Opening Pandora's box: the public declaration of competing interests.

To ensure the credibility of research findings and to maintain public trust in the research process, it is critical that research be conducted transparently. Consideration of how financial, institutional and personal interests influence how research is designed, conducted and published is an increasingly important part of this transparency. Conflict of interest exists when a researcher’s private interests (competing interests) diverge from his or her responsibilities to the conduct of a research study and the publication of the results.1 Unfortunately, these competing interests are often not recognized2 or declared3-4; the suspicion that there may be undeclared competing interests has the potential to undermine public confidence in medical science.

Open Medicine is delighted to publish a conflict of interest checklist by Rochon and colleagues5 that provides a structured approach to considering and documenting potential competing interests through the entire research process, from study inception to publication. Designed to be completed by each investigator, irrespective of his or her role in the study, the checklist also provides, for the first time, a single comprehensive document that stakeholders with a role in a project’s adjudication or evaluation—members of research ethics boards and funding bodies, journal editors and readers—can use to review and interpret researchers’ competing interests and respond if necessary. A critical benefit of the checklist is that it also provides a record that can be held in the public domain for scrutiny and evaluation over the course of a research project and for future reference.

We believe that the checklist has two key strengths: the process undertaken to develop it6 and its broad scope. The checklist was developed in three phases. First, the research team drafted initial checklist items on the basis of published literature on competing interests, and two drafts of the checklist were reviewed by 29 and 24 reviewers, respectively, drawn from 3 groups (18 experts, 12 team members and 4 research staff). In the second phase, a third draft was discussed by 28 participants in a day-long, face-to-face meeting. In the third, “consolidation,” phase, further drafts were pilot-tested for usability. An example document was completed using a hypothetical research project, and an explanatory document and interactive PDF version were developed and reviewed at a second face-to-face meeting. The research team’s receptivity to broad user input demonstrates a commendable commitment to genuine knowledge translation; they have captured a wealth of knowledge and are disseminating it in the form of a practical product.5

The scope of the checklist is also critical. The checklist contains four sections: administrative information, study information, personal financial information and authorship and contributorship information, covering processes from study inception to completion. Given the time frame that it covers and the breadth of information that it prompts users to collect, the checklist has value beyond that provided by the recently released International Committee of Medical Journal Editors (ICMJE) Uniform Disclosure Form for Potential Conflicts of Interest,7 which is designed to be completed by authors and submitted to journal editors at the time they submit their work for publication. The ICMJE form represents a final summary of an author’s relevant financial competing interests for the 36 months preceding submission of his or her work for publication. The ICMJE form is also more prescriptive than the checklist in identifying what constitutes a financial competing interest. Additionally, because the ICMJE statement is completed retrospectively, it does not offer a mechanism to identify competing interests as they arise and hence to manage them while the project is underway.

As requirements and strategies to report competing interests become increasingly important to the transparency of scientific research, there is a risk that researchers, as well as those who evaluate competing interests, will be satisfied with the simple fact that the conflicts have been reported, rather than seizing the opportunity as a community to better anticipate and manage conflicts of interest. The fact that Rochon and colleagues’ checklist is designed to be filled out throughout the lifetime of a study ensures that researchers will consider from the outset whether and how potential competing interests might influence such issues as study design and the choice of collaborators. This provides an opportunity for all stakeholders to identify potential biases and manage competing interests proactively, rather than discovering at the publication stage that there are irreconcilable conflicts. In this sense it also serves as an educational tool on the conduct of research and brings the attention of investigators to the kinds of information that should be disclosed so that stakeholders can interpret a competing interest.

The availability of a single comprehensive checklist that can be used by the different groups involved in evaluating a research project provides an opportunity for the research team to save funding and administrative resources by eliminating the need to repeatedly complete different forms collecting similar information. Completion of the checklist may also flag irreconcilable conflicts, thereby allowing stakeholders to avoid investing in unpublishable studies. Importantly, it also helps to ensure that patients participate only in clinical research that has the potential to produce outcomes that can be published and thereafter applied in practice.

We encourage readers to review both the ICMJE form and Rochon and colleagues’ checklist and provide feedback to their creators to maximize the utility of these resources to these research community and the broader public. To this end, Open Medicine’s editors encourage users of the checklist to complete its associated usability survey, available at www.openmedicine.ca/fcoichecklist/; the authors are seeking feedback on the clarity and ease of use of the checklist as well as on its comprehensiveness and relevance. Although the checklist was designed for clinical trials, it could clearly be used more broadly, such as for guideline panels and research projects employing other study designs. We also invite you to consider this in your responses. Rochon and colleagues are strongly committed to developing the best possible tool for identifying and managing competing interests; we urge you to be part of this exciting evolution.

As further iterations of this tool are developed, the editors encourage the authors to consider two ways to strengthen the checklist: by the inclusion of non-financial competing interests, and by giving attention to the issue of contributorship as well as authorship. Both of these features are included in the ICMJE disclosure form. Including contributors in the checklist would ensure that those who make substantial contributions to a manuscript but do not qualify for authorship according to the ICMJE criteria (e.g., professional medical writers) are subject to the same disclosure practices. Given the ongoing lack of agreement about what does or should constitute authorship,8 lack of awareness of the ICMJE authorship criteria among researchers, and problems of guest and ghost authorship,9 it may be useful to require that all contributors to a study declare their competing interests, rather than arbitrarily limiting this requirement to the subset of contributors who meet the narrower definition of “author.”

A final issue to be considered by authors, editors and readers is the location where completed and in-progress checklists should reside. We believe the checklists should be available in the public domain. If research is genuinely free from untenable competing interests, then there is no need to hide this information. We recognize that no research is likely to be entirely conflict free; publication of the checklist provides an opportunity to broaden discourse rather than point the finger. Under the current system, journal editors and members of funding bodies and research ethics boards have the onus of responsibility for determining whether untenable competing interests exist. The availability of publicly archived checklists would allow readers and other stakeholders to determine for themselves how much credence should be given to the results of a particular research project. It would also ensure that media representatives are able to access information about competing interests; ideally, they would incorporate this information into their reports about biomedical studies.

How could checklists be published in the public domain? One option would be publication on a publicly accessible section of the website of the funding body or the author’s institution. Alternatively, for clinical trials requiring registration, the checklist could be housed at the same registration point as the other information on the trial. Authorized people could update the checklist as the study progressed. Journal editors publishing any papers arising from a research project could also publish the version of the checklist updated at the time of publication, along with a link to the primary archive.

It is vital that we in the research community manage competing interests to ensure the integrity of research; building public reporting of competing interests into the research process is critical to the principles of transparency and accountability. Integrity is in the eye of the beholder, imperfect though that eye might be (even if belongs to a journal editor); limiting access to information about competing interests unnecessarily stymies others’ efforts to evaluate them and makes it even harder to evaluate the integrity of a research project. With careful attention to competing interests along the full length of the research path, their declaration at publication need not open a Pandora’s box of concerns.