Crystal clear

Much as was seen with the recent UK parliamentary expenses scandal, simply following guidelines in letter but not in spirit is not acceptable. An increasing awareness of issues regarding  authorship, conflicts of interest, and the general publication ethics of scientific research have led to the publication, in 2009, of a new set of guidelines, the Good Publication Practice for Communicating Company Sponsored Research: the GPP2 guidelines.

These guidelines were created by a working group, brought together by the International Society for Medical Publication Professionals (ISMPP), a not-for-profit organisation that has a mission to advance the medical publication profession through education and advocacy; drive integrity, excellence and transparency in medical publications; and lead the establishment and adoption of medical publication standards and best practices.

What is the reasoning behind creating the new guidelines? Recent discussion has focused on the need for an improvement in the way company-sponsored research is communicated. In addition, there is a general ethical scepticism about the aims of the sponsors of research. There are also specific concerns around the way they communicate (or do not communicate) the results of such research. Therefore, there has been a growing feeling in the industry that greater and greater transparency is the only way to improve ethical practices.

Blurred line
For example, numerous professional bodies have highlighted the valid role for professional medical writers in the reporting of scientific research, including the World Association of Medical Editors (WAME)  and the Association of American Medical Colleges (AAMC), but the line between appropriate medical writing support and ghost writing remains blurred in the lay press and with some journal editors. In addition, new legislation, such as US Food and Drug Administration Amendments Act (FDAAA), has also changed the way sponsors, publishers and editors must think about publication. As the publication landscape has changed so much since the original publication of GPP in 2003, it was necessary to update the guidelines to provide more current and relevant guidance.

How was gpp2 created?
Taking the excellent original GPP publication as a basis for discussion, the GPP2 steering committee, led by Chris Graf, drafted a revised guideline. They then recruited a large number of international volunteers to form an expert consultation panel. This resulted in the draft guidelines being circulated to 193 reviewers from a wide variety of job functions, including pharmaceutical professionals, biotechnology professionals, journal editors, journal publishers and representatives from academic centres and professional organisations.

Overall, 116 sets of comments were received, with the steering committee assessing and incorporating the most frequent and important comments for inclusion in the final draft.

What is new?
The enduring theme in GPP2 is to increase transparency in sponsored publications. There is a number of areas where this is applicable:

1. Contributorship model of authorship
Although authorship criteria are well defined by the International Committee of Medical Journal Editors (ICMJE) in Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication, an increase in transparency can be gained by adopting the contributorship model. The guidelines recommend that "…clear, concise descriptions of the role of each contributor during preparation of the article or presentation… are made in an acknowledgement…". Contributions that should be acknowledged include, but are not limited to, study conception or design, conducting or managing the study, interpreting data, performing statistical analysis, drafting a manuscript, or critically reviewing a manuscript.

2. Conflicts of interest/funding disclosures
Authors should disclose all potential financial and non-financial conflicts of interest when submitting their publication. The GPP2 guidelines recommend that all authors make such disclosures for all publications, irrespective of whether the journal or congress specifically requests them. GPP2 recommends that honoraria should not be paid to authors of peer review publications, although appropriate out-of-pocket expenses, such as travel to a congress to present the primary study results, are appropriate. In practical terms, the golden rule is to disclose whether you feel there is a conflict of interest or not. In other words, let others decide whether there is a conflict.

3. Role of the sponsor/medical writers
GPP2 encourages sponsors to take a proactive role in ensuring ethical publication of their research by developing a list of sponsor responsibilities, author responsibilities and joint responsibilities. Furthermore, these roles and responsibilities should be formalised by means of a written agreement between the sponsors and authors, with recommendations that this written agreement confirms the sponsor's agreement to provide the authors will full access to the data and, further, to provide the authors with the freedom to make public the study results. Practically speaking, these written agreements have been standard practice for the conduct of the trials themselves, but the exact form for publications may need to evolve over the next few months.

With regard to professional medical writers, GPP2 acknowledges that "professional medical writers are not ghostwriters", mirroring the stance of other professional bodies like WAME and AAMC. However, this important statement only holds true if the professional medical writers follow best practice guidelines. This should include not only full acknowledgement of the support provided (both in terms of contributions and funding), but should also be demonstrated by how the medical writer works with the author group.

For example, the authors should be determining the content of all written materials drafted by the medical writer, starting from analysing the raw data to generating the first outline. Each and every further draft by a medical writer should also be critically reviewed by the authors, who must approve or reject each change and ensure that the final version is approved as an accurate representation of their thoughts and interpretation.

It is acceptable, however, for administrative tasks to be delegated to the medical writer, such as setting up the submission of the approved manuscript.

Key statement
A further key statement in GPP2 makes it clear that professional medical writers should be named as authors on the publication if they have met all of the ICMJE criteria for authorship. Although a large degree of press focus has been on ghost authorship issues, it is also clearly stated in GPP2 that guest authorship is also unacceptable. Guest authorship is the practice of adding an author to the byline who has not met all the required ICMJE criteria, a practice more often seen in academia as a favour to colleagues.

4. Attempted publication of all trials
Much recent discussion has focused on whether all relevant information on specific products has been made available. In the GPP2 checklist, the authors recommend "making public or publishing results, regardless of outcome".

Of course, we have often heard anecdotally that "journals are not interested in publishing negative trials", to which the journals have responded that they are "definitely interested in truly negative trials, just not inconclusive trials". However, with the evolution of clinical trial registries, legislation on primary trial results and the increasing presence of online journals such as PLoS One and Archives of Drug Information that will publish all technically sound manuscripts irrespective of outcome, there is no longer any excuse not to publish trial results.

The proportion of pre-specified outcomes that can be reported in a single paper is likely to vary on a trial-by-trial basis, but it seems reasonable to expect that all pre-specified (by the data analysis plan) outcomes should be reported in some form, and that 'salami publication', or unnecessarily splitting results across multiple publications, should be avoided.

The future
Guidelines are only as useful as they are current. As key legislation across Europe changes, so the guidelines will need to adapt to reflect this. The GPP2 guidelines state that "posting clinical trial results according to the US FDAA Act of 2007 and the Joint Position on the Disclosure of Clinical Trial Information, whether before or after submission to a peer reviewed journal, should not preclude consideration for publication". With the upcoming European Commission regulations on clinical trial result reporting, this statement will probably need a more international context in future iterations.

Evolution
Furthermore, the GPP2 guidelines are a series of recommendations that should be used in conjunction with other applicable standards, such as those of the ICMJE and the Consolidated Standards of Reporting Trials (CONSORT) Group. It is certainly the hope of this co-author that the GPP guidelines continue to evolve as the industry evolves, and that all relevant groups incorporate these guidelines into their daily practices.

The recommendations in GPP2 should not come as a complete surprise to professionals whose job it is to communicate company-sponsored research, instead they should be seen as a set of guidelines that formalise best practice in this area.

Rebuild trust
By reading, understanding and incorporating recommendations from GPP2 into their standard operating procedures, companies (both pharmaceutical/biotechnology and medical communications industries) will be able to be confident in their ethical practices and start to rebuild trust and credibility with the public. After all, increasing the transparency of medical research publications can only help physicians and patients receive the most appropriate, evidence-based treatment.

The Author
Dan Bridges is a member of ISMPP one of the co-authors of GPP2, and works for Excerpta Medica - a company that provides medical communication services to the pharmaceutical/biotechnology industry. The opinions expressed here are those of the author, and do not represent the views of the other GPP2 authors, his company, or ISMPP.

To comment on this article, email pme@pmlive.com

Related articles:
Common goals