Reprinted from Current Medical Research & Opinion 2003;19(3):149-154
Download PDF (printable format): gpp.pdf
Aim
The aim of these guidelines is to ensure that publications are produced in a
responsible and ethical manner. They are designed to be applied in conjunction
with other guidelines such as those from the International
Committee of Medical Journal Editors the CONSORT
group, and individual journals. In addition,
they may be incorporated into the more detailed operating procedures of
individual companies.
Scope
These guidelines are designed for use by pharmaceutical companies, other
commercial organizations that sponsor clinical trials, and any companies or
individuals who work on industry-sponsored publications (e.g., freelance
writers, contract research organizations, and communications companies). For
simplicity, the terms 'company' and 'employee' are used in these guidelines, but
they should be taken to include all of these parties.
These guidelines cover publications in biomedical journals, including both traditional print and electronic journals, and oral / audiovisual presentations at scientific meetings. They cover peer-reviewed publications (such as original research articles, review articles, sponsored supplements, and abstracts) and non-peer-reviewed scientific communications (such as posters, lectures, book chapters, and conference proceedings). However, they do not cover promotional materials, which are regulated by specific national codes and legislation.
Publication standards
Companies should endeavour to publish the results from all of their clinical
trials of marketed products. These publications should present the results of the research
accurately, objectively, and in a balanced fashion. Anyone working on company
publications should follow relevant external guidance such as the 'Uniform
Requirements for Submission of Manuscripts to Biomedical Journals' issued by the
International Committee of Medical Journal Editors (ICMJE)1 and the CONSORT
statement2. Additional guidelines relating to publications from
company-sponsored research are outlined below.
Relationship between the company and external investigators
The contractual relationship between companies and external investigators or
consultants should be set out in a written agreement. This should cover
publication policies and ownership of data.
Companies should be responsible for coordinating the publication of multicentre trials to ensure that they are reported in a responsible and coherent manner (i.e., results from data subsets should not be published in advance of or without clear reference to the primary paper and should not constitute redundant or prior publication). Therefore, companies should maintain the right to be informed of any plans for publication and to review any resulting manuscripts before they are submitted. Companies should not suppress or veto publications; however, it may be appropriate to delay publications to protect intellectual property.
All authors, external and internal, should have access to the statistical reports and tables supporting each publication. When differences about the presentation or interpretation of findings arise between company scientists and external investigators, both parties should work to find a mutually acceptable solution through honest scientific debate.
Premature publication
While it is acceptable to present abstracts, posters, or lectures at biomedical
conferences before the full publication of results, care should be taken to
avoid premature or inappropriate publication (e.g., through press releases).
Most journals provide guidelines on what constitutes prior publication and
impose embargoes on contact with the press before publication. These are also
outlined in the ICMJE guidelines1. In the case of findings with major
implications for public health or of great commercial sensitivity, it may be
helpful to discuss with the journal editor the timing of publication and
proposed approaches to the media.
Duplicate / redundant publication / multiple submissions
Most peer-reviewed journals will consider only papers that have not appeared or
been accepted for publication in full elsewhere. Presentation at scientific
meetings does not constitute full publication, so prior publication of abstracts
or posters does not affect the consideration of full papers. These conditions
are set out in journals' instructions to authors and the ICMJE guidelines,1
which should be followed in all cases. Because journals do not accept duplicate
publications and because they do not want to waste the time of their reviewers,
it is not acceptable to submit a paper to more than one journal at a time.
Companies should avoid duplicate publication of the primary results of a study in peer-reviewed journals. Cases in which secondary publications might be acceptable include symposium proceedings, results of significant and scientifically sound alternative analyses, or grouping of data from more than one study. However, such publications should not precede the original publication, should reference the original publication, and should include a unique study identifier as described below. Full peer-reviewed publications should contain references to all previous presentations of the data (e.g., abstracts). Translations of papers into different languages are usually acceptable as long as the original source of the publication is clearly acknowledged.
Many major biomedical meetings discourage repeat presentations of findings that have previously been presented to substantial audiences; the guidelines for each individual meeting should be observed. However, there is no absolute rule against submitting several abstracts presenting the results of a single study to several conferences unless this breaches the guidelines of the individual meetings. Closed presentations to inform investigators of results should not jeopardize publication or wider presentation of results at public meetings.
Identification of studies
Identification of clinical trials by the use of a study, trial registry, or
protocol number helps readers and those performing systematic reviews by making
it clear when data from the same patients are being presented in different
publications (e.g., in abstracts and then a full paper, or when interim or
long-term follow-up findings or secondary analyses are presented). A unique
study identifier should therefore be included in all publications.
Authorship
The ICMJE guidelines1 are a good starting point for determining who qualifies to
be an author, but they do not provide detailed guidance applicable to all situations. Furthermore, some
journals have adopted a system of listing contributors rather than authors.
Therefore, the individual requirements of different journals should be
respected. Whatever criterion for listing is used, it should be applied in the
same way to both external investigators and company employees. Companies should
ensure that all authors fulfil the relevant criteria and that no authors who
meet the criteria are omitted from the submitted manuscript. The order in which
authors/contributors appear on a publication should be negotiated between all
authors/contributors. It may be helpful for companies to outline authorship
policies in the investigators' agreement.
Acknowledgments
The Acknowledgments section of a paper should list those people who made a
significant contribution to the study but do not qualify as authors. It should
also be used to acknowledge the study's funding and the company's involvement in
the analysis of the data or preparation of the publication unless this is
apparent from the list of contributors/authors.
The role of professional medical writers
The scientists, healthcare professionals, and statisticians who were involved
with the design, conduct, and interpretation of a study (either as company
employees or external investigators) should participate in the preparation of
publications arising from the data. However, since these people may lack the
time, expertise, or language skills to produce high-quality and timely
manuscripts, companies may wish to employ professional medical writers to
facilitate the publication process. The writer may provide publication expertise
and assistance with writing, editing or preparing manuscripts, or collating
comments from contributors. When a professional medical writer is involved with
a publication, the following guidelines should be followed to ensure that the
opinions of all authors are fully represented in the publication.
· The named author(s)/contributors must determine the content of the publication and retain responsibility for it.
· The medical writer should begin drafting the manuscript after consultation and discussion with the named author(s)/contributors. It is often helpful if the author(s)/contributors and the medical writer agree on an outline of the paper before detailed writing begins.
· The named author(s)/contributors should be given adequate time to comment on an early draft of the manuscript.
· The medical writer should remain in close and frequent contact with the author(s)/contributors throughout the development of the manuscript.
· The named author(s)/contributors should approve the final version of the manuscript before it is submitted.
· The lead author should be responsible for submitting the manuscript to the journal and acting as the primary contact for interactions with the journal editor.
· The contribution of the medical writer should be acknowledged.
The use of professional writers may be particularly helpful when companies publish the results from large, multicentre studies involving many contributors. The formation of a writing committee involving the medical writer may facilitate this process. While it is acceptable for professional writers or authors’ editors to assist authors who have written editorials or opinion pieces (e.g., to improve the written style of authors whose first language is not English), it is not usually appropriate for them to prepare the first draft of such articles.
Responsibility for implementing the guidelines
Company employees who are involved with publications and people who are hired by
companies to work in this area should be familiar with these guidelines.
Companies should ensure that appropriate management structures are in place to
implement the guidelines. Company procedures for the review of manuscripts
should ensure that approval for submission is given in a timely manner. (Most
companies have a procedure in place for medical/legal review or 'copy approval',
and it may be helpful to append details of this here.)
References
1
Uniform requirements for manuscripts submitted to biomedical journals and
separate statements from the International Committee of Medical Journal Editors.
http://www.icmje.org
2 The CONSORT statement. http://www.consort-statement.org