University Research: Most Federal Agencies Need to Better Protect
against Financial Conflicts of Interest (14-NOV-03, GAO-04-31).  
                                                                 
In fiscal year 2001, federal agencies provided $19 billion for	 
university research, a vital part of the nation's research and	 
development effort. GAO was asked to examine federal agencies'	 
actions to ensure that (1) the results of the university research
grants they fund are made available to the public and (2)	 
universities receiving such grants implement policies for	 
identifying and managing possible financial conflicts of	 
interest. GAO reviewed the actions of eight federal agencies and 
conducted a Web-based survey of 200 leading research		 
universities. GAO also met with officials in the Office of	 
Science and Technology Policy (OSTP) to discuss the National	 
Science and Technology Council's role in coordinating federal	 
science policy. 						 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-04-31						        
    ACCNO:   A08841						        
  TITLE:     University Research: Most Federal Agencies Need to Better
Protect against Financial Conflicts of Interest 		 
     DATE:   11/14/2003 
  SUBJECT:   Conflict of interests				 
	     Financial management				 
	     Financial management systems			 
	     Internal controls					 
	     Surveys						 
	     Web sites						 
	     Federal grants					 
	     Research grants					 
	     Research and development				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** 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-04-31

United States General Accounting Office

GAO

Report to the Honorable Richard C.

                              Shelby, U.S. Senate

November 2003

UNIVERSITY RESEARCH

  Most Federal Agencies Need to Better Protect against Financial Conflicts of
                                    Interest

                                       a

GAO-04-31

Highlights of GAO-04-31, a report to the Honorable Richard C. Shelby, U.S.
Senate

In fiscal year 2001, federal agencies provided $19 billion for university
research, a vital part of the nation's research and development effort.
GAO was asked to examine federal agencies' actions to ensure that (1) the
results of the university research grants they fund are made available to
the public and (2) universities receiving such grants implement policies
for identifying and managing possible financial conflicts of interest. GAO
reviewed the actions of eight federal agencies and conducted a Web-based
survey of 200 leading research universities. GAO also met with officials
in the Office of Science and Technology Policy (OSTP) to discuss the
National Science and Technology Council's role in coordinating federal
science policy.

GAO recommends that the Department of Education post the final technical
reports of the research it funds on its Web site. GAO also recommends that
the National Science and Technology Council coordinate the development of
a uniform federal requirement for identifying and resolving financial
conflicts of interest in federally funded research. In commenting on the
draft report, Education agreed with the recommendation to post research
results on its Web site. OSTP agreed with the recommendation for
developing a federal requirement.

www.gao.gov/cgi-bin/getrpt?GAO-04-31.

To view the full product, including the scope and methodology, click on
the link above. For more information, contact Robin Nazzaro at (202)
512-3841or [email protected].

November 2003

UNIVERSITY RESEARCH

Most Federal Agencies Need to Better Protect against Financial Conflicts of
Interest

Each of the eight federal agencies GAO examined relies on university
scientists who receive federally funded research grants to make the
results available to the public. Although university scientists
customarily seek to publish their research results in peer-reviewed
journals, agencies cannot require such publication as a condition for
funding because it is impossible to ensure in advance that the results
will be accepted for publication. Agencies do, however, explicitly
encourage funding recipients to make results public. The Departments of
Agriculture, Defense, and Energy; the Environmental Protection Agency
(EPA); and the National Aeronautics and Space Administration (NASA) also
disseminate the results of their funded research by posting them on their
Web sites (see table below). Officials from these agencies said that
posting the results is an effective way to share information among
scientists, as well as with the public. In contrast, the National
Institutes of Health (NIH) and the National Science Foundation (NSF) do
not post research results on their Web sites. According to NIH officials,
the risk associated with posting researchers' final reports before they
have been validated by peer review is too great in the biomedical field.
The Department of Education is considering how best to widely disseminate
the results of research it funds.

NIH and NSF are the only federal agencies that require universities to
implement policies for identifying and managing possible financial
conflicts of interest for the research they fund. The other six agencies
do not have financial conflict of interest standards for university
research grants. Of the 171 universities that responded to the GAO survey,
148 (87 percent) reported that all of their federally funded research is
covered by financial conflict of interest policies that are consistent
with either NIH's or NSF's standards. However, 17 universities reported
that they do not extend either agency's requirements to cover research
grants from other federal agencies. Unless federal agencies uniformly
require that universities implement conflict of interest policies, the
government cannot properly safeguard against financial conflicts of
interest that might bias federally funded research.

Agencies That Post Research Results on Their Web Sites

Contents

  Letter

Results in Brief
Background
Federal Agencies Rely on University Scientists and Agency Web

Sites to Disseminate Research Results Only NIH and NSF Require Financial
Conflict of Interest Standards

for Grant Recipients Conclusions Recommendations for Executive Action
Agency Comments and Our Evaluation Scope and Methodology

1 2 4

6

12 16 16 17 18

Appendixes                                                              
                Appendix I:  Agencies' Actions in Implementing the Shelby  21 
                                              Amendment                    
               Appendix II: Universities' Acceptance of Equity in Start-up 26 
                                              Companies                    
                               Comparison of NIH's and NSF's Financial     
              Appendix III:              Conflict of Interest              
                                              Standards                    28 
              Appendix IV:    Comments from the Department of Education    31 
                              Comments from the National Aeronautics and   
                Appendix V:                     Space                      
                                            Administration                 32 
                               Comments from the National Institutes of       
              Appendix VI:                      Health                     33

Tables    Table 1: Federal Agencies' Policies on Disseminating Research 
                                       Results                              8 
              Table 2: Agencies That Post Research Results on Their Web    
                                        Sites                              10 
              Table 3: NIH's and EPA's Disposition of FOIA Requests Citing 
                                                                       the 
                       Shelby Amendment, as of August 31, 2003             22 
Figures      Figure 1: The Year Universities First Received Equity in a    
                                                          Start-up Company 26

Contents

Abbreviations

EPA Environmental Protection Agency
FDP Federal Demonstration Partnership
FOIA Freedom of Information Act
NARA National Archives and Records Administration
NASA National Aeronautics and Space Administration
NIH National Institutes of Health
NSF National Science Foundation
OMB Office of Management and Budget
OSTP Office of Science and Technology Policy

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

A

United States General Accounting Office Washington, D.C. 20548

November 14, 2003

The Honorable Richard C. Shelby United States Senate

Dear Senator Shelby:

University research is a vital part of the nation's research and
development efforts. Because there is broad consensus that university
research is a longterm, national investment in the future, the federal
government has been the primary source of funding for this research. In
fiscal year 2001, federal agencies provided $19 billion, or about 60
percent of all funding for university research. Eight agencies-the
Departments of Agriculture, Defense, Education, and Energy; the
Environmental Protection Agency (EPA); the National Aeronautics and Space
Administration (NASA); the National Institutes of Health (NIH), within the
Department of Health and Human Services; and the National Science
Foundation (NSF)-obligated 97 percent of the university research funding,
with NIH and NSF alone accounting for $14.2 billion. The Office of Science
and Technology Policy (OSTP) coordinates the development of standard
practices among federal agencies through the National Science and
Technology Council. For example, the Council is undertaking a review of
policies, procedures, and plans relating to the business relationship
between federal agencies and research performers with the goal of
improving the performance and management of federally funded research.

Historically, the primary return on the federal government's investment in
university research was the advancement of scientific knowledge. Science
progresses through open communication among scientists and the sharing of
research results. Publication of research results in peer-reviewed
scientific journals is an academic tradition that both validates and
disseminates these results.

More recently, the federal investment in university research not only has
advanced scientific knowledge but also has yielded thousands of inventions
each year that have fostered the development of new technologies,
stimulated the creation of new jobs, and improved the quality of life. For
some universities, it has also yielded new streams of income that helped
to support their research and education missions. The Bayh-Dole Act of
1980 has facilitated commercialization of university technology by giving
universities, among others, the right to own their federally funded
inventions and license them to businesses. As the importance of university

research to technological innovation has increased, partnerships between
universities and businesses have also grown, giving rise to concerns that
financial conflicts of interest might restrict the dissemination of
research results or bias the conduct or results of federally funded
research.

Industry groups and others have also expressed concerns about the need, in
certain instances, for access to the scientific data that underlie the
published results of federally funded research. In response, in November
1999, the Congress enacted a provision, commonly called the Shelby
Amendment, requiring the Director of the Office of Management and Budget
(OMB) to amend Circular A-110 for universities, hospitals, and other
nonprofit organizations. Under this amendment, federal agencies are
directed to provide scientific data in response to a Freedom of
Information Act (FOIA) request if the published results of federally
funded research are used in developing a federal agency action that has
the force and effect of law.

You requested that we examine federal agencies' actions to ensure that (1)
the results of the university research grants they fund are made available
to the public and (2) universities receiving such grants implement
policies for identifying and managing possible financial conflicts of
interest. In addition, as agreed with your office, appendix I provides
information on agencies' actions in implementing the Shelby Amendment. Our
review focused on the Departments of Agriculture, Defense,1 Education, and
Energy; EPA; NASA; NIH; and NSF. As part of our review, we conducted a
Web-based survey of the 200 universities that received the most federal
funding for research and development in fiscal year 2000. We received
responses from 171 universities, an 86 percent response rate. The
universities' aggregated responses are available at
http://www.gao.gov/special.pubs/gao-04-223sp.

Results in Brief	Each of the eight federal agencies we examined relies on
university scientists who receive federally funded research grants to make
the results available to the public; five of these agencies also
disseminate results by posting them on their Web sites. Although
university scientists customarily

1Within Defense, we generalized the research grant terms and conditions of
the Office of Naval Research and the Air Force Office of Scientific
Research to the entire department. The Army does not have standard
research grant terms and conditions because each of its awarding offices
is responsible for its own unique terms and conditions.

seek to publish their research results in peer-reviewed journals, agencies
cannot require such publication as a condition for funding because it is
impossible to ensure in advance that the results will be deemed by peer
review to be acceptable for publication. Agencies do, however, explicitly
encourage funding recipients to make results available to the public, and
they consider scientists' publication records in reviewing grant
applications. Agriculture, Defense, Energy, EPA, and NASA also disseminate
the results of the research they fund by posting researchers' final
reports on their Web sites because, according to officials at these
agencies, Web sites offer an effective way to share information among
scientists, as well as with the public. In contrast, NIH officials told us
that they do not post researchers' final reports because, in the
biomedical field, the risks associated with posting results that have not
been scrutinized and validated by peer review are too great. Similarly,
NSF officials said that NSF does not post results, partly because some
scientific journals reject manuscripts if the results have already been
posted on the Web. Education currently is considering how best to respond
to the directive in the Education Sciences Reform Act of 2002 to widely
disseminate the findings and results of scientifically valid research in
education. We are recommending that Education post the results of the
research it has funded on its Web site to facilitate access to and
maximize the benefits of its research investment. Education agreed with
our recommendation.

NIH and NSF are the only federal agencies that require universities to
implement policies for identifying and managing possible financial
conflicts of interest for the research they fund; the other agencies do
not have any standards to protect against financial conflicts of interest
in university research. The NIH and NSF standards, promulgated in July
1995, place primary responsibility on universities to institute
appropriate policies and procedures for identifying and managing potential
conflicts of interest. Under these standards, a financial conflict of
interest might be resolved by eliminating or reducing the conflict or it
might be managed by establishing an oversight committee to monitor the
conduct and the reporting of the research. The other six agencies do not
have financial conflict of interest standards for universities, in part
because some of the agencies believe that identifying and managing
conflicts of interests is the responsibility of the universities. Of the
171 university respondents, 148 (87 percent) reported that all of their
federally funded research is covered by financial conflict of interest
policies that are consistent with NIH's and/or NSF's standards. However,
17 university respondents reported that they do not extend either the NIH
nor the NSF financial conflicts of interest requirements to cover research
grants funded by other federal agencies. Unless federal agencies

uniformly require that universities implement financial conflict of
interest policies, the government cannot properly safeguard against
conflicts of interest that might bias federally funded research. We are
recommending that the National Science and Technology Council coordinate
the development of uniform federal requirements for universities and other
funding recipients to identify and resolve financial conflicts of interest
that might bias the design, conduct, or reporting of federally funded
research. OSTP officials agreed with the thrust of our recommendation that
uniform federal requirements be developed to identify and resolve
financial conflicts of interest. However, the OSTP officials noted that
recent experience in developing a common rule for research misconduct has
demonstrated that the process of reaching consensus among federal agencies
can be difficult and prolonged.

Background	The Bayh-Dole Act of 1980 (Pub. L. No. 96-517, Dec. 12, 1980)
has fostered linkages between universities and businesses by giving
universities, other nonprofit organizations, and small businesses the
option to retain title to the inventions they make in the course of
federally funded research. Before 1980, federal agencies generally
retained title rights to any inventions made in the course of the research
they funded. Funding recipients seeking to commercialize such inventions
often faced long delays and uncertainty when they asked the funding
agencies to waive their rights. Since 1980, universities have upgraded and
expanded their technology licensing efforts, particularly in such fields
as biomedicine and computer technology. Federal agencies and industry also
substantially increased their funding of university research-federal
funding grew from $8 billion (in 2001 dollars) in fiscal year 1980 to
$19.2 billion in fiscal year 2001, and industry funding grew from $461
million (in 2001 dollars) to $2.2 billion during this period. For the
Association of University Technology Managers' survey for fiscal year
2001,2 U.S. university respondents reported that they (1) executed 3,282
technology licenses and options, (2) received $852 million in gross
license income, and (3) held equity in 348, or 70 percent, of the 494
start-up

2Beginning in 1993, the Association of University Technology Managers has
annually surveyed U.S. and Canadian institutions on their patenting and
licensing activities. The participants in the fiscal year 2001 survey were
142 U.S. universities, 28 U.S. hospitals and research institutes, 27
Canadian institutions, and 1 patent management firm.

companies that were formed around university-licensed technology.3 (See
app. II for information from our survey about universities' licensing
activities with start-up companies.)

OMB Circular A-110 establishes uniform requirements for the administration
of federal grants and cooperative agreements with institutions of higher
education, hospitals, and other nonprofit organizations. For example, the
circular requires that funding recipients submit performance reports to
the funding agency at least annually, with a final technical report
normally due within 90 days after the grant's termination or expiration.
However, the circular provides flexibility by allowing the agencies to
specify the content of these reports or to waive the final technical
report. The National Science and Technology Council, established by
Executive Order 12881 in November 1993, coordinates the development of
governmentwide science and technology policies. For example, the Council's
Subcommittee on Research Business Models is examining the effects of the
changing nature of scientific research on business models for conducting
federally funded research. In addition, 7 federal agencies, 84 research
universities, and 6 other research institutions participate in the Federal
Demonstration Partnership (FDP), which seeks to streamline the
administrative processes for implementing OMB Circular A-110.4

In July 1995, the Department of Health and Human Services, which includes
NIH, promulgated regulations on Objectivity in Research and NSF revised
its Investigator Financial Disclosure Policy to establish consistent
requirements for universities and most other grantees to identify and
manage financial conflicts of interest.5 Specifically, the NIH and NSF
standards require that funding recipients implement policies for

3According to the Association of University Technology Managers, taking
equity in a start-up company, partially in lieu of cash fees, is an
important licensing approach because start-up companies rarely have a
positive cash flow during their first years of operation and, therefore,
need to conserve cash for investing in product development.

4FDP is an outgrowth of the Florida Demonstration Project formed in 1986.
All of the federal agencies we reviewed, except Education, and 64 of our
171 university respondents participate in FDP and incorporate the FDP
terms and conditions in their grant agreements.

5NIH funds almost all of the university research within Health and Human
Services. The department also has promulgated regulations to protect human
research subjects. (See 45 C.F.R. Part 46.)

(1) scientists to disclose any "significant financial interests"6 to an
official designated by the institution and (2) institutions to determine
whether a real or apparent conflict exists and, if so, take appropriate
actions to manage, mitigate, or eliminate the identified conflict. Under
these regulations, a conflict of interest exists when the institution's
designated official determines that a significant financial interest could
directly and significantly affect the research design, conduct, or
reporting. The financial benefit may result, for example, from an
investigator owning stock in a company providing the research funding, or
from an investigator having ownership interest in a company that may
profit from a university invention. Conflicting interests are not
necessarily unacceptable, and many can be managed through disclosure and
oversight. The NIH regulation exceeds the scope of NSF's policy in some
areas. For example, it requires that universities and other funding
recipients report every identified possible conflict of interest, while
NSF requires that institutions report only those conflicts that have not
been resolved. (See app. III for a more detailed comparison of the NIH and
NSF requirements.)

  Federal Agencies Rely on University Scientists and Agency Web Sites to
  Disseminate Research Results

Federal agencies rely primarily on the university scientists who receive
research grants to make their research results available to the public.
Each agency encourages grantees to publish research results in the
scientific literature, a practice that is steeped in academic tradition.
Agriculture, Defense, Energy, EPA, and NASA also disseminate the results
of the research they fund by posting scientists' final technical reports
on their Web sites, and Education is considering whether to post research
results. While NIH, NSF, and Education do not post research results on
their Web sites, they post certain grant information, including abstracts
submitted at the time of the award.

6Significant financial interests are defined, among other things, to
include (1) the holdings of the investigator, the investigator's spouse,
and any dependent children that exceed a $10,000 equity interest or a
5-percent ownership interest in a single entity and (2) salary, royalties,
or other payments for the investigator and the investigator's spouse and
dependent children that exceed $10,000 over the next 12-month period.

    Agencies Rely on and Encourage Grant Recipients to Make Research Results
    Available

The eight agencies we examined rely on university scientists to
disseminate the results of the research they fund, and their policies
explicitly encourage principal investigators and universities to
disseminate those results through presentations at scientific conferences
and publishing in scientific journals. (See table 1.) Similarly, FDP's
model terms and conditions for research grants state, "The recipient is
expected to publish or otherwise make publicly available the results of
the work conducted under the award." Publishing federally funded research
results also is vital to university scientists because research
publications are key to obtaining future grant awards, gaining
professional recognition, and achieving tenure.

     Table 1: Federal Agencies' Policies on Disseminating Research Results

     Agency                  Policy                         Citation          
                  The principal investigator is                               
                expected to publish or otherwise            Cooperative State 
Agriculture                make                       Research, Education,
               publicly available the results of   and Extension Service      
               the work conducted under this       General Terms              
               award.                              
                                                         and Conditions       
               Publication of results of the                                  
     Defense   research project in appropriate      Office of Naval Research
               professional                        
               journal is encouraged as an             Research Terms and     
               important method of recording and           Conditions         
               reporting                           
                     scientific information.       
               The recipient is expected to        Air Force Office of        
               publish or otherwise make publicly  Scientific Research        
               available                           
                the results of the work conducted  Grant Terms and Conditions 
                      under this agreement.        

Education	Grantees are encouraged to publish the results of the work
conducted under this award.

  Individual grant agreementsa Energy Recipients are encouraged to disseminate
 results promptly to the Office of Science, Grant Application Guide scientific
                       community. Reporting Requirements

EPA  EPA encourages the independent              National Center for       
        publication of the results of its              Environmental          
          extramural research in appropriate   Research Terms and Conditions  
                 scientific journals.          
NASA NASA requires prompt public disclosure        Guidebook for Proposers 
        of the results of its sponsored                       Responding to a 
        research and, therefore, expects         NASA Research Announcement   
        significant findings from supported    
        research to be promptly submitted for  
        peer reviewed publication.             

NIH      [P]rincipal investigators and grantee NIH Grants Policy Statement 
               organizations are expected to make and Final                   
       the results and accomplishments of their    NIH Research Tools Policy  
       activities available to the                         (Dec. 23,          
        research community and to the public at              1999)            
                         large.                   
       Starting with the October 1, 2003, receipt 
                   date, investigators submitting 
           an NIH application seeking $500,000 or      Final NIH Statement on 
               more in direct costs in any single            Sharing Research 
       year are expected to include a plan for       Data (Feb. 26, 2003)     
       data sharing or state why data             
                sharing is not possible.          

NSF   Investigators are expected to promptly prepare and NSF Grants Policy 
                                     submit for publication            Manual 
       ... all significant findings from work conducted     
       under NSF grants.                                    
       Investigators are expected to share with other       
       researchers ... materials                            
       created or gathered in the course of work under NSF  
       grants. Grantees                                     
       are expected to encourage and facilitate such        
       sharing.                                             

Sources: Agriculture, Defense, Education, Energy, EPA, NASA, NIH, and NSF.

aEducation currently is revising its research dissemination policies in
response to the Education Sciences Reform Act of 2002.

Agencies also indirectly encourage the dissemination of research results
through their grant award practices. Officials at each agency said that
peer review panels consider the publication record of the applicant
(usually the principal investigator) in assessing the grant proposal. NSF,
for example, requires that principal investigators requesting grant
renewals include a list of publications generated with NSF's prior
support. Agriculture officials told us that they are less likely to
recommend renewal applications for continued funding if the funded
project's results have not been published. Publications indicate to the
agencies that the principal investigator has

made progress in his/her research and that the results are available to
other scientists in the field. However, a research project may not
generate publishable results because leading scientific journals require
that manuscripts be reviewed by other experts in the field to validate the
research findings prior to publication. The scientific journal may reject
a manuscript because, for example, the reviewers conclude that the work
adds little value to the field of study, the results are inadequately
supported, or the research failed.

All but five of the university respondents reported that they have a
policy or standard operating procedure that addresses whether sponsors are
allowed to delay the publication of research results under certain
circumstances, such as reviewing a manuscript for possible proprietary
information or for intellectual property. Three universities-the
California Institute of Technology, Howard University, and Iowa State
University-reported that they do not permit any publication delays, while
160 universities allow a sponsor to review a manuscript prior to
publication-typically from 30 to 90 days.7 However, 10 universities allow
a longer period of up to either 120 days or 180 days, and 1 university
allows up to 365 days for the sponsor to review a manuscript for
proprietary information.

Generally, research sponsors appear to adhere to the universities' time
frames for reviewing manuscripts. Administrators reported the following:

o 	Fourteen universities were aware of one or more cases during the past 3
years of a sponsor delaying the publication of unclassified and
nonsensitive research beyond the university's time limits.

o 	Three universities were aware of one or more cases during the past 3
years of a federal sponsor delaying the publication of research involving
sensitive, but not classified, information beyond the university's time
limits.

o 	Thirteen universities were aware of one or more cases during the past 3
years of a federal sponsor blocking, or attempting to block, publication
of research involving sensitive, but not classified, information.

7According to NIH, a 30-to 60-day delay to secure intellectual property
rights is generally viewed as reasonable. See "Developing Sponsored
Research Agreements: Considerations for Recipients of NIH Research Grants
and Contracts," (59 fr 55674).

However, several university administrators noted during the pretest of our
survey instrument that publication delays can occur without the
university's knowledge if the sponsor and the research team reach an
accommodation without notifying university administrators.

    Five Agencies Disseminate Research Results on Their Web Sites

As shown in table 2, Agriculture, Defense, Energy, EPA, and NASA use their
Web sites to post research results, in some form, for grants that they
issue.8 For example, EPA posts summaries of annual and final technical
reports on its National Center for Environmental Research Web site. These
summaries include research accomplishments or findings, the reporting
date, EPA agreement number, title, investigators, institution, research
category, project period, objective of research, progress summary,
conclusions (if applicable), publications/presentations, future
activities, supplemental keywords, and other relevant Web sites. EPA's Web
site also allows users to search for publications associated with a
particular grant. NASA primarily posts abstracts of final technical
reports on its Web site, although NASA plans to post mainly full technical
reports in 2004.

Table 2: Agencies That Post Research Results on Their Web Sites

                   Agency     Web-based information system Information posted 
                          Current Research Information       Annual and final 
              Agriculture System                                    technical 
                                                                      reports 
                  Defense Defense Technical Information    Final technical    
                          Center                           reports            
                   Energy               Information Bridge Final technical    
                                                           reports            
                      EPA              National Center For Annual and final   
                                             Environmental report             
                                                  Research          summaries 
                     NASA         Scientific and Technical  Abstract of final 
                                                 Aerospace          technical 
                                              Reports File            reports 

Sources: Agriculture, Defense, Energy, EPA, and NASA.

Note: Some agencies use their Intranet sites to provide greater access to
research results for organizations with which they have a business
relationship. For example, Defense's Scientific and Technical Information
Network uses a customer registration and information release process to
provide access to eligible contractors, academic institutions, and certain
other federal agencies.

8The organic statutes for Energy and NASA call for them to widely
disseminate the results of research that they fund.

While Education, NIH, and NSF do not post research results on their Web
sites, they post a project abstract written at the time of award stating
how the research will be conducted and what researchers hope to
accomplish. In November 2002, the Education Sciences Reform Act of 2002
(Pub. L. No. 107-279) established the Institute of Education Sciences and
directed it to widely disseminate the findings and results of
scientifically valid research in education. An Education official told us
that after the members of the Institute's National Board of Educational
Sciences have been appointed and confirmed, Education will consider how
best to fulfill this requirement, particularly for the results of
Institute-funded research that have not been peer reviewed. The official
noted that the Institute's National Center for Education Evaluation
currently disseminates the results of research performed under contract
either through research publications or through its Web site after the
results have been peer reviewed.

In addition to using their own Web sites, several agencies participate in
collaborative Web-based efforts to share information, including research
results. For example, Energy's Office of Scientific and Technical
Information maintains Federal R&D Project Summaries, a Web-based portal to
summary and award information for Energy, NIH, and NSF research grants.
The office also maintains GrayLIT Network, a portal to full-text reports
located on the Energy, Defense, EPA, and NASA information systems. In
addition, an interagency working group from 11 major science agencies
recently initiated the science.gov Web site, which provides a gateway to
federal research and development results and other scientific information.

Officials at the eight agencies identified both advantages and
disadvantages to posting all funded research results on agencies' Web
sites. Most of the agency officials told us that posting technical reports
on agencies' Web sites is an effective way to share information among
scientists in the field of research, as well as with the public. In
explaining why they have chosen not to post all results on their Web site,
NIH and NSF officials cited concerns that grant results posted prior to
peer review and publication may be incomplete or incorrect and could
mislead other researchers or the public. According to NIH officials, the
risk associated with posting results that have not been scrutinized and
validated by peer review is simply too great in the biomedical field. In
addition, NSF officials were concerned that a scientific journal would
reject a manuscript because it views reports posted on the Web as
publications. Some agency officials also expressed concern that a final
technical report might be posted before the university

files a patent application for an invention, thereby preventing it from
obtaining a patent.

Among the 171 university respondents to our survey, 91 universities (53
percent) supported posting the grantee's final technical reports on the
agency's Web site, and 31 universities (18 percent) opposed posting the
final technical report, while 49 universities (29 percent) either were
uncertain or did not respond. Primary advantages that universities cited
for posting final technical reports on an agency's Web site include
facilitating the access of other scientists to research results,
facilitating collaboration among scientists, providing prompt
dissemination of research results, and providing a public record if the
results of a research project are not published. Primary disadvantages
that universities cited for posting final technical reports are the
potential for (1) an invention to be prematurely disclosed, (2) a
scientific journal to reject a manuscript because it views posted reports
as publications, (3) proprietary information to be disclosed, (4) research
results to be prematurely disclosed, (5) incomplete or misleading report
results to be prematurely disseminated, (6) an investigator to be to
harassed by opponents to the research, and (7) universities to incur added
administrative costs in complying with agency requirements.

  Only NIH and NSF Require Financial Conflict of Interest Standards for Grant
  Recipients

NIH and NSF, the two largest federal supporters of university research,
are the only federal agencies we examined that have adopted standards
intended to protect against financial conflicts of interest among
university grantees. The other six agencies do not require universities
and other grantees to identify and manage possible financial conflicts of
interest involving their research. According to officials from these
agencies, it is the universities' responsibility to protect against
conflicts of interest in university research. While 87 percent of our
survey respondents reported that all of their federally funded research is
covered by financial conflict of interest policies that are consistent
with either NIH's or NSF's standards, 17 universities-including 5
universities in the University of California system-reported that they do
not extend either the NIH or the NSF financial conflicts of interest
requirements to cover research grants funded by other federal agencies.

    Agencies without Financial Conflict of Interest Standards Believe
    Universities Should Take the Lead in Implementing Policies

While both NIH and NSF promulgated regulations in 1995 that require
universities to implement financial conflict of interest policies, the
other six federal agencies do not require that their grantees have similar
standards. According to Agriculture and Energy officials, universities
should take responsibility for developing and implementing policies for
identifying and managing financial conflicts of interest involving their
scientists. Defense and NASA officials told us that they have not
experienced enough problems to justify adopting financial conflict of
interest standards for universities and other grantees. These officials
added that the potential for financial conflicts of interest in the
scientific fields that they fund is generally lower than in the biomedical
field. However, NSF supports research in many of the same fields of
research as these agencies.

    Most Universities Report That Their Financial Conflict of Interest Policies
    Are Consistent with NIH and/or NSF Standards

All of the 171 university respondents to our survey reported that they had
one or more policies for addressing possible financial conflicts of
interest by research investigators. Of the respondents, 148 universities
(87 percent) reported having financial conflict of interest policies
consistent with either NIH's or NSF's regulations that apply to all
federally funded research. More specifically, 135 universities (79
percent) reported that they have a single conflict of interest policy that
applies to all of their research. These universities' policies are
consistent with one of the 10 guidelines that the Association of American
Universities' Task Force on Research Accountability proposed for managing
individual conflicts of interest: "Treat research consistently, regardless
of funding source-all research projects at an institution, whether
federally funded, funded by a nonfederal entity, or funded by the
institution itself, should be managed by the same conflict of interest
process and treated the same." 9

In contrast, 17 universities reported that some of the federally funded
research they perform is not covered by financial conflict of interest
policies that are consistent with either NIH's or NSF's regulations. For
example, 5 universities in the University of California system reported
that their financial conflict of interest policies apply to research
funded by NIH

9Association of American Universities, Task Force on Research
Accountability: Report on Individual and Institutional Financial Conflict
of Interest, October 2001. In asserting that universities are accountable
for the research they perform, the task force provided 10 guidelines for
managing individual conflicts of interest and recommended a three-step
approach for addressing institutional conflicts of interest.

or NSF, but not to research funded by other federal agencies.10 The
Massachusetts Institute of Technology and Yale University reported that
they have specific policies that cover research funded by NIH and NSF,
while their institutional policies cover all other funded research. Six
other universities did not provide a response.

Overall, 124 universities strongly supported, and 25 universities somewhat
supported, creating a single financial conflict of interest policy for all
federally funded research. Among the other respondents, 19 universities
either did not have a strong opinion or did not respond to the question,
while only 3 universities either strongly or somewhat opposed a single
financial conflict of interest policy for all federally funded research.
The university respondents did not agree, however, on which agency's
standards should serve as the basis for a single federal policy: among the
133 universities that expressed an opinion, 72 preferred the NIH
regulation, 56 preferred the NSF regulation, while 5 stated that either
would be acceptable.

To implement their financial conflict of interest policies, 140 of the 171
universities (82 percent) reported that they require scientists to
indicate whether or not a conflict may exist when a grant proposal is
submitted; 108 universities (63 percent) require scientists to annually
submit financial disclosure forms to appropriate institution officials;
and 139 universities (81 percent) require scientists to update financial
disclosure forms during the year if new possible financial conflicts of
interest are identified. A policy that incorporates all three of these
requirements is consistent with the Association of American Universities'
Task Force on Research Accountability guideline: "Disclose financial
information to the institution-individuals engaged in research should
disclose on an annual basis all financial interests related to university
research, and provide updated information when new financial circumstances
may pose a conflict of interest and when grant applications are
submitted." All but 6 of the 171 universities reported that they require
at least one of these three types of financial disclosure.

10Universities in the University of California system also must comply
with state requirements under the Political Reform Act of 1974. The act
established a lower income and equity threshold than the NSF and NIH
regulations for disclosing financial interests, but the relationship
between the university scientist and an affected company is more tightly
limited to only those companies directly supporting the research through
gifts, grants, or contracts.

In addition, 56 universities reported that their policy requires that the
federal funding agency be notified whenever a financial conflict of
interest is identified. In comparison, 62 universities reported that their
policies require that only certain federal funding agencies be notified,
49 universities do not have a policy for notifying federal funding
agencies about identified financial conflicts of interest, and 4
universities did not respond about their notification policies.

Our survey results indicate that several universities have tightened their
policies for financial conflicts of interest in recent years to comply
with the NIH and NSF requirements. Specifically, all of our 171
respondents reported that they have financial conflict of interest
policies, while a survey reported in the November 2000 issue of the New
England Journal of Medicine found that 15 of the 250 institutional
respondents (6 percent) did not have a policy on conflicts of interest.11
In response to the November 2000 survey, NIH reviewed the financial
conflict of interest policies of a representative sample of more than 100
universities and other institutions. NIH found that, generally, the
institutions had developed policies that reflected a serious desire to
inform and assist their investigators in complying with NIH's regulation.
However, NIH found several specific areas of noncompliance and identified
four major areas of concern that the institutions' financial conflicts of
interest policies need to address: (1) many policies are not separated
from other institutional policies through a distinct part, appendix, or
document; (2) investigators face an increased burden because the many
policies do not provide electronic links to supporting information; (3)
many policies are confusing because their applicability and terminology
are unclear; and (4) many policies include numerous examples of vague
language or statements.

Upon review of our university survey results, officials at Agriculture,
Energy, EPA, and NASA told us that OMB should take the lead in developing
a uniform, governmentwide requirement for addressing possible financial
conflicts of interest that is consistent with NIH's and NSF's standards.
NIH and NSF officials also supported developing a uniform requirement that
is consistent with their standards. Defense officials said they were ready
to work with other federal agencies on

11S. Van McCrary et al., "A National Survey of Policies on Disclosure of
Conflicts on Interest in Biomedical Research," The New England Journal of
Medicine 343, No. 22 (2000): 16211626. The authors surveyed 297 medical
schools and other research institutions that received more than $5 million
in total grants annually from NIH or NSF to analyze their policies on
conflicts on interest.

governmentwide regulations, if regulations are warranted. OMB and OSTP
officials believe that the National Science and Technology Council, which
OSTP coordinates, is in the best position to develop a uniform financial
conflict of interest standard for federally funded research.

Conclusions 	A fundamental principle of scientific research is that wide
dissemination of research results is vital for validating these results
and advancing the field of science. Posting final research reports, or
similar information, on federal agencies' Web sites can advance scientific
research by providing other scientists with timely access to research
results and facilitating collaboration. Posting this information also
provides access to members of the public interested in the research and a
public record if the results of agency-funded research are not published,
thus maximizing the benefit of the federal investment. For these reasons,
five federal agencies, including Energy and NASA, already routinely
disseminate research results through their Web sites. While posting
research results might create concerns in some fields, such as biomedical
research, these concerns are less applicable for Education, which like
Energy and NASA, has a specific statutory requirement to widely
disseminate research results.

The growing relationship between universities and businesses since passage
of the Bayh-Dole Act has led to an increase in possible financial
conflicts of interest, as businesses have increased their funding of
university research and some universities have collected more than $10
million in royalties in a given year for technologies they have developed.
In response to the NIH and NSF requirements, all of the universities we
surveyed have implemented policies for identifying and managing possible
conflicts of interest. However, some universities have not extended their
policies to cover research funded by other agencies, which also provide
substantial amounts of research funding, and OMB Circular A-110 does not
address financial conflicts of interest. Unless all federal agencies
require that universities have appropriate conflict of interest policies,
the government cannot ensure that safeguards are in place to protect the
integrity of scientific research, and the public's investment.

Recommendations for 	To better ensure that the findings and results of
scientifically valid research in education are widely disseminated, we
recommend that the Secretary of

Executive Action 	Education direct the new Institute of Education Sciences
to post the final technical reports of the research it funds on its Web
site.

To safeguard against bias in the design, conduct, or reporting of
federally funded research, we recommend that the National Science and
Technology Council coordinate the development of uniform federal
requirements for universities and other funding recipients to identify and
resolve financial conflicts of interest. The NIH and NSF standards provide
a useful starting point for this requirement.

  Agency Comments and Our Evaluation

We provided Education, OSTP, Agriculture, Defense, Energy, EPA, NASA, NIH,
and NSF with a draft of this report for their review and comment.
Education agreed with our recommendation to post the results of the
research it has funded on its Web site, stating that the department is
currently exploring how to best implement the Education Sciences Reform
Act's provisions while not discouraging grantees from having their work
published in scientific journals. (See appendix IV for Education's written
comments.) We met with OSTP officials, including the Associate Director
for Science, who agreed with the thrust of our recommendation that the
National Science and Technology Council coordinate the development of
uniform federal requirements to identify and resolve financial conflicts
of interest. However, the OSTP officials noted that recent experience in
developing a common rule for research misconduct has demonstrated that the
process for reaching consensus among federal agencies can be difficult and
prolonged. We continue to believe that federal agencies should develop a
single, uniform requirement for financial conflicts of interest. Through
their experiences in implementing standards since 1995, NIH and NSF can
provide important insights into the benefits and costs of alternative
approaches in areas where their requirements differ.

The Deputy Administrator for Extramural Programs within Agriculture's
Cooperative State Research, Education, and Extension Service stated, in
oral comments, that the Service agreed with our recommendation and will,
where appropriate, implement financial conflict of interest standards
similar to those of NIH and NSF. Defense, Energy, EPA, NASA, NIH, and NSF
agreed with the factual presentation of the report. (See app. V for NASA's
written comments, and app. VI for NIH's written comments.) Several
agencies also provided specific comments to improve the report's technical
accuracy, which we incorporated as appropriate.

  Scope and Methodology

To assess the actions that federal agencies have taken to ensure the
public's access to authoritative and unbiased scientific research at
universities, we examined the policies and procedures of the eight federal
agencies that primarily fund university research-Agriculture, Defense,
Education, Energy, EPA, NASA, NIH, and NSF. Specifically, we performed the
following audit steps:

o 	To assess agencies' actions to ensure that the results of the
university research they fund are made available to the public, we
reviewed each agency's policies and procedures for disseminating research
results and interviewed agency officials. We also accessed the final
technical reports for several university grant projects from the Web sites
of the five agencies that post research results.

o 	To assess agencies' actions to ensure that universities implement
policies for identifying and managing possible financial conflicts of
interest, we examined whether each agency has regulations or policies
requiring that universities identify and manage possible financial
conflicts of interest. We also interviewed cognizant officials about their
procedures for ensuring that universities are implementing financial
conflict of interest policies. We did not examine the extent to which
agencies have taken additional actions to protect against financial
conflicts of interest for research involving human subjects, a topic
examined in a November 2001 GAO report.12

o 	To assess agencies' actions to implement the Shelby Amendment, we
examined the 1999 legislation; OMB's revisions to Circular A-110; and the
actions each agency has taken to implement the circular's revisions. We
also discussed these actions with cognizant agency officials, asked them
whether they had received any FOIA requests that cited the Shelby
Amendment, and, if so, asked them to provide information about each such
request. We then reviewed the agency's disposition of these FOIA requests.

In addition to our review of federal agencies' actions, we conducted a
Webbased survey of the 200 universities and colleges that received the
most federal research funding in fiscal year 2000. The survey contained 42

12See U.S. General Accounting Office, Biomedical Research: HHS Direction
Needed to Address Financial Conflicts of Interest, GAO-02-89 (Washington,
D.C.: Nov. 26, 2001).

questions that asked about (1) their policies and procedures for ensuring
that federally funded research results are made available to the public,
(2) their views of the advantages and disadvantages of posting a grant's
final technical report to the agency's Web site, (3) their conflict of
interest and financial disclosure policies, (4) any FOIA requests federal
agencies had received that asked for access to research data, and (5) data
on their research funding and technology transfer activities.

We pretested the content and format of the questionnaire with research
office administrators at the Georgia Institute of Technology, Emory
University, Washington University, the University of Missouri, the
University of Colorado, the Colorado School of Mines, George Washington
University, and the University of Maryland. During the pretest, we asked
the administrators to determine whether the survey questions were clear,
the terms used were precise, and the questions were unbiased. We also
assessed the usability of the Web-based format. We made changes to the
content and format of the final questionnaire based on pretest results.

We received responses from 171 of the 200 universities surveyed, for a
response rate of 86 percent. Respondents included 44 of the 50
universities that received the most federal funding in fiscal year 2000.
We performed analyses to identify inconsistencies in the data and resolved
them. The universities' aggregated responses are available at
http://www.gao.gov/special.pubs/gao-04-223sp.

We conducted our review from August 2002 through September 2003 in
accordance with generally accepted government auditing standards.

We are sending copies of this report to the appropriate House and Senate
Committees, the Director of OSTP, the Secretary of Agriculture, the
Secretary of Defense, the Secretary of Education, the Secretary of Energy,
the Administrator of EPA, the Administrator of NASA, the Director of NIH,
the Director of NSF, and the Director of OMB. We also will make copies
available to others upon request. In addition, the report will be
available at no charge on the GAO Web site at http://www.gao.gov.

If you or your staff have any questions about the report, please contact
me at (202) 512-3841. Key contributors to this report were Richard
Cheston, Vondalee Hunt, Ulana Bihun, Donald Pless, and Lynn Musser.

Sincerely yours,

Robin Nazzaro Director, Natural Resources and Environment

Appendix I

Agencies' Actions in Implementing the Shelby Amendment

The Omnibus Consolidated and Emergency Supplemental Appropriations Act,
1999, (Pub. L. No. 105-277) required the Director of the Office of
Management and Budget (OMB) to amend Circular A-110 by incorporating a
provision known as the Shelby Amendment.1 Among other things, the Shelby
Amendment requires that (1) federal awarding agencies ensure that all data
produced under an award will be made available to the public through the
procedures established under the Freedom of Information Act (FOIA) and (2)
if the agency obtaining the data does so solely at the request of a
private party, the agency may charge a reasonable user fee equaling the
incremental cost of obtaining the data. The Shelby Amendment grew out of a
controversy that arose over the Environmental Protection Agency's (EPA)
proposal to tighten Clean Air Act standards for small airborne
particulates in 1997. EPA's proposed rule cited the published results of a
30year epidemiological study funded by the National Institutes of Health
(NIH) and conducted by Harvard University. Various industry groups that
opposed EPA's proposed regulation asked to review original data of the
study. However, Harvard denied the requests, citing both confidentiality
agreements with human subjects and the volume of data accumulated.

On November 6, 1999, OMB published revisions to Circular A-110 in the
Federal Register in response to the Shelby Amendment. Under the revision,
a subject institution must provide the research data to the funding agency
in response to a FOIA request if a federal agency has used the published
research findings in developing an agency action that has the force and
effect of law. In March 2000, 15 federal agencies published an interim
final rule in the Federal Register that codified the OMB Circular A-110
revision. These agencies included Agriculture, Defense, Energy, Education,
EPA, the National Aeronautics and Space Administration (NASA), and NIH.
National Science Foundation (NSF) officials told us that NSF incorporated
the revision by reference to OMB Circular A-110 in its grant agreements.

Only NIH and EPA have received FOIA requests citing the Shelby Amendment.
In reviewing the requests, both agencies determined that the requests did
not meet the OMB Circular A-110 criteria. (See table 3.) Of the 40
requests received by NIH, 20 requested copies of either funded grant
applications or contract records, not research data; 9 requested data
generated from grants funded prior to the effective date of the NIH

1OMB Circular A-110 establishes uniform administrative requirements for
grants and agreements with institutions of higher education, hospitals,
and other nonprofit organizations.

                                   Appendix I
                     Agencies' Actions in Implementing the
                                Shelby Amendment

regulation implementing the Shelby Amendment; and 4 were withdrawn. NIH
officials told us that NIH determined that the remaining seven requests
were not applicable to the Shelby Amendment; however, information on the
basis for this decision was unavailable because NIH had destroyed the FOIA
files 2 years after its final response, in accordance with the National
Archives and Records Administration's (NARA) records retention schedule.
EPA denied both requests it received because the requested data were
generated by projects funded prior to the effective date of its regulation
implementing the revision to OMB Circular A-110.

    Table 3: NIH's and EPA's Disposition of FOIA Requests Citing the Shelby
                        Amendment, as of August 31, 2003

Request Request date Response date Agency disposition

NIH

                     Page 22 GAO-04-31 University Research
                                                                                                                                                                                                                                                                                                                                                                                                                                                                  Request                               
                                                                                        because                                           because                                           because                                           because                                           because                                           because                                                                      because                            did not                               
                                   Not                                                 the FOIA                                          the FOIA                                          the FOIA                                          the FOIA                                          the FOIA                                          the FOIA                                                                     the FOIA                           identify     the date of               
                                   applicable     before                               file had                                          file had                                          file had                                          file had                                          file had                                          file had                                                                     file had                                the    publications               
 Nov. Apr.  Request                because          the               Not                  been                         Not                  been                         Not                  been                         Not                  been                         Not                  been                         Not                  been                        June  Request               Not                  been                             grants      cited, the    
  4,  29,     was                  the           effective            applicable.     destroyed    retention            applicable.     destroyed    retention            applicable.     destroyed    retention            applicable.     destroyed    retention            applicable.     destroyed    retention            applicable.     destroyed    retention 9   May   23,     was    10           applicable.     destroyed    retention 11                for        projects    
 1999 2002 withdrawn               projects        date               Information            in    schedule             Information            in    schedule             Information            in    schedule             Information            in    schedule             Information            in    schedule             Information            in    schedule    16,2000 2000 withdrawn              Information            in    schedule                  which     were funded    
                                   for which      (April              on the         accordance                         on the         accordance                         on the         accordance                         on the         accordance                         on the         accordance                         on the         accordance                                                    on the         accordance                               data    before April    
                                   data are        2000)              basis for            with                         basis for            with                         basis for            with                         basis for            with                         basis for            with                         basis for            with                                                    basis for            with                               were       2000, the    effective
                      Nov.         sought                   Jan. July NIH's              NARA's               Feb. June NIH's              NARA's               Feb. July NIH's              NARA's               Feb. Apr. NIH's              NARA's               Feb. Aug. NIH's              NARA's               Mar. Mar. NIH's              NARA's                                          June July NIH's              NARA's                 Aug. Dec.  sought;                     date of
                       5,          were                     20,   3,  decision is       records                1,   9,  decision is       records                1,  18,  decision is       records                1,  17,  decision is       records               29,   2,  decision is       records               31,  31,  decision is       records                                          28,   6,  decision is       records                  4,  14,  However,                     the NIH
                      1999 Pending funded                   2000 2000 unavailable                             2000 2000 unavailable                             2000 2000 unavailable                             2000 2000 unavailable                             2000 2000 unavailable                             2000 2000 unavailable                                                        2000 2000 unavailable                               2000 2000    given                    regulation

Appendix I
Agencies' Actions in Implementing the
Shelby Amendment

                         (Continued From Previous Page)

                     Page 23 GAO-04-31 University Research
                                                                                                                                                                                                         Not                                                                                                                                                                according                                                                                                                                                                                                                                                                                                                                                                                                                                      Not                                                 
                                                  Not             before              Not                                   Not                                   Not             before                 applicable      before              Not                                   Not                                   Not             before               The study     to                                         Not                                    Not                                      Not                                   Not                                      Not                                       Not                                    Not                                                   before                    Not                                                          applicable                         Not                     
                                                  applicable      April               applicable                            applicable                            applicable      April                  because         April               applicable                            applicable                            applicable      April               referenced     information     predated                applicable                                applicable                               applicable                            applicable                               applicable                             applicable                             applicable                               Not             April                  applicable                                                         because                      applicable    
                                                  because       2000, the   Jan. Feb. because                          Nov. because                     Oct.      because       2000, the   Oct.         the           2000, the   Jan. Feb. because                     Mar.      because                          June because       2000, the               in the       provided by       the      June          because                     June         because                                  because                               because                                  because                           Oct.  because                     Oct. Nov.  because                       Oct. Feb. applicable    2000, the     Dec.        because                       Jan. July  Request                requester                       because      
Request Request Response   Agency    12           the grants    effective   19,  28,  requester     application,       15,  requester     application,  16,       the grants    effective   16,          projects      effective   14,  11,  requester     application,  20,       requester     application,       14,  the grants    effective               request      the            effective   24,          requester     applications,  24,          requester     application,               requester     application,            requester     application,               requester     application,         2,  requester     applications,   3,   7,  requester     applications, 28 28,   6,  because       effective  29 12,        requester     applications, 30  9,  21,     was    31               sought    data 32           requester     
         date     date   disposition              for which      date of    2001 2001 sought          not data         2001 sought          not data    2001      for which      date of    2001         for which      date of    2002 2002 sought          not data    2002      sought          not data         2002 for which      date of               could not     requester,      date of    2002           sought        not data     2002         sought          not data                 sought          not data              sought          not data                 sought          not data          2002   sought        not data     2001 2002   sought        not data       2002 2003 the grant      date of      2002         sought        not data       2003 2003 withdrawn                  only a                        sought      
                                                  data were      the NIH              only a                                only a                                data were      the NIH                 data are       the NIH              only a                                only a                                data were      the NIH                  be         if the          the NIH                    only                                   only a                                   only a                                only a                                   only a                                    only                                   only                                  for which      the NIH                    only                                                            copy of                        only a      
                                        Dec. Dec. sought        regulation            copy of                     Aug.      copy of                          May  sought        regulation               sought        regulation            copy of                          Apr. copy of                     Apr.      sought        regulation  June July identified.    study          regulation               copies of                                 copy of                     July         copy of                     July Aug. copy of                     Aug.         copy of                     Sept.      copies of                              copies of                                data were     regulation         Sept. copies of                                             Feb. Feb.   contract            June Aug. copy of a     
                                        20,  26,  were                                the funded                  27,       the funded                       30,  were                                   were                                the funded                       18,  the funded                   1,       were                      17,  18,      In         existed, it                               funded                           July   the funded                   2,          the funded                  25,  21,  the funded                  29,          the funded                   27,         funded                                 funded                                 sought was                        15,    funded                                               3,  20,    records,            23,  18,    funded      application,
                                        2000 2000 funded                              grant                       2001      grant                            2002 funded                         Pending funded                              grant                            2002 grant                       2002      funded                    2002 2002  addition,                                     Pending   grant                           22,2002 grant                       2002 Pending grant                       2002 2002 grant                       2002 Pending grant                       2002         grant                                  grant                                  funded                           2003    grant                                               2003 2003        not            2003 2003   grant         not data

                                   Appendix I
                     Agencies' Actions in Implementing the
                                Shelby Amendment

                         (Continued From Previous Page)

Request Request date  Response date           Agency disposition           
     33                                 Not applicable because requester      
           June 27, 2003 Sept. 26, 2003 sought only a copy of a funded grant  
                                                application, not data         
           July 28, 2003 Not applicable         Request was withdrawn         
                                          Not applicable because requester    
           July 28, 2003 Aug. 28, 2003   sought only copies of funded grant   
                                               applications, not data         
                                          Not applicable because requester    
           July 28, 2003    Pending      sought only copies of funded grant   
                                               applications, not data         
                                        Not applicable because requester      
           Aug. 1, 2003  Sept. 5, 2003  sought only a copy of a funded grant  
                                                application, not data         
                                        Not applicable because requester      
           Aug. 1, 2003     Pending     sought only a copy of a funded grant  
                                                application, not data         
                                        Not applicable because requester      
           Aug. 10, 2003    Pending     sought only a copy of a funded grant  
                                                application, not data         
                                        Not applicable because the grants for 
           Aug. 19, 2003 Sept. 3, 2003     which data were sought were funded 
                                        before April 2000, the effective date 
                                                of the NIH regulation         

EPA

Dec. 9, 1999 April 24, 2001  Not applicable because the projects for which 
                                                 data were sought were funded 
                                         before the effective date            
                                Not applicable because the projects for which 
Dec. 9, 1999 April 24, 2001                   data were sought were funded 
                                         before the effective date            

Sources: NIH and EPA.

More recently, OMB published a proposed bulletin and guidelines to ensure
that agencies conduct peer reviews of the most important scientific and
technical information relevant to regulatory policies that they
disseminate to the public, and that the peer reviews are reliable,
independent, and transparent.2 The guidance would supplement the
requirements that many agencies have for peer review of "significant
regulatory information," which is scientific or technical information that
qualifies as "influential" under OMB's information quality guidelines and
is relevant to regulatory policies.3 Specifically, the proposed guidelines
state that, to the extent permitted by law, an agency shall have an
appropriate and scientifically- rigorous peer review conducted on all
significant regulatory information

2OMB published a notice in the Federal Register on September 15, 2003,
announcing a proposed OMB bulletin under Executive Order No. 12866 and
supplemental information quality guidelines.

3The proposed guidelines define significant regulatory information to
exclude most routine statistical and financial information; studies that
have already been adequately peerreviewed; and science that is not
directed toward regulatory issues, such as most of the scientific research
conducted by NIH and NSF.

Appendix I
Agencies' Actions in Implementing the
Shelby Amendment

that the agency intends to disseminate. In addition, the proposed
guidelines state that, to the extent permitted by law, an agency shall
have formal, independent, external peer review conducted for so-called
"especially significant regulatory information" which would apply to
significant regulatory information if (1) the agency intends to
disseminate the information in support of a major regulatory action, (2)
the dissemination of the information could otherwise have a clear and
substantial impact on important public policies or important private
sector decisions with a possible impact of more than $100 million in any
year, or (3) the Administrator of the Office of Information and Regulatory
Affairs determines that the information is of significant interagency
interest or is relevant to an administration policy priority.

Appendix II

Universities' Acceptance of Equity in Start-up Companies

Among the 171 respondents to our survey, 155 universities reported that
they, or their affiliates, have the option to accept equity as a means of
payment for licensed technology. As shown in figure 1, since the enactment
of the Bayh-Dole Act in December 1980, these universities have
increasingly begun receiving equity in start-up companies in lieu of
receiving license fees and royalties. Prior to the act, only 10
universities accepted equity in start-up companies.

Figure 1: The Year Universities First Received Equity in a Start-up
Company

As of March 2003, 123 universities reported that they held equity in at
least one start-up company, and 44 of these universities reported that
they held

Appendix II
Universities' Acceptance of Equity in Start
up Companies

equity in at least 10 start-up companies. The Massachusetts Institute of
Technology held equity in 116 start-up companies at that time.
Furthermore, 93 universities reported that they held, on average, less
than 10 percent of the start-up companies' equity, and 31 universities
reported that they held, on average, 10 percent or more of the start-up
companies' equity. While 16 universities limit equity ownership to at most
10 percent, 116 universities reported that their institutional policy does
not restrict the percentage of equity ownership they can hold in a
start-up company.

Appendix III

Comparison of NIH's and NSF's Financial Conflict of Interest Standards

On July 11, 1995, the Department of Health and Human Services, which
includes NIH, promulgated regulations and NSF revised its Investigator
Financial Disclosure Policy to establish consistent requirements for
universities and other grantees, with certain exceptions, to identify and
manage real or apparent financial conflicts of interest. The stated
purpose of these requirements is to ensure a reasonable expectation that
the design, conduct, and reporting of research will be unbiased by any
conflicting financial interest of the investigator. The effective date of
these standards was October 1, 1995.

Both NIH and NSF define a "significant financial interest" as anything of
monetary value with the following exceptions:

o  salaries, royalties, and remuneration from the applicant institution;

o 	any ownership interest in the institution, if the institution is an
applicant under the Small Business Innovation Research program;

o 	income from seminars, lectures, teaching engagements, and service on
advisory committees or review panels;

o 	an equity interest that-when aggregated for the investigator, spouse,
and dependent children-does not exceed $10,000 and does not represent more
than 5 percent ownership interest in a single entity; or

o 	salary, royalties, or other payments that-when aggregated for the
investigator, spouse and dependent children-do not exceed $10,000 over the
next 12 months.

The NIH regulations (42 C.F.R. Part 50 and 45 C.F.R. Part 94) require that
each institution, except Phase I applicants for the Small Business
Innovation Research program, takes the following actions:

o 	Maintains a written, enforced policy on conflict of interest complying
with the regulations, and inform investigators of the policy. The
institution must take reasonable steps to ensure that subgrantees comply
with its policy.

o 	Designates an institutional official who will review financial
disclosure statements.

Appendix III
Comparison of NIH's and NSF's Financial
Conflict of Interest Standards

o 	Requires that each investigator submit to the institutional official,
by the time the application is submitted for funding, a listing of
significant financial interests that would reasonably be affected by the
research.1

o 	Provides guidelines for designated officials to identify conflicts of
interest and take necessary action to manage, reduce, or eliminate those
conflicts. Under the regulations, a conflict of interest exists when the
designated official reasonably determines that a significant financial
interest could directly and significantly affect the design, conduct, or
reporting of the funded research.

o 	Maintains records for 3 years after the date of the submission of the
final report of expenditures.

o 	Establishes adequate enforcement mechanisms and provide for appropriate
sanctions.

o 	Certifies in each application for funding that the institution has an
administrative process to manage conflicts of interest and that, prior to
any expenditure of funds, the institution will report the existence of a
conflict and assure that it is being managed, reduced, or eliminated.

If an investigator fails to comply with the institution's policies and
has, thereby, biased the research, the institution must report the
noncompliance immediately to NIH and inform NIH of the action that has
been, or will be, taken. If this failure occurs in a project whose purpose
is to evaluate the safety or effectiveness of a drug, medical device, or
treatment, the institution must require that it be disclosed in each
public presentation of the results of the research.

NSF's policies were developed in close conjunction with the NIH
regulations but differ in the following significant respects:

o  NSF has no conflict of interest requirement governing subgrantees.

o  NSF exempts all entities with less than 50 employees from its standard.

1The investigator is the principal investigator and any other person
responsible for the design, conduct, or reporting of research funded by
NIH. For financial interest purposes, the term also includes the
investigator's spouse and dependent children.

Appendix III
Comparison of NIH's and NSF's Financial
Conflict of Interest Standards

o 	NSF requires that records be retained for 3 years after the termination
of the award instead of 3 years after the last financial statement has
been submitted.

o 	NSF requires that the institution provide notification of a conflict of
interest only if the institution is unable to resolve the conflict.

o 	NSF permits research to proceed, in spite of disclosed conflicts, if
the review determines that restrictions would be ineffective or that the
benefits of proceeding outweigh the consequences of any negative impact.
NIH does not address this issue in its policy.

                                  Appendix IV

                   Comments from the Department of Education

Appendix V

Comments from the National Aeronautics and Space Administration

Appendix VI

Comments from the National Institutes of Health

GAO's Mission	The General Accounting Office, the audit, evaluation and
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.

  Obtaining Copies of GAO Reports and Testimony

The fastest and easiest way to obtain copies of GAO documents at no cost
is through the Internet. GAO's Web site (www.gao.gov) contains abstracts
and fulltext 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 www.gao.gov and
select "Subscribe to e-mail alerts" under the "Order GAO Products"
heading.

Order by Mail or Phone	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 441 G Street NW, Room LM Washington, D.C.
20548

To order by Phone: 	Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

To Report Fraud, 	Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm

  Waste, and Abuse in E-mail: [email protected]

Federal Programs Automated answering system: (800) 424-5454 or (202)
512-7470

Public Affairs	Jeff Nelligan, Managing Director, [email protected] (202)
512-4800 U.S. General Accounting Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

                               Presorted Standard
                              Postage & Fees Paid
                                      GAO
                                Permit No. GI00

United States
General Accounting Office
Washington, D.C. 20548-0001

Official Business
Penalty for Private Use $300

Address Service Requested
*** End of document. ***