Good Publication Practice: why we need another set of guidelines
E Wager, EA Field, NB Glazer, L Grossman
Reprinted with permission from Current Medical Research & Opinion
Introduction
Pharmaceutical companies' relations with
clinicians, academics, medical journals, and the public have often been
characterized by conflicting interests and tensions, and these negative
aspects have received considerable attention [1,2]
. Yet these different constituencies
often work closely together, especially during clinical trials, and
successful collaboration is critical to the development of new medicines.
While the conduct of the clinical trials themselves is heavily regulated,
until recently much less attention has been paid to the process of
publishing their findings.
The evolution of GPP
In
November 1998, journal editors, academics / investigators and
pharmaceutical company employees involved with publications took part in a
retreat organized by the Council of Biology Editors (now Council of
Science Editors) [3]. Over the course of the
meeting it became clear that there was a lack of understanding about the
ways in which the different constituencies operated and concern about the
ways in which publications arising from company-sponsored research were
sometimes developed. Those of us present from within the industry and
closely involved with the publication of company-sponsored clinical trials
agreed that it would be helpful to identify some principles and common
standards to address the concerns about publication practices. We set up a
Working Group that drafted ‘Good Publication Practice: Guidelines for
Pharmaceutical Companies’ (or GPP, see Appendix). These guidelines are
designed to increase the transparency of the processes involved in the
publication of industry-sponsored trials and to establish standards for
these. Although they predate the most recent statements by the
International Committee of Medical Journal Editors (ICMJE) [1] we believe that they remain
timely and pertinent to them.
We
consulted widely within our several companies and eventually agreed on a
document that addressed the important issues. During 2000 we sent this document
to 70 major pharmaceutical companies and publicized its existence in
several journals [4-7].
Although we believed that GPP would
have the greatest impact if it was adopted by individual companies, we
also discussed the guidelines with the Pharmaceutical Manufacturers
Association (PhRMA) in the United States, and the Association of the
British Pharmaceutical Industry (ABPI). The guidelines have also been
presented at meetings of the Council of Science Editors, European
Association for Science Editors, the American Medical Writers Association
and the Cochrane Collaboration.
Since
the initial meeting in 1998, the membership of the Working Group has
evolved as members changed jobs or companies. It is probably typical of
current patterns of employment in the pharmaceutical industry that, of the
six original members of the group who signed the letter in
JAMA
[4], three have
moved to new companies, one has changed responsibility within the company,
one has gone freelance and one has retired. Approval was also delayed or made
difficult because of company mergers, which have occurred in three of our
six original employing companies.
For all these reasons, we (the current members of the Working
Group) have decided to publish the guidelines in our individual capacities
rather than as representatives of any particular companies, but we
acknowledge the support of our various employers over this period, the
contributions of previous members of the group, and the numerous other
people who have contributed to the development of the
guidelines.
With this
publication, we hope that the GPP guidelines will be discussed further and
many companies will wish to endorse them.
Why do we need more
guidelines?
Publication in peer-reviewed journals is
an integral part of biomedical research. While it is not immune from
inappropriate behavior and even malpractice, it is less heavily regulated
than other aspects of the process. Many of the issues addressed by the GPP
guidelines, such as failure to publish results from negative or
disappointing studies and inappropriate allocation of authorship, are not
unique to pharmaceutical-industry sponsored trials. However, responsible
companies cannot ignore them and are often in a good position to address
them. Documents such as the CONSORT statement [8], the ICMJE’s Uniform
Requirements
[9], and
journals’ instructions to authors are helpful, but none was designed
specifically for company sponsors of large trials, and they do not address
all the concerns that have been raised.
What issues do
the GPP guidelines seek to address?
The
two main themes of the GPP guidelines are publication bias and the
relationship between pharmaceutical companies and academic investigators.
Publication bias may result from either the non-publication of
inconclusive or unfavourable findings or by redundant publication of
positive findings. These problems, which are not unique to
industry-sponsored trials, may be caused by a number of factors
[10] but are well
documented
[11-13]. The GPP guidelines aim to
reduce publication bias in three ways. They encourage companies to
endeavour to publish results from all their studies and to avoid redundant
publication. However, they
recognize that results may legitimately be presented at several scientific
conferences, and that secondary analyses or follow-ups may be appropriate.
The guidelines therefore recommend the inclusion of unique trial
identifiers in all publications to increase transparency and facilitate
the preparation of systematic reviews.
The
successful conduct and publication of large-scale clinical trials require
close collaboration and partnership between clinicians and company
scientists. Suggestions that companies should have less involvement in
preparing papers [1,2] go against the greater
transparency that has been achieved by the contributorship approach to
listing authors [14,15] and prevents recognition of
the important intellectual and scientific contributions of company
employees
[16].
The
role of professional writers working for pharmaceutical companies is dealt
with in detail. This has been an area of particular concern, and some have
suggested that the practice should be discouraged altogether
[17, 18]
. However, we believe that
preventing professional writers from assisting with publications would
exacerbate the problems of non-publication and delayed publication and
that, when such writers are an integral part of the publication process,
openly acknowledged, and working within the guidelines, they can improve
both the quality and the timeliness of publications [19,
20].
The
GPP guidelines apply to publications arising from industry-funded clinical
studies. This includes trials
used to support licensing applications (Phase II and III) and those funded
by manufacturers after products are approved (Phase IV). The guidelines do not cover
studies performed and published independently by investigators (even when
these involve some company support, e.g. supply of drugs), although we
hope the principles may still be helpful in those cases. The GPP
guidelines also apply to other types of publication that are initiated by
companies, such as review articles and secondary papers.
The
GPP guidelines are designed to be followed by pharmaceutical companies and
any company or individual working on their behalf, such as contract
research organizations, communications agencies and freelance
contractors. They also set
out some of the responsibilities of healthcare professionals working with
companies as investigators or authors of publications.
We
hope that companies will base policies and procedures on the guidelines
and devise their own ways of ensuring that they are followed. Therefore we have aimed to set out
principles rather than dictate specific procedures or mechanisms. Since these are voluntary
guidelines, the language is that of recommendation rather than an
imperative (i.e. they set out what companies should do rather than what
individuals must do).
The
GPP guidelines for pharmaceutical companies do not aim to replace existing
documents such as CONSORT
[8] or the ICMJE
recommendations
[9], and we hope
that companies will also consult these and incorporate them into their
policies and practices.
Although the guidelines were written with
pharmaceutical companies in mind, many of the issues they address occur in
other sectors. In particular, publication bias caused by under-publication
of negative or disappointing findings is known to affect studies
regardless of the source of their funding [11,
21]
. Therefore we hope that other
funding bodies, academic institutions and perhaps research review boards /
ethics committees [22]
might seek to
ensure that results from all studies are published. This principle is now included in
the latest version of the Declaration of Helsinki
[23] which may encourage
individual clinicians to take responsibility for this.
Our
aim in publishing the GPP guidelines is to stimulate discussion between
journals, investigators and trial sponsors and to provide guidance to
those who seek it. We also hope that pharmaceutical companies and others
involved in developing publications will endorse them. However, we
recognize that developing guidelines is an iterative process, and it is
never possible to consult with everybody who might have something useful
to contribute. We also recognize that experience of implementing the
guidelines in different companies may raise points that require
clarification or expansion. Therefore we plan to review the document at
regular intervals. Ideally such a review would take place at a forum in
which the different constituencies are equally represented, perhaps along
the lines of the initial retreat, with a similar small Working Group
convened to act on any recommendations.
We
hope that the GPP guidelines represent a first step in establishing a
common standard for the publication of industry-sponsored studies, and
that regular review and discussion will lead to continually rising
standards.
1.Davidoff F, DeAngelis C, Drazen JM, Hoey J,
Hřjgaard L, Horton R et al. Sponsorship, authorship, and accountability.
Ann Int Med 2001;135:463-6.
2.Bodenheimer T. Uneasy alliance. Clinical
investigators and the pharmaceutical industry. N Engl J Med
2000;342:1539-44.
3.Wager E. Common aims / different languages:
increasing understanding among medical journals, academia and industry.
CBE Views 1999;22:41-2.
4.Wager E, Tumas JA, Field EA, Glazer NB,
Schulz G, Grossman L. Improving the conduct and reporting of clinical
trials. JAMA 2000;283: 2788-9
5.Wager E, Tumas JA, Field EA, Glazer NB,
Schulz G, Grossman L. Improving the conduct and reporting of clinical
trials. Can J Gastroenterol 2000; 14: 749.
6. Wager E, Tumas JA, Field EA, Glazer NB,
Schulz G, Grossman L. Good
publication practice. Ned Tijdschr Geneeskd 2000; 144:399-400.
7. Sharp D. Drug industry code proposed on
‘ghost’ writing. Lancet 2000; 355:1084.
8.Begg C, Cho M, Eastwood S, Horton R, Moher
D, Olkin I et al. Improving the quality and reporting of randomized
controlled trials. JAMA 1996;276:637-9.
9.International Committee of Medical Journal
Editors. Uniform requirements for manuscripts submitted to biomedical
journals. www.icmje.org.
Accessed 25th March 2003.
10. Gibbs TG, Wager E.
Realities of trial registration: the Glaxo Wellcome experience.
International Journal of Pharmaceutical Medicine 2000;14:203-5.
11. Easterbrook PJ, Berlin
JA, Gopalan R, Matthews DR. Publication bias in clinical research. Lancet
1991;337:867-72.
12. Dickersin K. The
existence of publication bias and risk factors for its occurrence. JAMA
1990;263:1385-9.
13. Stern JM, Simes RJ.
Publication bias: evidence of delayed publicatoin in a cohort study of
clinical research papers. BMJ. 1997;315:640-5.
14. Davidoff F. News from
the International Committee of Medical Journal Editors. Ann Intern Med
2000;133:229-231.
15. Rennie D, Flanagin A,
Yank V. The contributions of authors. JAMA 2000;284:89-91.
16. Wager E. Drug industry
is increasingly allowing employees to be named as authors (letter). BMJ.
1996;312:1423.
17. Reed CR, Camargo CA.
Recent trends and controversies in industry-sponsored clinical trials.
Academic Emergency Medicine 1999;6:833-9.
18. Committee on Publication
Ethics. The COPE Report 2000: Annual Report of the Committee on
Publication Ethics. London: BMJ Books; 2000.
19. Grossman L. Ghostwriting
(letter). Lancet 1998;351:1741.
20. Wager E. Ghostwriting
(letter). Lancet 1998;351:1741.
21. Blumenthal D, Campbell
EG, Anderson MS, Causino N, Seashore Lewis K. Witholding research results
in academic life science. JAMA 1997;277:1224-8.
22. Antes G, Chalmers I. Under-reporting of clinical trials is unethical. Lancet 2003;361:978-9.
23. World Medical Association. Declaration of
Helsinki. Ethical principles for medical research involving human
subjects. 2000. 52nd WMA
General Assembly, Edinburgh, Scotland, October 2000.