial disclosure reports. One of FDA?s centers
has provided guidance by creating a clinical review template for drug
marketing application reviewers that includes brief guidance on reviewing
financial disclosures.

In December 2000 HHS developed draft guidance entitled ?Financial

Relationships in Clinical Research: Issues for Institutions, Clinical
Investigators, and IRBs to Consider When Dealing With Issues of Financial
Interests and Human Subject Protection: Draft Interim Guidance.? This
guidance drew on information obtained at a conference HHS held in August
2000 on financial conflicts of interest in clinical research 32 and comments
it received. 33 The document contains guidance for institutions, clinical
investigators, and IRBs to assist in their deliberations concerning
financial relationships and potential and real conflicts of interest. The
document is also intended to facilitate disclosure of such conflicts in
consent forms. This document was posted on the OHRP Web site on January 2001
but has not been published in the Federal Register. According to HHS
officials, the draft is being revised and will be published as ?points for
consideration.? While it provides promising guidance for identifying and
managing individual investigator financial conflicts of interest, it is
limited in its discussion of institutional financial conflicts of interest.
The draft guidance states that institutions should have policies and
procedures on institutional financial conflicts of interest; establish an
institutional conflict- of- interest committee to review potential conflicts
and their management when considering entering into business agreements; and
document and disclose to the IRB institutional financial relationships with
a commercial sponsor of a study. But the document does not provide detailed
guidance on the appropriate ways of addressing institutional conflicts of
interest, particularly institutional relationships with university- related
start- up companies.

31 The responsible centers are the Center for Drug Evaluation and Research,
the Center for Biologics Evaluation and Research, and the Center for Devices
and Radiological Health. 32 For a transcript of the conference, see http://
ohrp. osophs. dhhs. gov/ coi/ 8- 15. htm and http:// ohrp. osophs. dhhs.
gov/ coi/ 8- 16.htm. 33 See http:// ohrp. osophs. dhhs. gov/ humansubjects/
finreltn/ finmain. htm HHS? Draft Interim

Guidance on Financial Conflicts of Interest Is Promising but Limited With
Regard to Institutional Conflicts

Page 28 GAO- 02- 89 Financial Conflicts in Biomedical Research

HHS received 36 comments on its draft guidance from health care
professionals, institution officials, and representatives of the patient
community, FDA, and academic associations. Some members of the research
community expressed concern about the guidance?s usefulness and
appropriateness. These groups also commented that the academic community had
not yet fully discussed institutional financial conflicts of interest and
was still grappling with a definition. Some research community members
disagreed with giving responsibilities regarding financial conflicts of
interest to already overburdened IRBs, which could distract them from their
role of protecting human research subjects. Another stated that the draft
interim guidance emphasized academic institutions without taking into
account the perspective of other types of research centers, such as
hospitals and freestanding centers.

After reviewing the draft guidance and comments, the National Human Research
Protections Advisory Committee (NHRPAC), an advisory group to HHS,
recommended that the Secretary of Health and Human Services move to release
the guidance. NHRPAC also recommended that, in the absence of consistent
federal regulations, institutions should use the PHS threshold for
disclosure of financial interests but that, ultimately, the PHS and FDA
thresholds should be harmonized. All research subject to HHS regulations,
funded privately or publicly, then would be held to the same standards.
Steps toward harmonization, in NHRPAC?s view, would include regulatory
measures that go beyond the draft interim guidance. In addition, NHRPAC
stated that IRBs should not have to collect, analyze, and provide remedies
for financial conflicts of interest but should rely on a conflict- of-
interest entity (such as a committee or an individual charged with conflict-
of- interest review responsibilities) to handle the matters and report
formally to the IRB as part of the research application. NHRPAC supported
HHS? efforts to identify and define institutional financial conflicts of
interest and methods to manage them and suggested that such interests could
be disclosed to the institution?s conflict- of- interest entity. NHRPAC
recommended that specific, detailed information be provided in the informed
consent process when an actual conflict of interest has been identified
during financial disclosure and, in cases in which a potential conflict is
conceivable, to make general information about financial interests
available, with more detailed information available upon request. Finally,
NHRPAC recommended that institutions audit and monitor compliance with their
own institutional policies and procedures and develop and enforce
disciplinary standards for violations. The final version of the guidance is
scheduled for completion this fall.

Page 29 GAO- 02- 89 Financial Conflicts in Biomedical Research

The five universities in our study implemented the PHS regulation on
individual investigators? financial interests in different ways, and they
had or were developing policies and procedures to address aspects of
institutional financial conflicts of interest. The universities expressed
interest in learning about others? policies and procedures, such as how
investigators? financial disclosure information was communicated to IRBs or
ways the universities monitored financial conflicts of interest. Having
information on the best practices of institutions for dealing with
investigator and institutional financial conflicts of interest could help
institutions develop policies and procedures that would best meet their
needs.

HHS? proposed guidance on financial relationships in clinical research is
promising and will help institutions implement the PHS regulation on
investigators? financial interests. With some revision, this guidance could
link the HHS financial interest regulations with the human subjects
protection regulations, making sure that IRBs are aware of financial
conflicts of interest to help minimize risks to study subjects. However, the
guidance is limited in its treatment of institutional conflicts of interest.
As financial relationships between institutions and industry proliferate,
the need for guidance in this area increases.

Research institutions are not required to apply their financial conflict-
ofinterest policies and procedures, as the five we studied did, to both
publicly funded and privately funded research. Furthermore, a significant
and growing amount of biomedical research is now conducted outside of
universities by entities that may not be operating under broad financial
conflict- of- interest policies and procedures. Addressing potential
financial conflicts of interest in these other settings will be important to
ensure the integrity of research and the well- being of human research
subjects.

To ensure the integrity of biomedical research and the protection of human
research subjects, HHS needs to improve the implementation of its financial
interest regulations and its oversight of financial conflicts of interest.
Specifically, we recommend that the Secretary of Health and Human Services
take the following actions:

 Develop and communicate information on best practices for institutions to
consider for identifying and managing investigator and institutional
financial conflicts of interest in biomedical research.

 Develop specific guidance or regulations concerning institutional
financial conflicts of interest. Conclusions

Recommendations for Executive Action

Page 30 GAO- 02- 89 Financial Conflicts in Biomedical Research

HHS reviewed a draft of this report and provided comments, which are
included as appendix III. HHS said that the report gives a useful overview
of how some academic research institutions handle financial conflicts of
interest and clinical research issues. HHS concurred with our
recommendations.

With regard to our recommendation to develop information on best practices,
HHS stated that NIH has efforts under way to collect such information by
making site visits to institutions and analyzing financial conflict- of-
interest policies from institutions. NIH plans to post this information on
its Web site. Regarding our recommendation to develop guidance or
regulations concerning institutional conflicts of interest, HHS said that
NIH?s Regulatory Burden Reduction Committee has begun to address
institutional conflicts of interest. To the extent that specific policies or
guidance on human subjects protection and financial conflict of interest are
developed, HHS said it will be coordinated within the Department.

HHS made several specific comments. It noted that financial conflicts of
interest occur in the context of all areas of research, not just clinical
research. We agree with this assessment, but our report focuses on
biomedical research funded or regulated by HHS. HHS suggested that we expand
on the rationale for selecting the five universities in our report in order
to better explain the institutional variability we observed. We did not add
any information because we believe appendix II clearly states our selection
criteria and the sample is too small to draw conclusions about how specific
characteristics of the universities relate to policy differences.

HHS also noted that one reason NIH typically obtains only limited
information about financial conflicts of interest from institutions is that
any information NIH has about these matters would be subject to disclosure
under the Freedom of Information Act. We agree that financial details
disclosed by investigators to NIH potentially are subject to disclosure
under the Freedom of Information Act. However, as FDA has recognized in its
treatment of such information, the likelihood of such disclosures is slim,
and only when necessary to effect a public purpose that outweighs a
particular privacy interest. FDA decides such matters on a case- by- case
basis and has recognized that, in some cases, there may be legitimate public
interests in the financial information of investigators that warrants its
disclosure. In its comments, HHS also questioned the purpose for which
follow- up information would be gathered. We revised the report to avoid
implying that NIH should routinely seek further information and to emphasize
instead that NIH already has authority to obtain additional Agency Comments

Page 31 GAO- 02- 89 Financial Conflicts in Biomedical Research

information on the conflict of interest if it chooses to do so. We believe,
however, that there may be instances where NIH may need to know the nature
and details of a financial conflict of interest to determine whether it was
acted on appropriately.

HHS also stated that concerns remain that the PHS regulation on financial
interests does not specifically or adequately address the impact of
financial relationships on the interests and welfare of human subjects and
added that an IRB may not be the most appropriate body to consider financial
conflicts of interest. We have added a discussion about the absence of a
link between the HHS financial interest regulations and the human subjects
protection regulations. We agree with HHS that an IRB may not be the most
appropriate body to review investigators? financial interests and that an
IRB can also learn about any risks from conflicts of interest by receiving
information from a conflict- of- interest committee or by asking for
information directly from investigators.

HHS also provided technical comments, which we incorporated where
appropriate.

As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days from
the date of this letter. We will then send copies to the Secretary of Health
and Human Services, the Director of OHRP, the Acting Director of NIH, the
Acting Principal Deputy Commissioner of FDA, appropriate congressional
committees, and others who are interested. We will also make copies
available to others on request.

If you or your staff have any questions, please contact me at (202) 512-
7119 or Marcia Crosse at (202) 512- 3407. Other major contributors are
listed in appendix IV.

Sincerely yours, Janet Heinrich Director, Health Care- Public Health Issues

Appendix I: Federal Regulations Pertaining to Financial Interests in
Research

Page 32 GAO- 02- 89 Financial Conflicts in Biomedical Research

Agency, regulation, and effective date Applicable party Disclosure
requirements Reporting requirements

Public Health Service (including NIH) Regulation: Responsibility of
Applicants for Promoting Objectivity in Research for Which PHS Funding Is
Sought 42 C. F. R. pt. 50 Subpart F; effective Oct. 1, 1995

Institutions or individuals applying for PHS funding.

Significant financial interests of the investigator (or spouse or dependent
child):

Significant financial interests are defined as interests that would be
affected by research or in entities whose financial interests reasonably
appear to be affected by research, including equity interests exceeding
$10,000 or 5 percent ownership in a single entity; salaries, royalties, or
other payments (not from applicant institution) expected to total more than
$10,000 in the next year; and patents.

Required financial disclosures of investigators must be provided to the
institution by the time the grant application is submitted to PHS.

Grant applications to PHS must certify that the institution has implemented
a written and enforced administrative process to identify, manage, reduce,
or eliminate conflicting interests; that all conflicts have been reported;
and that each conflict will be managed, reduced, or eliminated before the
expenditure of PHS funds.

Investigators must update financial disclosure reports annually or as new
interests are obtained. Food and Drug Administration Regulation: Financial
Disclosure by Clinical Investigators 21 C. F. R. pts. 54, 312, 314, 320,
330, 601, 807, 812, 814, and 860; effective Feb. 2, 1999

Applicants who submit marketing applications for human drugs, biological
products, or medical devices and submit clinical studies in support of those
applications.

Financial interests and arrangements of the investigator: A financial
interest or arrangement subject to disclosure includes (1) an arrangement
between the sponsor and the investigator (or spouse or dependent child) in
which the value of the investigator?s compensation could be influenced by
the study outcome; (2) significant payments from sponsor to investigator or
institution supporting investigator activities that are valued at more than
$25, 000 beyond the costs incurred in conducting the study; (3) proprietary
interests, including patents, held by the investigator in the product; or
(4) significant equity interests in the sponsor of a covered study whose
value cannot be readily determined through reference to public prices or
valued at more than $50,000 if a company is a publicly traded corporation.

Applicant also must disclose any steps taken to minimize the potential for
bias.

Investigators must provide the sponsor with sufficient, accurate financial
information needed to allow subsequent disclosure or certification.

The applicant must submit, for each investigator who participates in a
covered study, either certification that no financial interest or
arrangement listed in the regulation exists or disclose the nature of the
interest or arrangement to the agency. Certifications and disclosures must
accompany marketing application.

Investigators must update financial disclosure reports during the course of
the study or for 1 year following its completion.

Appendix I: Federal Regulations Pertaining to Financial Interests in
Research

Appendix II: Scope and Methodology Page 33 GAO- 02- 89 Financial Conflicts
in Biomedical Research

To address our objectives, we reviewed the HHS regulations pertaining to
financial interests in biomedical research. In addition, we interviewed
officials at the Food and Drug Administration (FDA), the National Institutes
of Health (NIH), and the Office for Human Research Protections (OHRP). We
also interviewed staff at the American Association of Medical Colleges, the
Association of Academic Health Centers, the Association of American
Universities, the National Association of College and University Business
Officers, the National Bioethics Advisory Commission, 1 and HHS? Office of
Inspector General. We also visited five universities that received federal
funding for biomedical research in order to understand how they were
implementing the HHS financial interest regulations. Our sample selection
and data collection are described in the following sections.

Our sample included public and private academic institutions. Accordingly,
this report does not address how financial conflicts of interest in clinical
research are managed at hospitals or other research institutions.

Our selection criteria were universities

 that received large amounts of research funding from NIH (top 20
universities);

 had extensive technology transfer activities, according to the Association
of University Technology Managers? (AUTM) 1999 licensing survey; 2

 had not been extensively scrutinized, audited, or targeted recently for
review by NIH?s Office of Extramural Research or OHRP; and

 that were located in different geographic areas of the United States. We
visited the following academic institutions: University of California- Los
Angeles; University of North Carolina, Chapel Hill; University of
Washington, Seattle; Washington University, St. Louis; and Yale University,
New Haven.

Given our selection criteria, our sample is biased toward large research
universities with complex organizational structures. Medium and small

1 The National Bioethics Advisory Commission was a presidential commission
established in 1995 to advise the National Science and Technology Council
and other government entities regarding bioethical issues arising from
research on human biology and behavior.

2 AUTM does not provide specific information on biomedical- related
activities. Appendix II: Scope and Methodology

Sample Selection

Appendix II: Scope and Methodology Page 34 GAO- 02- 89 Financial Conflicts
in Biomedical Research

universities may not necessarily have comparable organizational structures.
Consequently, our study results are not generalizable to all universities.

At each of the five universities, we interviewed the following officials:
the institution official responsible for research; the head of the conflict-
ofinterest committee or the institution official responsible for managing
conflict- of- interest issues, or both; the chairperson or a member of the
institutional review board (IRB), or both; the head of the technology
transfer office; and two investigators selected by the university (one
receiving NIH funding for research and another receiving private funding).

We reviewed the universities? policies and procedures on financial conflicts
of interest, sponsored research, outside professional activities, and equity
acquisition. We also reviewed a sample of investigators? financial
disclosures for fiscal years 1999 and 2000. Some universities provided
copies of these financial disclosures and the university?s management plans
with the names of investigators and sponsors removed.

To obtain information on the percentage of university clinical investigators
with financial interests related to their research, we requested information
on the total number of clinical investigators receiving sponsored research
funding and the number of those clinical investigators who disclosed
financial interests each year from 1995 through 2000. We also requested
information on whether the research funding was private or public, the type
of financial interests disclosed (for example, income, equity interests, or
intellectual property rights), and the type of management strategies
employed.

We conducted our work from February through September 2001 in accordance
with generally accepted government auditing standards. Site Visits

Appendix III: Comments From the Department of Health and Human Services

Page 35 GAO- 02- 89 Financial Conflicts in Biomedical Research

Appendix III: Comments From the Department of Health and Human Services

Appendix III: Comments From the Department of Health and Human Services

Page 36 GAO- 02- 89 Financial Conflicts in Biomedical Research

Appendix III: Comments From the Department of Health and Human Services

Page 37 GAO- 02- 89 Financial Conflicts in Biomedical Research

Appendix III: Comments From the Department of Health and Human Services

Page 38 GAO- 02- 89 Financial Conflicts in Biomedical Research

Appendix III: Comments From the Department of Health and Human Services

Page 39 GAO- 02- 89 Financial Conflicts in Biomedical Research

Appendix IV: GAO Contact and Staff Acknowledgments

Page 40 GAO- 02- 89 Financial Conflicts in Biomedical Research

Marcia Crosse, (202) 512- 3407 In addition to the person named above, Anne
Dievler, Bertha Dong, Romy Gelb, Julian Klazkin, and Elizabeth Morrison made
important contributions to this report. Appendix IV: GAO Contact and Staff

Acknowledgments GAO Contact Staff Acknowledgments

Related GAO Products Page 41 GAO- 02- 89 Financial Conflicts in Biomedical
Research

Human Subjects Research: HHS Takes Steps to Strengthen Protections, but
Concerns Remain (GAO- 01- 775T, May 23, 2001).

VA Research: Protections for Human Subjects Need to Be Strengthened

(GAO/ HEHS- 00- 155, Sept. 28, 2000).

NIH Research: Improvements Needed in Monitoring Extramural Grants

(GAO/ HEHS/ AIMD- 00- 139, May 31, 2000).

NIH Clinical Trials: Various Factors Affect Patient Participation

(GAO/ HEHS- 99- 182, Sept. 30, 1999).

Technology Transfer: Reporting Requirements for Federally Sponsored
Inventions Need Revision (GAO/ RCED- 99- 242, Aug. 12, 1999).

Technology Transfer: Number and Characteristics of Inventions Licensed by
Six Federal Agencies (GAO/ RCED- 99- 173, June 18, 1999).

Medical Records Privacy: Access Needed for Health Research, but Oversight of
Privacy Protections Is Limited (GAO/ HEHS- 99- 55, Feb. 24, 1999).

Technology Transfer: Administration of the Bayh- Dole Act by Research
Universities (GAO/ RCED- 98- 126, May 7, 1998).

NIH Extramural Clinical Research: Internal Controls Are Key to Safeguarding
Phase III Trials Against Misconduct (GAO/ HEHS- 96- 117, July 11, 1996).

Scientific Research: Continued Vigilance Critical to Protecting Human
Subjects (GAO/ HEHS- 96- 72, Mar. 8, 1996). Related GAO Products

(290071)

The General Accounting Office, the investigative arm of Congress, exists to
support Congress in meeting its constitutional responsibilities and to help
improve the performance and accountability of the federal government for the
American people. GAO examines the use of public funds; evaluates federal
programs and policies; and provides analyses, recommendations, and other
assistance to help Congress make informed oversight, policy, and funding
decisions. GAO?s commitment to good government is reflected in its core
values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents is through the
Internet. GAO?s Web site (www. gao. gov) contains abstracts and full- text
files of current reports and testimony and an expanding archive of older
products. The Web site features a search engine to help you locate documents
using key words and phrases. You can print these documents in their
entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as ?Today?s Reports,? on its Web
site daily. The list contains links to the full- text document files. To
have GAO E- mail this list to you every afternoon, go to our home page and
complete the easy- to- use electronic order form found under ?To Order GAO
Products.?

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U. S. General Accounting Office P. O. Box 37050 Washington, D. C. 20013

To order by phone: Voice: (202) 512- 6000 TDD: (301) 413- 0006 Fax: (202)
258- 4066

GAO Building Room 1100, 700 4th Street, NW (corner of 4th and G Streets, NW)
Washington, D. C. 20013

Contact: Web site: www. gao. gov/ fraudnet/ fraudnet. htm, E- mail:
fraudnet@ gao. gov, or 1- 800- 424- 5454 (automated answering system).

Jeff Nelligan, Managing Director, NelliganJ@ gao. gov (202) 512- 4800 U. S.
General Accounting Office, 441 G. Street NW, Room 7149, Washington, D. C.
20548 GAO?s Mission

Obtaining Copies of GAO Reports and Testimony

Order by Mail or Phone Visit GAO?s Document Distribution Center

To Report Fraud, Waste, and Abuse in Federal Programs

Public Affairs
*** End of document. ***