Ghostwriting in Medical Literature
Minority Staff Report
111th Congress
United States Senate Committee on Finance
Sen. Charles E. Grassley, Ranking Member
June 24, 2010
Table of Contents
BACKGROUND ............................................................................................................................ 1
FINDINGS ..................................................................................................................................... 4
A. Despite acknowledgment of medical writers for “editorial assistance,” the role of
pharmaceutical companies in medical publications remains veiled or undisclosed ............ 4
B. Some medical schools explicitly prohibit ghostwriting in their policies ............................. 7
C. Detection of ghostwriting by medical schools is limited ..................................................... 9
D. Strengthening journal authorship policies appears to have limited effect on ghostwriting
and disclosure of industry financing of medical articles ................................................... 11
E. National Institutes of Health does not have explicit policies on disclosure of industry
financing of ghostwritten articles....................................................................................... 14
CONCLUSIONS.......................................................................................................................... 14
Attachments 1-6
1
BACKGROUND
In 2005, the Senate Committee on Finance (Committee) initiated an inquiry into
educational grants for continuing medical education (CME) programs. This inquiry began after
reports that drug companies were using the grants to promote off-label uses of their drugs, i.e.,
uses that had not been approved by the Food and Drug Administration. The findings of that
inquiry were released in a Committee staff report in April 2007.1 The Committee’s inquiry
revealed that the pharmaceutical industry spent more than a billion dollars a year to fund CME
programs.
Since releasing that report, Ranking Member Charles Grassley expanded his inquiries
into the financial relationships between drug and device companies and academic physicians and
scientists. These financial relationships include payments to physicians and scientists for
consulting services, speaking engagements, and research activities. Senator Grassley also
examined the policies and reporting practices at over a dozen medical schools in the United
States and found that the medical schools have not adequately monitored the outside income of
their researchers and faculty. The National Institutes of Health (NIH) relies on an institution to
report and manage its faculty members’ conflicts of interest in order to maintain the integrity of
federal funding for biomedical research. However, Senator Grassley found cases where there
were vast disparities between the amounts of payments leading physicians and scientists received
from drug companies and the amounts they reported to their institutions.
Senator Grassley’s inquiries led to the introduction of the bipartisan Physician Payments
Sunshine Act, which he co-authored with Senator Herb Kohl. This bill was incorporated into the
recently passed health care reform legislation, the Patient Protection and Affordable Care Act,2
which was signed into law by President Obama on March 23, 2010. Beginning in March 2013,
drug, device, and biologics manufacturers will be required to report annually payments they
made to physicians nationwide.
About two years ago, Senator Grassley inquired about an industry practice to get articles
published in major medical journals touting the benefits of a company’s product without public
disclosure that the company initiated and paid for the development of the articles. Specifically,
Senator Grassley wrote to Merck & Co., Inc. (Merck) and Scientific Therapeutics Information
(STI), a medical publishing company, following the publication of a study in the Journal of the
American Medical Association (JAMA).3 In that study the authors examined Merck’s
manipulation of scientific literature through ghost writers to market the painkiller Vioxx.
Notably, based on their review of court documents the authors of the JAMA article concluded
that “review manuscripts were often prepared by unacknowledged authors and subsequently
1 Committee Staff Report to the Chairman and Ranking Member, Use of Educational Grants by Pharmaceutical
Manufacturers, S. Prt. 110-21, April 2007, available at
http://finance.senate.gov/newsroom/chairman/release/?id=af4af834-3fab-4293-be6d-ca7f1246484f.
2 Pub. L. No. 111-148, § 6002 (2010).
3 Merck removed Vioxx from the market in 2004 because of cardiovascular risks.
2
attributed authorship to academically affiliated investigators who often did not disclose industry
financial support.”4
Medical ghostwriting is a practice where pharmaceutical or device companies hire
medical education, marketing or communications companies to draft articles that are presented to
prominent physicians and scientists to sign on as authors to increase the likelihood that the article
will be published in important medical journals. Ghostwritten articles include articles that are
drafted by pharmaceutical or device company employees who are not acknowledged in the final
publication. The articles may be review articles, editorials or primary research papers, and they
are typically presented to physicians and scientists affiliated with academic institutions. The
physicians and scientists agree to sign on even if they may not be intimately familiar with the
underlying data or relevant research or provided limited input on the article. Authors who make
little to no contribution to a publication are also referred to as “guest” authors.
Senator Grassley is concerned about the lack of transparency that exists in medical
ghostwriting. Not only are the articles typically initiated and paid for by a pharmaceutical or
device company, but also more significantly, the final publications do not disclose the
company’s role and financial support for the article. Ghostwritten articles can have a significant
impact on, among other things, physician prescribing practices. When prominent physicians and
scientists lend their names to an article, it raises the credibility of the findings and conclusions
presented. This, in turn, can affect the pocketbook of the American taxpayer since Medicare and
Medicaid pay billions of dollars for prescription drugs. In addition, manipulation of medical
literature could lead physicians to prescribe drugs that are more costly or may even harm
patients.
A prominent researcher and professor of medicine at a leading medical school informed
Committee staff that in the late 1990s and early 2000s it was common practice for
pharmaceutical companies to approach him and his colleagues with requests to review and sign
on as primary authors to company studies. Specifically, a company would tell a physician or
scientist at an academic institution that it completed a study and would like his or her
interpretation and feedback on the results of that study. In return, the company would offer that
physician or scientist lead authorship on the paper, which was written by someone other than the
physician or scientist. The physician or scientist would not be paid for being the author of the
paper, but he or she may receive compensation for the time and effort related to reviewing and
commenting on the study results. The researcher who contacted Senator Grassley stated that
while he was aware of colleagues accepting the companies’ offers, his own policy is “I won’t
touch it if I wasn’t involved in the concept of the study.”
In December 2008, Senator Grassley also wrote to Wyeth5 about allegations that the
company was engaging in the practice of ghostwriting. The Committee was provided with
documents from recent lawsuits involving Wyeth’s hormone therapy products. The documents
showed that Wyeth hired a medical communications and education company, DesignWrite Inc.
(DesignWrite), to draft review articles regarding the breast cancer risk of hormone therapy
4 Joseph S. Ross, Kevin P. Hill, et al., “Guest Authorship and Ghostwriting in Publications Related to Rofecoxib,”
JAMA, 2008; 299 (15): 1800-1812.
5 Pfizer, Inc. acquired Wyeth on October 15, 2009.
3
products and then invited academic researchers to sign on as the primary authors. The
documents also revealed information about the process. In one email dated March 24, 1999,
DesignWrite’s Vice President for Scientific Affairs described the company’s role to Wyeth:
The first step is to choose the target journal best suited to the manuscript’s
content, thus avoiding the possibility of manuscript rejection. We will then
analyze the data and write the manuscript, recruit a suitable well-recognized
expert to lend his/her name as author of the document, and secure his/her approval
of its content. After the client has reviewed and released the manuscript for
submission, DesignWrite will see it through the necessary production stagescreating
camera-ready figures and tables and the text according to the journal
guidelines-and submit the package…to the appropriate journal editor. Any
revisions requested by the journal will be handled by DesignWrite in conjunction
with the client and the author. Should the journal reject the manuscript,
DesignWrite will restyle it for submission to another journal within 10 working
days. [Emphasis added] See Attachment 1.
Because medical schools and medical journals have a role in promoting greater
transparency and accountability in the development and authorship of medical literature, Senator
Grassley wrote to leading medical schools and journals and asked about their positions and
policies on ghostwriting. Letters of inquiry were sent to the following medical schools:
 Columbia University College of Physicians and Surgeons (Columbia)
 Duke University School of Medicine (Duke)
 Harvard University School of Medicine (Harvard)
 Johns Hopkins University School of Medicine (Johns Hopkins Medicine)
 Stanford University School of Medicine (Stanford Medicine)
 University of California, San Francisco School of Medicine (UCSF)
 University of Pennsylvania School of Medicine (Penn Medicine)
 University of Washington School of Medicine (UW Medicine)
 Washington University in St. Louis School of Medicine (Washington University)
 Yale University School of Medicine (Yale)
Letters of inquiry were sent to the following medical journals:
 American Journal of Medicine
 Annals of Internal Medicine
 Annual Review of Medicine
 Archives of Internal Medicine
 Nature Medicine
 PLoS Medicine
 The Journal of the American Medical Association (JAMA)
 The New England Journal of Medicine (NEJM)
4
The following report presents the Committee staff’s findings to date, based on (1) a
review of documents provided to the Committee by the medical schools, the medical journals,
DesignWrite, Merck, STI, and Wyeth; (2) a review of court documents and publications the staff
collected online and through literature search; and (3) interviews with published academic
researchers, physicians, and attorneys examining conflicts of interest and/or the practice of
ghostwriting.
FINDINGS
A. Despite acknowledgment of medical writers for “editorial assistance,” the role of
pharmaceutical companies in medical publications remains veiled or undisclosed
Senator Grassley wrote to Wyeth in December 2008 after receiving documents from
recent lawsuits involving the company’s hormone therapy products. In his letter, the Senator
highlighted three publications as examples of articles that Wyeth had paid the medical
communications firm, DesignWrite, to outline and draft for target authors.6 According to Wyeth,
44 review articles relating to the Premarin products were authored by outside academic experts
with “publication support” from DesignWrite. DesignWrite’s documents show that the company
“initiated a comprehensive, peer-reviewed publication program in support of the PREMARIN
Family of Products” in 1997. One goal of the communications plan was to “restore confidence”
in the Premarin family of products.
In Wyeth’s response to Senator Grassley, the company described the extent of the
authors’ involvement in the development and review of the three publications cited in the letter.
The timing of the authors’ input varied during the development of each article. In one case,
input and involvement appeared to have been initiated with the first draft of the manuscript. In
the other two cases, Wyeth stated that the authors were involved at an earlier stage in the
manuscript’s development, providing input on the outline of the drafts as well as the draft
manuscripts themselves.
Nevertheless, as Senator Grassley pointed out in his letter to Wyeth, the final journal
publications only acknowledged the medical writers for their “editorial assistance” or
“assistance.” The articles did not disclose that Wyeth had initiated and paid DesignWrite for the
development of the manuscripts and that the medical writers were hired and compensated by
DesignWrite. Wyeth stated that DesignWrite was compensated for its work in getting
manuscripts drafted and submitted for publication but payments were not allocated for individual
articles.
6 John Eden, “Progestins and Breast Cancer,” Am J Obstet Gynecol, 2003; 188(5): 1123-1131; Lila E. Nachtigall,
“Sex Hormone-Binding Globulin and Breast Cancer Risk,” Prim Care Update OB/GYNs, 1999; 6(3): 39-45; Trudy
L. Bush, Maura Whiteman, and Jodi A. Flaws, “Hormone Replacement Therapy and Breast Cancer: A Qualitative
Review,” Obstet Gynecol, 2001; 98(3): 498-508.
5
Wyeth’s role did not end with underwriting the cost of manuscript development. It
appears from documents in the Committee’s possession that Wyeth also reviewed and provided
input on the outlines and drafts of manuscripts prepared by the medical writers. However, that
involvement also was not disclosed in the final publication. An article published in PLoS
Medicine in February entitled, “Ghostwriting at Elite Academic Medical Centers in the United
States,”7 illustrates the points at which a company may be involved in a journal article’s
development. A company may play a role in reviewing and approving an article at different
stages of the article’s development. The company may influence the outline, the draft before it is
sent to a target author for review, the draft after the author provides comments or edits, or the
medical writer’s responses to reviewer comments on the article during the peer review process.
Another point where the company may play a role is during final approval of the manuscript
before the author submits it to the journal. According to an email from DesignWrite to an author
dated November 17, 1999, it may be the company that gives an article the green light for
submission. See Attachment 2.
Documents Merck submitted to the Committee also demonstrate the company’s
involvement in developing articles for journal publication. For example, STI’s June 1999
proposal to Merck on secondary publications shows that STI proposed an outline, author and
time frame for Merck’s approval to proceed on a specific article. The rest of the timeline shows
that the first draft was submitted to the company for review and comment before it was sent to
the author for input. Subsequent drafts were also reviewed and approved by Merck before the
articles were submitted for publication. See Attachment 3.
The editor-in-chief of the medical journal PLoS Medicine expressed her concern
regarding medical ghostwriting in a statement that was filed in support of the Public Library of
Science’s motion to intervene and motion for access to discovery materials in the Prempro
Products Liability Litigation. In that statement she said:
Medical ghostwriting is a particularly troubling form of manipulation. When
they are appropriately acknowledged for their involvement, medical writers
paid by drug companies…may have a legitimate role in helping shape papers
for publication. However, when the medical writers’ involvement is hidden
they become ghostwriters, and hence they are unaccountable for their work.
The fact that ghostwriters are paid for by drug companies, and that their role is
by definition hidden, suggests that it is likely that they will write about a
company’s products in a biased way….When ghostwriters are used, readers
are unaware that the company was ever involved in shaping the article’s
contents. Instead, the published article bears only the names of the academic
physicians or scientists, who are often highly renowned and trusted in their
fields. By keeping the company’s role in the article hidden, the article has
7 Jeffrey R. Lacasse and Jonathan Leo, “Ghostwriting at Elite Academic Medical Centers in the United States,”
PLoS Med, 2010; 7(2): 1-4.
6
greater credibility in the eyes of the medical community, and thus greater
opportunity for influencing the prescribing behavior of physicians.8
“Editorial assistance” may suggest to the reader and journal editors that there was
involvement of a third party medical writer, hired by a pharmaceutical company through a
medical educational, communications or marketing company, to develop publications in support
of the pharmaceutical company’s products. However, published articles have not consistently
acknowledged editorial assistance. For example, one of the Vioxx-related articles that Merck
paid STI to help develop does not mention STI or the medical writer that assisted Merck and the
author with the article.9
Furthermore, the term “editorial assistance” is poorly defined and does not shed light on
the specific roles played by a pharmaceutical company or other commercial entities. Unless an
article presents a company-funded study, there may be no mention of financial support from the
pharmaceutical company. Companies compensate the medical educational, communications or
marketing company for their work on the articles, but they do not always pay the authors of the
articles. For example, Merck stated that there were four articles related to Vioxx that the
company engaged STI to assist with production. Two of the authors declined compensation
while the other two authors received an honorarium of $1500.
Wyeth informed the Committee that it revised its publication policies in 2006 to include
greater disclosure of the role that Wyeth and medical writers play in developing a Wyeth-funded
publication. Specifically, Wyeth’s policy stated:
The Acknowledgements section of a publication should…also be used to
acknowledge the project’s funding and Wyeth’s involvement in the analyses of
the data or preparation of the publication.
* * *
When professional writing staff assist an author in interpreting data and/or
producing a publication, the author should recognize their contributions
appropriately in the resulting publication.
* * *
It is Wyeth’s expectations that authors and speakers will fully disclose all
financial and material support for research related to the publication, disclose all
potential conflicts of interest related to the publication, and affirm that they have
read and approved the final publication.10
8 Declaration of Virginia Barbour in support of the Public Library of Science’s Motion to Intervene and Motion for
Access to Discovery Materials, In re Prempro Products Liability Litigation (D. Ark. 2009), available at
http://www.plosmedicine.org/static/ghostwritingDeclaration.pdf.
9 William R. Garnett, “Clinical Implications of Drug Interactions with Coxibs,” Pharmacotherapy, 2001; 21(10):
1223-1232.
10 “Policy on External Publications and Presentation,” Wyeth Pharmaceutical Policies, Policy 513, May 4, 2006.
7
The question, however, is how was this policy enforced and who ensured that the authors made
the appropriate disclosures to the medical journals.
Since Pfizer acquired Wyeth last fall, Pfizer’s policies are now in effect. On Pfizer’s
website is its policy on public disclosures of Pfizer-sponsored clinical studies, including a
requirement that authors of study publications “acknowledge individuals who provide editorial
support and disclose the funding source.”11 Pfizer’s internal document on public disclosure and
authorship also includes policies governing other manuscripts submitted to peer-reviewed
medical and scientific journals, such as review articles, secondary articles, supplements, abstracts
and book chapters. According to that document, if Pfizer pays an author or a healthcare
institution for the development of a publication, Pfizer makes the payment directly to the author
or institution. Furthermore, if Pfizer pays a third party for editorial/writing support and the
medical journal does not specify how Pfizer support is to be acknowledged in the publication,
Pfizer’s policy provides acknowledgment language that should be submitted to the journal. For
example, the acknowledgment/disclosure statement for editorial support would be:
“Editorial/medical writing support was provided by at and was
funded by Pfizer Inc.”
In addition, according to Pfizer’s internal document, to ensure that authors are informed
of Pfizer’s policies on authorship and disclosure of Pfizer’s support, the company sends each
potential author a letter on its policies and asks the authors to acknowledge that they will adhere
to the policies. If an author refuses to sign an acknowledgment, then the matter is referred to the
legal department.
B. Some medical schools explicitly prohibit ghostwriting in their policies
Last November, Senator Grassley wrote to 10 leading medical schools asking whether or
not the schools have written policies regarding ghostwritten articles as part of his continuing
effort to shed light on the ties between the pharmaceutical industry and medical professionals.
Based on a review of the responses, which were submitted in December 2009, the Committee
staff found that:
 Six of the medical schools have policies that explicitly prohibit ghostwriting—Columbia,
Johns Hopkins Medicine, Stanford Medicine, UCSF, UW Medicine, and Washington
University.
 In addition to prohibiting ghostwriting, UW Medicine’s policy prohibits guest (honorary,
courtesy, prestige) and gift authorship. Yale’s policy also prohibits gift authorship. Guest
authorship is often granted to increase the credibility of the work or the likelihood of
publication, even though the individuals do not meet the criteria required for authorship.
Gift authorship is offered to individuals who do not contribute intellectually or
11 Pfizer’s policy on public disclosure and authorship of Pfizer-sponsored clinical trials is available at
http://www.pfizer.com/research/research_clinical_trials/registration_disclosure_authorship.jsp.
8
substantially to the work as repayment or in exchange for another benefit or anticipated
benefit.
 Three of the medical schools’ policies—Harvard, Duke, and Yale—prohibit guest,
honorary or courtesy authorship but not ghostwriting explicitly. Harvard, however, stated
that it prohibits the practice of ghostwriting and is examining ways to strengthen its
current prohibitions related to ghostwriting. Yale does not use the term guest or honorary
but it prohibits faculty from adding as co-authors “highly respected individuals merely as
an attempt to increase the likelihood of publication.” In addition, Yale stated that the
university believes its current standards for authorship, which requires that the individual
contribute in a meaningful way, effectively prohibit ghostwriting because “the purported
authors have not made meaningful contributions to the content nor have they given
attribution to the true authors.” The Dean of the School of Medicine also sent out an
email in December 2009 to Yale medical faculty and students regarding ghostwriting. He
wrote that the “practice is strictly contrary to the values and principles of academic
medicine and is prohibited by existing Yale University policy.”
 Penn Medicine does not use the term “ghostwriting” in its authorship policies, but stated
that it has policies against plagiarism and it considers ghostwriting to be the equivalent of
plagiarism.
In addition, based on the responses submitted, six of the medical schools—Duke,
Harvard, Johns Hopkins Medicine, UW Medicine, Washington, and Yale—appear to have
adopted the same criteria for authorship that have been recommended by the International
Committee of Medical Journal Editors (ICMJE).12 In general, under the schools’ policies, to
qualify as an author, the individual must meet at least all three of the following:
1) contribute substantially/significantly to the conception, design, execution and/or data
acquisition or interpretation of the underlying paper;
2) participate in the drafting, reviewing and/or revising of the manuscript for intellectual
content; and
3) provide final approval of the manuscript to be published.
Authorship policies alone, however, do not address the lack of transparency in the role of
a pharmaceutical or device company in the development of scientific articles. In addition to
explicit policies regarding ghostwriting and/or guest, honorary or gift authorship, seven medical
schools13 have policies requiring acknowledgment of individuals and/or entities that contribute to
the publication but do not meet the criteria for authorship. These policies generally state that
contributors should be disclosed or acknowledged in the final publication. Columbia’s policy
specifically states that “any articles or other materials written in conjunction with commercial
entities must include full disclosure of the role of each author, as well as other contributions or
participation by such commercial entities.”
12 International Committee of Medical Journal Editors, Uniform Requirements for Manuscripts Submitted to
Biomedical Journals: Ethical Considerations in the Conduct and Reporting of Research: Authorship and
Contributorship, available at http://www.icmje.org/ethical_1author.html.
13 Columbia, Duke, Harvard, Johns Hopkins Medicine, UW Medicine, Washington, and Yale.
9
While the responses from the 10 medical schools to Senator Grassley show that most
expressly prohibit ghostwriting in their policies, a recent survey published in PLoS Medicine
shows that a smaller percentage of the top 50 academic medical centers have such explicit
policies. In February 2010, Drs. Jeffrey Lacasse and Jonathan Leo published a survey of
academic medical center policies to determine the extent to which these policies explicitly
prohibit ghostwriting.14 The authors found that ten of the academic medical centers explicitly
prohibited ghostwriting. They also found that three centers have authorship policies that prohibit
ghostwriting in practice but do not mention the term “ghostwriting” and thirteen authorship
policies fail to ban all aspects of ghostwriting, mainly because the policies do not require that all
qualified authors be listed.
C. Detection of ghostwriting by medical schools is limited
In their responses, all of the medical schools reported that they do not allow or condone
ghostwriting. For example, some referred to the practice as “unacceptable,” a “violation of
university policy,” “inconsistent with principles of sound research and scholarship,” and
“contrary to the values and principles of academic medicine.” As discussed earlier in this report,
some of the medical schools have policies that specifically prohibit ghostwriting and require
acknowledgment of third parties that contribute to or assist with faculty publications. While it is
important to have clear policies on authorship and acknowledgment of the contributions of nonauthors,
including the pharmaceutical or device company that financed development of the
article, these policies require enforcement to be effective. However, like the disclosure of
faculty financial conflicts of interest, the universities rely on a faculty honor system. Senator
Grassley found from his inquiries that the system has it flaws when it comes to disclosure of
financial interests.
UCSF also pointed out that ghostwriting is not as easily detected as plagiarism, where the
original or aggrieved author would report the incident. Specifically, the university noted an
incident in 2008 when UCSF received an allegation from an author that his work had been
plagiarized by a UCSF faculty member. As a result of UCSF’s investigation into the matter, the
university found that the article was ghostwritten and plagiarized by a medical writer. According
to UCSF, an industry representative had presented the paper to the faculty member for review
and signature. The UCSF faculty member reviewed the article, made minimal modifications and
signed it without realizing that the ghost writer had plagiarized another author’s work. The
faculty member wrote to the publisher and retracted his name from the article. The investigating
committee at UCSF recommended that the university develop resources to educate everyone
about the serious matter. UCSF stated that because the faculty member did not know that the
article had been plagiarized, no sanctions were imposed for plagiarism.
Two other universities also reported allegations related to ghostwriting. Washington
University described an allegation in December 2005 that a faculty member’s paper had been
substantially influenced by a medical writer paid by industry. The medical writer was not
acknowledged in the published article. The university investigated and concluded that the
14 Lacasse and Leo, supra note 7.
10
faculty member “liberally edited the draft and that the opinions were that of the faculty member.”
However, the university also informed the faculty member that failure to acknowledge the
contribution of the medical writer violated the spirit of university policy. The university has
since updated its policy to explicitly prohibit ghostwriting.
UW Medicine also received an allegation in 2005 that a faculty member failed to disclose
involvement of a third party that was paid by a drug company to develop and draft a manuscript.
The university’s investigation concluded that the faculty member had made significant
intellectual contribution to the work and had a role in writing and reviewing the manuscript. The
university stated that such participation did not violate university policy, but “the incident was
the motivating event for the UW SOM to adopt its Ghost Authorship Policy in 2007.”
Seven medical schools stated that they had not received any allegations of ghostwriting
since 2004. Several of the schools added that faculty members are not required to report their
publication activities to the university. However, they noted that some reporting may occur
annually or periodically to department chairs, for example, when a faculty member is being
considered for promotion or as part of an annual review or grant submission. Thus, allegations
may be handled at the department level and never brought to the attention of the dean or other
senior university administrators.
In addition, pharmaceutical companies usually pay the medical education,
communications, or marketing companies or the medical writers for drafting the manuscript
instead of the authors listed in the publication, so these payments do not appear in financial
disclosure forms submitted by university faculty. As the documents from the Prempro litigation
show and as Wyeth told Committee staff, authors are not typically paid for “authoring” the
articles. Because the authors do not receive direct financial support from the company for the
articles, they typically do not acknowledge any outside funding. Disclosures of financial support
may occur when there is financial support for the underlying study, but in those cases there is
still a lack of acknowledgment of a company’s role in developing a manuscript. At times, a
company may offer authors an honorarium, but the author may refuse such money. For example,
one expert stated in an email to STI, the medical communications company hired by Merck, that
there was no expectation for an honorarium and “I really do not feel it is appropriate to be paid
for this type of effort.” See Attachment 4.
Duke does not explicitly prohibit ghostwriting, but the medical school stated, “it is
expected that a faculty member will maintain records that document his/her active participation
in the design, conduct or analysis of research, and will not accept authorship on clinical studies
where such active involvement cannot be demonstrated.” However, without department
verification of records or information maintained by the faculty member, the institution would
have to trust that the faculty member is compliant or rely on third party reports of potential
misconduct. UW Medicine stated that department chairs are responsible for disseminating and
enforcing the policies, but it is not clear what steps would be taken by the department chairs to
ensure compliance. Columbia also believes that requiring “public disclosure of the contributions
or participation of commercial entities to medical articles and papers is more effective than
internal disclosure to Columbia alone.”
11
Johns Hopkins Medicine includes in its “Rules and Guidelines for Responsible Conduct
of Research” a professional obligation for faculty, students, and fellows to inform superiors if
they have reservations about the integrity of the work of a university colleague. Johns Hopkins’
guidelines also state that the institution “recognizes the risks to persons who report research or
professional misconduct and has made every effort to protect them as well as those who might be
accused in error” and the institution “will adhere to federal rules and guidelines regarding the
protection of whistleblowers, as applicable.”
D. Strengthening journal authorship policies appears to have limited effect on
ghostwriting and disclosure of industry financing of medical articles
Last summer, Senator Grassley wrote to leading medical journals regarding their position
and policies on ghostwriting—American Journal of Medicine, Annals of Internal Medicine,
Annual Review of Medicine, Archives of Internal Medicine, JAMA, Nature Medicine, NEJM and
PLoS Medicine. Based on a review of authorship policies and journal responses, which were
submitted in July 2009, as well as other information and materials obtained by the Committee,
the staff found the following:
 All eight medical journals require that authors disclose potential conflicts of interest,
competing interests and/or potential biases.
 Annals of Internal Medicine, Archives of Internal Medicine, and JAMA explicitly address
ghostwriting and use the terms “ghostwriting” or “ghost authorship” in their policies and
guidelines for authors. In addition, the Archives of Internal Medicine and JAMA follow
the AMA Manual of Style: A Guide for Authors and Editors.15 This style guide explains
ghost authorship and discusses the requirements of authorship and guidelines for ensuring
that proper credit is given to individuals who contributed to the work, such as medical
writers and others who participate substantially in the writing and editing of the
manuscript.
 According to the documents provided to the Committee, the Annals of Internal Medicine,
the Archives of Internal Medicine, and JAMA also require the public disclosure and
acknowledgment of individuals who contribute to a publication, such as medical writers,
industry employees, and/or other contributing non-authors. The AMA Manual of Style
also notes that JAMA “discloses the affiliation and funding of individuals who contribute
to manuscripts but who are not authors.” In addition, it notes that such disclosure is
supported by the American Medical Writers Association and the European Medical
Writers Association “as it is more helpful to editors, reviewers, and readers than are
vague statements about writing or editorial assistance that gives no indication about
financial relationships.”
 JAMA requires authors to sign statements certifying that “all persons who have made
substantial contributions to the work reported in this manuscript (e.g., data collection,
15 AMA Manual of Style: A Guide for Authors and Editors (10th ed.), available at
http://www.amamanualofstyle.com.
12
analysis, or writing or editing assistance) but who do not fulfill the authorship criteria are
named with their specific contributions in an Acknowledgment in the manuscript.” The
Archives of Internal Medicine has a similar form. See Attachment 5.
 The Annals of Internal Medicine requires authors to sign statements attesting that “all
individuals who contributed to the manuscript have been appropriately acknowledged,
and also that all contributors who are not authors are named in the Acknowledgment
section.” See Attachment 5.
 The NEJM author disclosure form, which is the same form developed by ICMJE, asks if
the author or the author’s institution was paid for preparing the manuscript, and if so, by
whom and what was the nature of the compensation. However, it does not ask if
someone else was paid to assist with the manuscript. The Annals of Internal Medicine
disclosure form also asks if the author received payment for “involvement in the
preparation” of the manuscript.
 PLoS Medicine follows the guidelines of ICMJE on authorship and stated that the
involvement of any professional medical writer must be declared. Its policy also refers
authors to the European Medical Writers Association (EMWA) guidelines, which
recommend that medical writers and their funding source be acknowledged. Specifically,
EMWA states, “Identifying the writer, either as an author or contributor or in the
acknowledgements section, helps readers, reviewers, and journal editors to understand
how the manuscript was developed, and recognizes the writer’s involvement. Identifying
the writer’s funding source ensures transparency and makes readers aware of any
potential conflicts of interest.”16
 Nature Medicine stated that it has an authorship policy that defines what it means to be an
author and that the policy should prevent some authors’ willingness to appear on papers
in which they did not contribute. The journal added, however, that it is not strictly a
medical journal but rather a journal for and by basic researchers so “ghostwriting is not a
problem significant enough to warrant an official position against it.” A JAMA survey
presented at the International Congress on Peer Review and Biomedical Publication in 2009
found that the prevalence of ghost authorship was lowest in Nature Medicine compared
to the other medical journals but it also found that the prevalence of honorary authorship
was the highest in Nature Medicine.17
 Annual Review of Medicine stated that no ghostwriting is permitted and its approach for
avoiding conflicts is through careful author selection. Topics are chosen by its editorial
committee about 18 months in advance of publication, and the journal invites authors it
considers best qualified to write the articles. NEJM also stated that most editorials and
review articles are solicited by an NEJM editor.
16 Adam Jacobs and Elizabeth Wager, “European Medical Writers Association (EMWA) guidelines on the role of
medical writers in peer-reviewed publications,” Curr Med Res Opin, 2005; 21(2): 317-321.
17 Joseph Wislar, Annette Flanagin et al., “Prevalence of Honorary and Ghost Authorship in 6 General Medical
Journals, 2009,” Paper presented at the Sixth International Congress on Peer Review and Biomedical Publication,
Vancouver, British Columbia, Canada, Sept. 10, 2009.
13
 The American Journal of Medicine stated that it rejects any manuscript that appears to be
written by someone other than the authors of the article.
 PLoS Medicine stated that the journal has had a longstanding interest in documenting
ghostwriting and its negative effects. In fact, PLoS Medicine has published several
articles and opinion pieces on the subject over the last three years.
Despite efforts by journals in recent years to strengthen their authorship and publication
requirements, a survey released last year found that the prevalence of ghostwriting has not
changed much in the past decade. In 2009, JAMA presented an abstract on its survey of authors
of 630 research articles, review articles and editorial/opinion articles that were published in six
medical journals in 2008.18 The six journals were Annals of Internal Medicine, JAMA, Lancet,
Nature Medicine, NEJM, and PLoS Medicine. JAMA found that 26 percent of the articles had
honorary authors, 8 percent had ghost authors, and 2 percent had both. The survey also showed
that these numbers did not differ significantly from a 1996 study, which found 19 percent
honorary authorship, 11 percent ghost authorship, and 2 percent of both.19 In addition, JAMA
reported that the prevalence of ghost authors was highest in NEJM (11 percent), lowest in Nature
Medicine (2 percent), and “no significant differences were found between journals requiring
author contribution disclosures and those that do not.” Thus, it appears that despite policies to
ensure that all authors who contribute to a publication are identified and that the authors listed in
fact contributed substantially to the publication, the prevalence of ghostwriting remains largely
unchanged.
Nevertheless, clear and strong policies are needed not only to ensure that all listed
authors meet the authorship criteria but also, more importantly, to ensure appropriate public
disclosure of the roles of non-authors, in particular the role of a pharmaceutical or device
company in initiating and paying for a manuscript. Although the JAMA survey did not find that
the prevalence of ghostwriting has changed much, it did find a slight decrease. The challenge is
educating authors of their responsibilities under the policies and ensuring compliance.
Following up on red flags would also allow journals to prevent manuscripts without appropriate
acknowledgments and disclosures from being published. For example, MedPage Today quoted
Dr. Harold Sox, editor of the Annals of Internal Medicine, in April 2008, stating that, “Ideally,
we’d be calling and saying, what exactly did they do,” but the journal didn’t have a policy on
following up routinely.20
In addition, an editor-in-chief of a medical specialty journal contacted Senator Grassley
and said that his journal handles articles that it suspects were ghostwritten by questioning or
editing the articles. He informed Committee staff that at least one third of the papers submitted
to his journal were written by science writers hired by an agency and paid for by a
pharmaceutical company. The editor added that in some cases, it was clear to him that the
academic expert had limited input in the writing of the article and that while authors now
18 Id.
19 Annette Flanagin, Lisa A. Carey et al., “Prevalence of Articles with Honorary Authors and Ghost Authors in Peer-
Reviewed Medical Journals, JAMA, 1998; 280(3): 222-224.
20 John Gever, “Fewer Medical Journal Articles Planted by Phantom Authors,” MedPage Today, April 14, 2008.
14
acknowledge that they received editorial assistance, what is unclear is whether or not the
academic expert evaluated the implications of what he was submitting for publication. That
editor was also concerned that medical literature “has become inundated with repetitive
promotional articles.”
E. National Institutes of Health does not have explicit policies on disclosure of industry
financing of ghostwritten articles
In August 2009, Senator Grassley wrote to NIH to bring to the agency’s attention several
NIH-funded researchers who were authors on what appeared to be ghostwritten articles paid for
by the pharmaceutical industry and ask NIH about its policies on ghostwriting. NIH responded
that it did not use the term “ghostwriting” in its policies. Instead, NIH stated that Federal
regulations and policies related to Public Health Service supported research could be applicable
to ghostwriting and noted that a case involving NIH-funded researchers may be appropriate for
consideration as a case of plagiarism.
The NIH also stated that it “acknowledges the importance of journal policies and
practices that promote transparency by disclosing investigators’ financial interests.” However,
as discussed earlier in this report, if a company does not provide direct financial support to the
author of an article, then the company’s role may not be captured in the author’s financial
disclosures to the journal editors or in the published article. NIH distributes billions of dollars of
extramural funds each year to support medical research that helps improve the health of the
American people. Thus it is important that NIH allocates those public funds responsibly. NIH
may want to consider requiring grantees to certify that they disclosed all potential conflicts of
interest, not only financial interests. Furthermore, during his inquiries into NIH’s policies on
financial conflicts, Senator Grassley found that NIH relies on the academic institution to
document, manage and address its faculty members’ financial conflicts. The institutions,
however, do not verify the disclosures, thus they cannot address conflicts that are not accurately
reported to them.
CONCLUSIONS
Just as the Physician Payments Sunshine Act is intended to promote transparency in the
financial relationships between healthcare companies and physicians that may influence medical
behavior, Senator Grassley’s inquiry into ghostwriting is intended to shed light on interactions
between the pharmaceutical industry and academic physicians and scientists. In the interest of
transparency and accountability, all parties who contribute substantively or financially to a
publication should be acknowledged. Only then can readers understand the context of a study
and be aware of any commercial interests that initiated and influenced the results or
recommendations presented in the publication.
The pharmaceutical company who pays a third party to develop articles that present the
company’s products in a positive light stands to benefit from a manuscript that is published in a
medical journal. As Wyeth’s own presentation slide states, “A scientific publication plan is as
15
vital as a carefully designed media plan in overall product marketing.”21 Academics who author
the articles also benefit, even if they do not receive payment from the company for an article.
Successful publications raise the authors’ visibility in their fields and may lead to promotions or
more research funding opportunities. For physicians and scientists in academia, it’s “publish or
perish”—a phrase that is commonly used to describe the pressure that faculty feel to publish
frequently in order to further their careers.
It is encouraging that many of the medical schools and journals that responded to Senator
Grassley’s inquiries have policies against ghostwriting. More significantly, it was welcome
news to find that some of the schools and journals already require the disclosure or
acknowledgment of individuals or entities involved in developing a manuscript. These
relationships, however, should be disclosed regardless of whether or not a commercial entity
specifically provided funding to the faculty or to an institution for work on the article. As stated
in this report, a pharmaceutical or device company does not typically provide financial support to
the faculty or institution for the articles, but instead to a third party who drafts the article for the
faculty member to sign on as the author.
In addition, while acknowledging “editorial assistance” may provide a clue that the
article was potentially ghostwritten, it does not give the reader any information about the roles of
others, in particular the pharmaceutical or device company, that may have helped shape the
publication. If a pharmaceutical or device company initiated, drafted, reviewed or paid for a
manuscript, the final publication must disclose that information and include a detailed
description of the company’s contribution to the article.
Drs. Lacasse and Leo stated that when academic medical centers do not prohibit their
faculty from participating in ghostwriting, the centers:
enable the pharmaceutical company to covertly shape the medical literature in
favor of commercial interests. When a pharmaceutical salesperson hands a
clinician an article reprint, the name of the institution on the front page of the
reprint serves as a stamp of approval. The article is not viewed as an
advertisement, but as scientific research.22
The Institute of Medicine (IOM) also recommended that academic medical centers adopt
policies prohibiting ghostwriting. It raised concern about the practice, stating:
Such arrangements (which are essentially gifts) send the wrong message about the
values of intellectual independence, professional ethics, accountability, and
evidence-based medicine. In the context of research, they raise questions about
the objectivity of research reports that other researchers as well as practitioners
and developers of practice guidelines rely on.23
Specifically, the IOM recommended:
21 Available on the Drug Industry Document Archive at http://dida.library.ucsf.edu/pdf/atb37b10.
22 Lacasse and Leo, supra note 7.
23 Institute of Medicine, Conflict of Interest in Medical Research, Education, and Practice (2009).
16
For all faculty, students, residents, and fellows and for all associated
training sites, academic medical centers and teaching hospitals should
adopt and implement policies that prohibit…educational presentations or
scientific publications that are controlled by industry or that contain
substantial portions written by someone who is not identified as an author
or who is not properly acknowledged.24
In addition, Dr. Lisa Bero and Ms. Jenny White of UCSF recommended uniform
standards for all journals so that companies will not be able to target journals that have weaker
policies on authorship and conflicts of interest. Dr. Bero and Ms. White found in their review of
journal policies regarding ghostwriting that while the journals had policies requiring disclosure
of conflicts of interest, a majority did not address ghost authorship. They also noted that journals
varied in their effectiveness in verifying full disclosure of conflicts of interest and authorship.25
Once clear and strong policies regarding authorship and disclosures/acknowledgments
are in place, the challenging task is ensuring compliance. Academic institutions rely on their
faculty to report accurately and honestly. Unfortunately, Senator Grassley found several cases
where medical faculty failed to report millions of dollars received from pharmaceutical
companies on their financial disclosure forms.26 Similarly, unless a journal looks out for signals
or clues that an article may have been ghostwritten and follows up on what is disclosed to them
by the authors, they must rely on the authors to report accurately and honestly. It is important
that the academic institutions and medical journals educate authors on their policies and author
responsibilities. It may be the case that faculty are not knowledgeable of the requirements and
their failure to report is not an intention to withhold or deceive. The authors of one opinion piece
published in PLoS Medicine in February 2009 believe that “existing guidelines already
emphasize the need for appropriate disclosure of writing assistance” and suggested that “one of
the most practical ways to tackle ghostwriting could be the mandatory use of a checklist that
could help editors detect ghostwriting and help authors avoid ghostwriters.”27 The checklist
would include a question asking if the source of funding for a medical writer’s services is
identified in the acknowledgments. The authors proposed that the checklist could be included in
a journal’s instructions to authors.
According to a MedPage Today article, Dr. Adriane Fugh-Berman, a researcher at
Georgetown University, proposed that journals could ask more questions of named authors to
curb the practice of ghost authoring.28 Dr. Fugh-Berman pointed out that the American Family
Physician, for example, asks specific questions in its author conflicts of interest disclosure form
about manuscript writing assistance and the involvement of a medical communications company
24 Id.
25 Lisa Bero and Jenny White, “Ghost Writing: How Some Journals Aided and Abetted the Marketing of a Drug,
Paper presented at the Sixth International Congress on Peer Review and Biomedical Publication, Vancouver, British
Columbia, Canada, Sept. 10, 2009.
26 See http://finance.senate.gov/newsroom/ranking/release/?id=67b016e1-58dd-4f7f-acaa-05fce22393a1; see also
http://finance.senate.gov/newsroom/ranking/release/?id=91aa840c-e86d-476d-a8fa-300220d602e0.
27 Peter C. Gotzche, Jerome P. Kassirer, Karen L. Woolley et al., “What Should Be Done to Tackle Ghostwriting in
the Medical Literature?” PLoS Med, 2009; 6(2): 122-125.
28 Gever, supra note 20.
17
or professional writers. If an author answers yes, the form asks who paid for the assistance. The
form also asks authors whether or not “a pharmaceutical company, public relations firm or any
commercial entity sponsor[ed] the substance or creation of [the] article directly or indirectly.”
See Attachment 6.
PLoS Medicine also stated that before it accepts a paper for submission, the editors ask
the authors if anyone prompted or paid them to write the article and the extent to which a
professional writer contributed to the article. In addition, the journal pointed out that when
individuals other than the authors inquire about the progress of the peer review of a study, it is a
red flag and prompts the editors to immediately contact the author. Such follow-up may help
capture third party contributions that were not initially disclosed.
In addition, the Centers for Medicare and Medicaid Services (CMS) allows its contractors
to use peer-reviewed literature to determine if an off-label use of a drug or biologic is a
medically accepted indication for purposes of coverage. CMS should consider requiring that the
peer-reviewed articles follow strict authorship and disclosure/acknowledgment policies. Such
policies might prevent coverage decisions based on the findings and recommendations of articles
that do not have adequate disclosures that they were initiated and paid for by a pharmaceutical
company.









Manuscript Title
By signing below, all authors acknowledge that they have read 1) the statement
on authorship, dual commitment, and contribution to authorship and 2) the
statement on copyright transfer or federal employment. They also agree to assign
the responsibility for changes to galley proofs to one designated author. The name
of this author, as well as the name of the corresponding author and the author to
receive reprint requests, should be indicated in the Author Information section,
below. Authors of editorials, letters, On Being a Doctor/Patient pieces, and Book
Notes need not designate author contributions. The corresponding author must
also sign the acknowledgment statement at the bottom of this page. Authors who
do not already receive Annals will receive a complimentary copy of the issue in
which their article appears (indicate by checking the box under Author Information).
This document may be photocopied for distribution to coauthors for their
signature. Every author must complete a form. Please mail all copies to Harold
Sox, MD, Editor, Annals of Internal Medicine, 190 N. Independence Mall West,
Philadelphia, PA 19106-1572, USA, or fax to 215-351-2619. Please put your
manuscript number at the top of the form.
Authorship, Dual Commitment, and Contributions to Authorship
● I have contributed directly to the intellectual content of this paper and have
agreed to have my name listed as an author on the final, revised version.
● For papers with more than one author: I agree to allow the corresponding
author to make decisions regarding prepublication release of information in the
paper to the media, federal agencies, or both.
● Financial interests, direct or indirect, that exist or may be perceived to exist for
individual authors in connection with the content of this paper have been disclosed
to Annals in the cover letter. Sources of outside support of the project are
named in the cover letter, and the role of funding organizations, if any, in the
conduct of the study is described in the Methods section of the manuscript.
In the spaces marked “Contribution Codes,” authors should mark those code
letters from the box that designate their own substantive contribution(s) to the
paper. Any contribution not described in the box should be indicated in the
space for “Other contributions.”
Contribution Codes
a Conception and design
b Analysis and interpretation of the data
c Drafting of the article
d Critical revision of the article for important
intellectual content
e Final approval of the article
f Provision of study materials or patients
g Statistical expertise
h Obtaining of funding
i Administrative, technical, or logistic support
j Collection and assembly of data
Copyright Transfer
Annals of Internal Medicine is owned, published, and copyrighted by the American
College of Physicians (ACP). In the event that the ACP publishes my work
and in consideration of the editing and publication of my work and the professional
benefits relating hereto, I transfer to the ACP all right, title, and interest to
all parts of the written work named above. The ACP shall own the work, including
1) copyright; 2) the right to grant permission to republish the article in whole
or in part, with or without fee; 3) the right to produce preprints or reprints and
translate into languages other than English for sale or free distribution; 4) the
right to republish the work in a collection of articles in any other mechanical or
electronic format. In addition, I affirm that the work has not been previously
published, is not subject to copyright or other rights except my own to be transferred
to the ACP, and has not otherwise been submitted for publication, except
under circumstances communicated to the ACP in writing at the time the work
was first submitted.
Authors of manuscripts reporting research funded by NIH are granted permission
to provide a copy of the accepted manuscript to the NIH for public archiving
in PubMed Central. “Accepted manuscript” refers to the prepublication version
for which Annals has issued a notice of final acceptance. Submission of
copies of the final published version (i.e., PDF or the HTML version downloaded
from www.annals.org) to PubMed Central would violate the copyright agreement.
U.S. Federal Employees: I was a U.S. federal employee when this work was done and the manuscript was prepared for publication. This work is not protected by the
Copyright Act and ownership cannot be transferred. Initials:
Author Information
Signature Name Date signed
Institution Street address City State ZIP code
Phone number Fax number Contribution codes (from box)
1) 2) 3)
Other contributions
Please check all of the following boxes that apply to you:

Author responsible for galley proofs
Author to receive reprint requests
Corresponding author
ACP member/Annals subscriber
Acknowledgment
I attest that all individuals who contributed to the manuscript have been appropriately acknowledged, and also that all contributors who are not authors are named in the
Acknowledgment section and have agreed in writing to be named.
Yes
No
Corresponding Author Signature (needed only on corresponding author’s form) Date Signed
Annals of Internal Medicine Authors’ Form


Link to Story