Volume 33, Issue 2 , Pages 81-86, February 2010
Conflict of Interest in Scientific Publications: A Historical Review and Update
Article Outline
- Abstract
- History of COI and the JMPT
- Defining and Recognizing COI
- ICMJE Uniform Requirements for Manuscripts COI Form
- COI and Health Care Professions
- Limitations of Declaring COI
- Conclusion
- Funding Sources and Potential Conflicts of Interest
- References
- Copyright
Abstract
There is a stigma surrounding the reporting of conflicts of interest (COIs) that should be overcome. This article provides a discussion about COI, a brief historical review of requirements, and an update of policy for the Journal of Manipulative and Physiological Therapeutics that includes the alignment of the International Committee of Medical Journal Editors new COI form requirement. This article reviews types of COI and suggests that professions (eg, chiropractic, physical therapy, acupuncture) not directly affiliated with pharmaceutical and device companies, though they may be faced with different circumstances, should still comply with current COI reporting standards. There should be no disgrace in properly declaring interests; instead, COI declaration should be considered an important part of publication and recognized as an honorable action. Declaration of COI helps to provide transparency and disclosure to all involved including editorial staff, peer reviewers, and the readers.
Conflict of interest (COI) policies continue to evolve as the relationships between researchers, their research, funding bodies, journal staff, and publishers become more complex. One may argue that the COI of those researchers who are involved with pharmaceutical or device companies are different from those doing research in manipulative therapies or other conservative care professions. However, transparency of affiliations and full disclosure should be offered for all professions no matter what the research focus.1, 2, 3, 4 The purpose of this article is to review historically the COI policy for the Journal of Manipulative and Physiological Therapeutics (JMPT), to discuss COI issues for professions not directly affiliated with pharmaceutical and device companies (eg, chiropractic, physical therapy, acupuncture), and to provide an update on how the JMPT is aligning its policy to current standards as set forth by the International Committee of Medical Journal Editors (ICMJE).5
History of COI and the JMPT
The JMPT published its first issue under the editorship of Roy W. Hildebrandt, DC, in 1978 (Fig 1). It should not be surprising that the first set of instructions for authors did not include a statement requiring authors to declare COI. The concept of declaring COI was not standard, no other biomedical journal included such a policy,1 and COI was just beginning to be used in codes of ethics for the medical profession in the 1970s.6 One of the first biomedical journals to implement a COI policy was the New England Journal of Medicine in 1984.1, 7, 8 However, it was not until later, in 1994, that the JMPT included the first mention of COI in the instructions for authors; an editorial discussed the topic 4 years earlier.9 The updated 1994 policy did not require that the authors declare COI but that they should be ready to answer any questions that the editor may have (Table 1).
Fig 1.
Cover of the first JMPT volume 1 number 1.
Table 1. Conflict of interest policies for the JMPT
| Year implemented | Policy | Importance |
|---|---|---|
| 1994 | “Conflict of interest—in recognition that it may at times be difficult to judge material from authors where proprietary interests are concerned, authors should be ready to answer requests from the editor regarding potential COIs. The editor makes the final determination concerning the extent of information released to the public.” | First mention of COI in the instructions for authors |
| 2005 | “Conflict of interest exists when an author has financial or personal interests that may influence his or her actions in regard to the authors' work, manuscript, or decisions. Conflicts of interest that exist, or that are perceived to exist, for individual authors in connection with the content of this paper should be disclosed to the JMPT in the cover letter to the editor and in the authorship form. In recognition that it may be difficult to judge material from authors where proprietary interests are concerned, authors should be ready to answer requests from the editor regarding potential COIs. The editor makes the final determination concerning the extent of information released to the public.” | First requirement for authors to declare COI at time of submission of manuscript |
| 2009 | Funding and COIs section included for each peer-reviewed paper. | First standard section in peer-reviewed manuscripts declaring COI |
| 2010 | Conflict of interest Authors—Each author is required to complete a COI form (created by the ICMJE) and submit this form at the time of initial submission. COI exists when an author has financial or other interests that may possibly influence his or her actions in regard to research efforts, manuscript development, or decisions. COI that exist, or that are perceived to exist, for individual authors in connection with the content of the paper shall be disclosed to the JMPT at the time of submission. In addition to the form, any concerns or additional COI issues that are not covered in the form should be included in the cover letter to the editor. Authors must also disclose to the editor in the cover letter the COIs of any other person or entity involved with the paper (eg, nonauthor, contributor, funding body). In recognition that it may be difficult to judge material from authors where COI are concerned, authors should be ready to answer requests from the editor regarding potential COI. Editorial decisions will not be based solely on declared COI. Manuscripts will go to blinded peer review and be judged on their quality and contribution to the literature. Should a manuscript get accepted for publication, the COI statement will be published with the final manuscript. The editor makes the final determination concerning the extent of information included in the published paper. It is expected that authors are truthful when declaring conflicts on their submission materials. An editor's role is not to be police officer or detective, so the burden is upon the author to properly declare COI. If an author did not accurately and completely declare COI upon submission, and this is discovered after publication, the editor will follow up with an ethics investigation. Fail to disclose COI may result in rejection or retraction of the paper, prevention of future submissions, and notification of ethical misconduct to the proper authorities. Editorial staff and peer reviewers—it is expected that those involved with handling manuscripts for the journal will properly disclose their financial and professional interests that may be viewed as potential COIs and recuse themselves from any actions in which their COIs would hamper their judgment or actions. Peer reviewers should inform the editor if they feel they are not able to properly peer review a manuscript and recuse themselves from reviewing that manuscript. Editorial staff should disclose information that readers may perceive might influence decisions in journal editing. Disclosure statements about potential COIs for the journal staff should be published regularly. Please refer to ICMJE Web site for more information on COI. | Alignment of journal COI policy with ICMJE standards First requirement of a COI form Inclusion of all people handling manuscripts in COI policy, including authors, peer reviewers, and editorial staff First publication of disclosure statements for journal editorial staff |
It was not until 2005 that the JMPT included a policy that clearly defined COI and required authors to declare their COI in a cover letter to the editor (Table 1). At that time, affiliations or declared COI would be printed in the author affiliations section and not in a separate section dedicated to COI. In 2009, a separate heading was included in peer-reviewed papers to state funding and declared conflicts for authors. This standardized section was included to help readers know where to find this information so that they may read the paper with a full understanding of authors' disclosures.
In 2010, the JMPT implemented the ICMJE's policy on COI, which provides a structured form in which each authors must declare their financial and professional interests. Although the updated ICMJE process may initially seem cumbersome, it helps to define COI, improve clarity, and better communicate this information with the editor and readers. The JMPT Editorial decisions are not based solely on declared COI. Manuscripts will go to blinded peer review and be judged on quality and contribution to the literature. Should a manuscript get accepted for publication, a COI statement will be published with the final manuscript.
Defining and Recognizing COI
According to the ICMJE, the body that has established the uniform requirements for manuscripts, “Conflict of interest exists when an author (or the author's institution), reviewer, or editor has financial or personal relationships that inappropriately influence (bias) his or her actions (such relationships are also known as dual commitments, competing interests, or competing loyalties). These relationships vary from negligible to great potential for influencing judgment. Not all relationships represent true conflict of interest” (http://www.icmje.org/ethical_4conflicts.html).
The World Association of Medical Editors outlines several types of COI (http://www.wame.org/conflict-of-interest-in-peer-reviewed-medical-journals). These include the following.
Financial Ties
A financial tie is included if an author has or will receive money, equipment, or anything of value (eg, salary, royalties, patents, gifts, stocks, honoraria, reimbursement for travel or meeting registration) related to the study or study topic. Money received directly or indirectly is included in this category. An author may own a patent, stock options, or receive consultancy fees related directly to a product or company that was discussed in an article. Another relationship may be when a researcher's spouse or family member receives a salary from a company that makes the product under investigation or may have financial investments in that company. This information should also be declared. Information about type of practice is also included here. The World Association of Medical Editors COI document states, “Clinicians have a financial competing interest if they are paid for clinical services related to their research, for example, if they write, review, or edit an article about the comparative advantage of a procedure that they themselves provide for income.” This would suggest that an author who practices or teaches a particular technique should declare this as a potential COI.
Academic Commitments
Academic commitments are the beliefs held by the researcher about science, methods, and ideas. It is possible that 2 scientists may each use sound research methods but their beliefs and approaches may be very different. For example, one may be biased toward qualitative research and the other toward quantitative research. Because of these differences, they may have conflicts or bias against each other's work and therefore judge the other's work more harshly than it deserves. The opposite may be true in that 2 researchers of similar mindset may judge each other's work more favorably than it deserves and may miss flaws that should be mentioned during peer review.
Personal Relationships
Personal relationships include close relationships such as with family, friends, competitors, and peers as they relate to the article or research. Personal relationships may be biased either favorably (eg, friends) or unfavorably (eg, opponents) and should be declared.
Political or Religious Beliefs
These include any strong views or agendas that may interfere with sound scientific process. For example, an author who strongly believes in antiabortion or antiimmunization movements may be biased against research that support opposing views, regardless of the scientific quality of the manuscript.
Institutional Affiliations
Institutional affiliations include relationships between those who are affiliated with a company or institution that sells, markets, distributes, or benefits from a particular product. This may include direct relationships, such as a researcher who is employed by the company that makes the product. An indirect relationship may be when a researcher is studying a device made by a company that is contributing funding to his institution. Although the funding is not directly going to the author or the research project, the author may receive pressure from the administration to show positive findings in order for the institution to maintain continued funding. Therefore, this relationship should be declared in the COI statement.
Having or declaring COI does not imply wrongdoing; instead, full disclosure assists with honesty and transparency.1 Because someone has an interest in the work that they are involved in does not mean that they are being deceitful or unethical. The goal is to declare these potential conflicts within reason and not hide this information from the editorial staff or readership. For the JMPT, the goal of declaring COI is not to prevent anyone from publishing in the journal or contributing to the scholarly process but instead to be forthcoming.
ICMJE Uniform Requirements for Manuscripts COI Form
A new ICMJE form was developed and adopted by major biomedical journals as a standard for reporting COI (http://www.icmje.org/coi_disclosure.pdf). The ICMJE form asks the authors to declare both financial and nonfinancial information.10 The content includes the following:
At first glance, this process may seem burdensome and extensive. However, for those authors who are submitting to journals on a regular basis because many journals are requiring this form, filling out the form once and then updating when there is new information is all that is needed. Thus, this form will save time in the long run. Overall, the declaration is simple and depending upon how many entities the author is involved with, should only take a few minutes to complete. For many authors who are not involved with pharmaceutical and device companies, the form is very easy to complete. Once it is completed, the same form can be used for future submissions to other journals that require ICMJE COI declaration. The ICMJE also provides a sample form so that authors who are new to disclosing conflicts may see an example of how the form should be completed (http://www.icmje.org/sample_disclosure.pdf). As the ICMJE updates the form and procedures, the JMPT will follow these updates. Using the ICMJE COI form should help reduce confusion about what authors should declare and provide consistency between publications.
In addition to authors, it is expected that anyone making decisions about a manuscript, such as peer reviewers or those making accept/reject decisions, should be transparent and disclose information. Peer reviewers and editorial staff should consider potential COIs and disclose these when processing or handling manuscripts.5, 11
Regarding peer reviewers, it is expected that an editor will select fair and unbiased peer reviewers to review manuscripts and avoid those who may be biased strongly either for or against the manuscript due to potential COIs. Reviewers should be selected based upon their ability to critically appraise the information and their content area of expertise. It is also the responsibility of a peer reviewer to declare to the editor if they are strongly biased or are unable to provide fair peer review, and they should recuse themselves from the review.1, 5 Although it may be easy to suggest that an editor avoid selecting a biased peer reviewer or one who may be familiar with authors' work, it is sometimes impossible when the field is very narrow and the people involved are limited, such as in specialty areas or professions that have few numbers. If the field is small, there may be only a few who are capable of providing appropriate review. In these cases, the editor must do his/her best to select fair reviewers to judge the work.
COI and Health Care Professions
The medical equipment and pharmaceutical industries have a strong influence on the medical profession,6 and we have recently witnessed an increase in the examples of ethical misconduct.1, 12 A researcher involved with a pharmaceutical company may face issues regarding influence of the industry and large financial incentives. These COIs may be strong, but these types of financial conflicts seem to be more clearly recognizable than other types of COI. Professions that are not directly involved with pharmaceutical companies or device companies may have a different set of difficulties with recognizing and declaring COI.
Researchers studying complementary and alternative medicine or nonpharmaceutically based professions (eg, chiropractic, physical therapy) or those who are involved in researching noncommercially driven techniques (eg, techniques that do not require the sale of a product, such as use of mind body, exercise, manual therapies) may have fewer financial temptations because of fewer funding opportunities. However, this does not mean that COI is not present. Conflict of interest may arise between proponents of one technique or profession against another. For example, a researcher who is a manual therapist from one profession may perceive that his/her method is superior to another profession although certification and years of training may be the same. The researcher may bias his/her representation of the one technique over another or deliberately introduce bias into the study design. Various types of methodology or philosophical underpinnings may also create COIs. Research approaches used by vitalists are different than those used by mechanists.13, 14 Those who believe that the chiropractic profession should be focused only on treating spinal conditions may be biased against those who believe that the sole purpose of chiropractic is to find and fix “spinal subluxations” and remove “nerve interference” or those who have a broad-based, primary care view of chiropractic. It may be difficult for scholars from these diametrically opposed groups to recognize that their philosophies may create bias and therefore disclose their ideals as COI. The COI issues in these factions may be more difficult to recognize and challenging to address than simple financial conflicts.
Limitations of Declaring COI
It is important to state what a COI declaration does not do. No matter how many policies or good intentions a journal may have, authors may still neglect to declare their COI. This may be because they are unaware that a conflict exists or they may be concerned that declaration may taint their reputation. An editor's role is not to be a police officer or detective, so the burden is upon the authors to properly declare COI. It is expected that authors are truthful when declaring conflicts when submitting their materials. If authors do not accurately and completely declare their interests upon submission, and it is discovered later, the editor will follow up with an ethics investigation. The results may include rejection or retraction of the paper and notification of the proper authorities.
Some journals have a policy to reject certain papers in which authors have declared COI.1, 15 This may potentially cause problems by creating an environment in which authors may avoid declaring COI. Not all authors who are asked to declare COI are properly doing so.16 As well, some suggest that declaring COI does not prevent the interest from influencing the researcher and the resulting manuscript. It is suggested that declaring COI may have a negative effect and make the author feel free to do as they please and not properly represent the research findings.17 The COI declaration also does not prevent research fraud. These are issues that are not easily solved using COI.
Another question that should be raised about COI is “How far should we go in declaring interests?” Should one declare all items of value received from a company (eg, that give away pen that was picked up from a booth while attending the last conference) or should one only declare if the value is over a certain amount (eg, $100, $1000, or $10 000?). When choosing a particular cutoff amount for value to declare, does $9995 influence someone less than $10 000? Selecting a cutoff amount becomes a slippery slope. Should we list the lunch we ate at a scientific conference that was partially sponsored by a company on the COI form or only those private lunches at the hospital sponsored by pharmaceutical companies? How far back should we declare? Is a financial or professional relationship from 4 years ago less influential than one 3 years ago? How far out does one declare relationship influences? If one's daughter attends pharmacy college and she has received a scholarship from a pharmaceutical company, should this be declared? What about our second cousin twice removed? What happens with the possible COI that an author may not be able to recognize? Imagine what would happen if all potential or perceived COI were required to be published in a manuscript that has 5 authors and each author has a career length of a minimum of 10 years. The information would not only be cumbersome for readers to sift through, I suggest that it would actually create lack of transparency. Ultimately, COI is a challenging issue that we will do our best to address, but there is no perfect system for declaring all relevant COIs for everyone. Our next best step is to use the standard ICMJE form and evolve with the process as more information comes to light.
Conclusion
There is a stigma surrounding the reporting of COI that should be overcome. There is no disgrace in properly declaring financial or other interests related to one's work. Instead, COI declaration should be considered an important part of publication and recognized as an honorable action. Declaration of COI is not meant to punish or reprimand those involved but to provide transparency and disclosure to the readers. The JMPT continues to evolve with standard publication practices and is now requiring the ICMJE COI form for each other. We hope that this will help our authors more easily recognize and declare COI and help our readers better understand author declarations.
Funding Sources and Potential Conflicts of Interest
No funding was received for this article. Disclosure statement: Claire Johnson, DC, MSEd, is the editor of the JMPT; a full-time professor at the National University of Health Sciences; peer review chair for the Association of Chiropractic Colleges; a board member of NCMIC; and a member of the American Chiropractic Association, American Chiropractic Board of Sports Physicians, International Chiropractors Association, Association for the History of Chiropractic, Counsel of Science Editors, American Public Health Association, Committee on Publication Ethics, World Association of Medical Editors, American Medical Writers Association, and American Educational Research Association.
References
- . Ethical issues in biomedical publication. Baltimore (Md): Johns Hopkins University Press; 2000;
- . Conflict of interest policies among institutions and organizations offering chiropractic continuing education. J Manipulative Physiol Ther. 2009;32:303–308
- . Managing conflicts of interest in continuing medical education: a comparison of policies.. J Chiropr Educ. 2009;23:36–39
- . Transparency of research methods: proud to be a Naked Emperor. J Manipulative Physiol Ther. 2005;28:377–378
- Guidelines of the International Committee of Medical Journal Editors on editorial conflict of interest. http://www.icmje.org/#conflicts
- . Conflicts of interest: recognition, disclosure, and management. J Am Coll Surg. 2003;196:505–517
- . Financial conflicts of interest in biomedical research. N Engl J Med. 1993;329:570–571
- . Dealing with conflicts of interest. N Engl J Med. 1984;310:1182–1183
- . Questions of authorship and financial conflict of interest. J Manipulative Physiol Ther. 1990;13:61–62
- Uniform format for disclosure of competing interests in ICMJE journals. Lancet. 2009;374:1395–1396
- . Editors' declaration of their own conflicts of interest. CMAJ. 2004;171:475–476
- . Scientists behaving badly. Nature. 2005;435:737–738
- . The evolution and importance of spinal and chiropractic research. J Manipulative Physiol Ther. 1992;15:31–35
- . When worldviews collide: maintaining a vitalistic perspective in chiropractic in the postmodern era. J Chiropr Humanit. 2005;12:2–7
- . The Lancet's policy on conflicts of interest. Lancet. 2004;363:2–3
- . Accuracy of conflict-of-interest disclosures reported by physicians. N Engl J Med. 2009;361:1466–1474
- . The dirt on coming clean: perverse effects of disclosing conflicts of interest. J Legal Studies. 2005;34:1–25
PII: S0161-4754(10)00022-9
doi:10.1016/j.jmpt.2010.01.004
© 2010 National University Health Sciences. Published by Elsevier Inc. All rights reserved.
Volume 33, Issue 2 , Pages 81-86, February 2010