Conflict of interest

Conflicts of interest

One of the core issues in research integrity is conflicts of interest. Indeed, it would not be an exaggeration to say that there would be very few issues in research integrity if there were no conflicts of interest. Why are CoI so integral to integrity? Because integrity is about doing what is right despite any temptation to do what is personally advantageous. Cicero was perhaps the first to identity this key conflict between what he called “the honourable and the useful”. If you are a junior researcher and your group leader wants his name to be included on anything you write even if she did not contribute, this might seem like an authorship rather than a CoI issue. But if it was not a CoI issue all junior researchers would simply say “no way” in response. In fact, they do not (tend to do so), because they realise that their jobs and careers depend to a large extent on keeping their bosses happy. Thus the researcher’s interest in respecting the rules of authorship (the honourable) is in conflict with his interest in remaining employed (the useful). Similar conflicts come into play throughout the research environment.

In a conflict of interest, a person has two competing interests that could bias their judgment or behaviour. In this section we explore three major and one minor type of conflicts of interest (financial;, personal, intellectual and medical), and clarify what is meant by perceived and potential conflicts of interest. (It is actually quite difficult to classify the above example as it involves personal conflict, but is really in essence a financial conflict of interest - though the researcher might really love his job and want to keep doing science, the real concern in most cases will be ensuring income for oneself and one’s family. As such, capitalism is perhaps the root cause of most conflicts of interest and thus of most research misconduct.)

Financial conflicts of interest

In medical research, the focus until very recently has been on financial conflicts of interest. Here, the concern is not so much loss of financing as in the example above, but the lucrative incentives that doctors can attract for involvement in research, and which can influence their decisions and judgment. For example, a doctor might be paid by a pharmaceutical company to do some research and write a paper for a journal. He might think that he remains entirely objective, but the evidence shows that authors paid by industry are more likely to give positive evaluations of drugs being tested. Because of this, all financial conflicts of interest of all authors must normally be disclosed when a paper is submitted to a journal. This applies not only to current COI, but also to any over the last few years (specific duration is journal-dependent).

Having a COI is not necessarily a breach of research integrity, as long as that COI is declared. Disclosing it allows readers to consider whether the payment or other financial interest might have biased the reporting of the study (or its design or analysis, or even the decision to conduct the study). But failure to disclose a COI is a serious breach of research integrity because it represents deception that threatens transparency and robs readers of important information about how to interpret research results. Notably, failure to disclose COI is not classified as misconduct by the definitions in use in the United States, which define the ‘big three’ of plagiarism, falsification and fabrication as misconduct and failure to disclose COI as a “detrimental research practice” (see linked section for critique of this term).

Personal conflicts of interest

Personal conflicts of interest are much less heavily regulated than financial CoI, yet they are perhaps even more important. Here, the concern is not being biased by financial interests, but the possibility that personal connections (whether positive or negative) could threaten the objectivity of research decisions. For instance, many journals state that authors should not nominate reviewers with whom they have recently collaborated or who work in the same department. This is because collaborators are more likely to give favourable reviews to their colleagues. Even if journals operate a blind review system (see linked section), collaborators are likely to know or guess the authors of a given paper - and unscrupulous authors can even tip off colleagues that they will be nominated as reviewers. There have been several cases of so-called ‘review rings’ where researchers agree to review for each other and give positive evaluations. In addition, journals allow authors to indicate non-preferred reviewers, in order to avoid referees known to have particular COIs against particular authors. While the checks implemented by journals are good in principle, journals lack the resources to police reviewer nominations with any rigour so the system relies on author honesty.

But personal COI extends beyond potentially biased reviewer nominations, as personal connections operate on a variety of levels. Even if they have never worked together, researchers can meet and socialise, generating biases. Opponents can clash at conferences, generating animosity that last for years. Personal COIs do not need to be declared, but both positive and negative ones should be actively avoided when nominating reviewers or referees anywhere in the research process and when making hiring decisions.

Intellectual conflicts of interest

Intellectual conflicts of interest are even more ephemeral than personal ones, but they do exist.  Researchers who have worked for several years on one topic are likely to believe in a particular truth or paradigm, and become biased against any alternative explanation. For example, take the topic of breast cancer screening. Some research groups only ever publish papers in favour of such screening; others publish only papers that are against screening. It is important to note the connection between intellectual and personal conflicts of interest. If someone has a different intellectual view, one can become personally biased against that researcher in ways that bias decisions and interactions. Another type of intellectual COI concerns affiliation. If a researcher is writing about breast cancer screening and is a member of a Breast Cancer Screening Charity, this affiliation might be seen as intellectually biasing and should be declared.

Non-affiliation-related Intellectual conflicts of interest are insidious and hard to detect or declare. One way to both try to avoid them and to warn readers about them is to practice self-reflection. For example, in writing a new paper about a familiar topic, I might strive to maintain an open mind and not take any conclusions for granted. When submitting the new paper, I should declare that “I have written papers critical of X in the past” to inform readers about my past views on the subject. But even if researchers are proactive and engage in reflection and declaration of intellectual COIs, journals often remove all conflicts that are not connected with financing and affiliation.

Medical conflicts of interest

There are also medical conflicts of interest (ref Shaw). These are a subset of intellectual (and sometimes personal) conflicts of interest and concern how one’s own medical experience (including those involving family members) can affect one’s research in medicine or other health-related topics. In one case, a principal investigator’s running of a cancer trial was called into question when his child developed the same condition. This personal factor coloured his views on the disease and his treatment of participants. In this case, he should have stepped aside and asked someone else to run the trial - but he did not and this led to difficulties.

In other cases one’s own intellectual views on a particular medical practice can be affected by one’s personal experience. For example, if someone conducts research on the social implications of deafness, it might be prudent to disclose whether he or she is actually deaf. If someone is writing about obesity and smoking, one’s views could be dependent on whether one is an obese smoker. These factors should generally be declared, but journals are likely to remove them.

Perceived and potential conflicts of interest

Finally, much of the literature about conflicts of interest involves discussion of “perceived” and/or “potential” COI. What is the difference between a perceived COI, a potential COI, and even a perceived potential COI? A perceived COI is in the eye of the beholder and might not be a real COI. For example, if a researcher based in a city that happens to host several pharmaceutical companies writes an article critical of alternative medicine, some supporters of alternative medicine might perceive a conflicts of interest where none exists. All perceived COIs are perceived as potential COIs, but many will not nee potential COIs. A potential COI is one that should be declared even if it does not bias the researcher. For example, there might be some very objective researcher somewhere who reports results accurately even if paid handsomely by a Pharma company. Potential COIs do not really need to be distinguished from COIs because both need to be declared

bnn

1. International Ethical Guidelines for Health-related Research Involving Humans, Fourth Edition. Geneva. Council for International Organizations of Medical Sciences (CIOMS); 2016, Guideline 25. https://cioms.ch/wp-content/uploads/2017/01/WEB-CIOMS-EthicalGuidelines.pdf.

2. The European Code of Conduct for Research Integrity. https://ec.europa.eu/research/participants/data/ref/h2020/other/hi/h2020-ethics_code-of-conduct_en.pdf.

3. Council of Europe Additional Protocol to the Convention on Human Rights and Biomedicine on Biomedical Research.

4. Explanatory Report to Additional Protocol to the Convention on Human Rights and Biomedicine on Biomedical research. https://rm.coe.int/16800d3810.

5. Declaration of Helsinki (2013). https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-pr....

  1. Research Ethics Program, UC San Diego. http://research-ethics.org/topics/overview/.
  2. The Norwegian National Research Ethics Committees (etikkom.no).  https://www.etikkom.no/en/ethical-guidelines-for-research/guidelines-for-research-ethics-in-science-and-technology/commissioned-research-openness-and-conflicts-of-interest.
  3. The Office of Research Integrity. Conflicts of Interest and Commitment. https://ori.hhs.gov/conflicts-interest-and-commitment.
  4. ENERI E-Manual. Conflicts of Interests.
  5. Emanuel E.J, Thompson D.F. The concept of conflict of interest// Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008.
  6. Shawwa K, Kallas R, Koujanian S, Agarwal A, Neumann I, Alexander P, et al. (2016) Requirements of Clinical Journals for Authors’ Disclosure of Financial and Non-Financial Conflicts of Interest: A Cross Sectional Study. PLoS ONE 11(3): e0152301.
  7. L.A. Bero, Q. Grundy. Why Having a (Nonfinancial) Interest Is Not a Conflict of Interest. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5176169/pdf/pbio.2001221.pdf.  
  8. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0152301...
  9. Abby R. Rosenberg. „Get the Consent”—Nonfinancial Conflict of Interest in Academic Clinical Research https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5455692/.

Learning objectives

  • to identify the types of the conflicts of interest in biomedical research
  • to identify safeguards against CoIs
  • to analyze ethical and integrity implications of the CoI
  • to explore these issues in the light of cases, and propose the ways to prevent CoIs and solve them

Introduction

The problem of the conflicts of interest has become one of the major issues in research integrity and research ethics. Why do conflicts of interest matter both in terms of integrity and ethics?

First of all, the presence of CoI can violate research integrity. For example, it can compromise the reliability of research results, because of possible bias in different stages of research – planning, choosing methodology, recruiting research participants, analyzing data, (not) publishing the results). This may have more general consequence, such as a decrease of public trust in research (researchers/research institutions). Secondly, CoI can also violate principles of research ethics, for example, having an impact on safety of research participants (including future patients). The evidence reveals that strong financial motivation might be associated with less rigorous reporting of negative research results what can have consequences for both actual research participants and future patients (e.g. Vioxx case).[1]-[2] It can also have negative effects on informed consent of research participants, as researchers who are e.g. financially motivated to recruit big number of patients in a short period of time may (subconsciously) make pressure for their participation, present information about the proposed research in a biased way, etc.

Thus a conflict of interest is a set of circumstances or conditions in which professional judgement of a primary interest, such as the integrity and quality of the research, tends to be unduly influenced by a secondary interest, such as personal financial gain.[3]

The primary interest of any professional is to develop and promote the ultimate goals of their profession. For example, following the International Code of Medical Ethics, one of the main aims and obligations of medical doctors is to “be dedicated to providing competent medical service in full professional and moral independence, with compassion and respect for human dignity”. [4] The main obligation of biomedical researchers is to conduct research generating new scientific knowledge and to contribute to health and wellbeing of society. While medical doctors are used to act as researchers in clinical research, they are automatically faced by the necessity to manage at least two competing interests – duty to care (as a primary duty) vs responsibility to seek for new knowledge[5] [6], e.g. in case where a treating physician involves her patient in a double-blind randomized early phase clinical trial. Thus any other interests either they are financial, personal or professional constitute the secondary interest.

Most of the debate and regulations have been focused on financial CoI for at least two major reasons – firstly, significant commercialization of biomedical research and the growth of investments into biomedical research from the industry since the end of the last century; and secondly, the fact that financial gain creates major tensions, and it is relatively more tangible, objective and measurable if compared to other interests, such as e. g.  personal beliefs.

Financial CoI

Stems from either direct payments from the sponsor of research study, from, holding stocks in the sponsoring company, receiving financial remuneration for services (e.g. for consultations), having any other financial relationships with the producer of investigational product. Many organizations have adopted guidelines on financial CoIs. Some of them propose specific amount of money to serve as threshold for disclosure, e.g. Public Health Service[7] indicate 5000 USD received in 12 months period, the American Association of Medical Councils – up to 10000 USD, while the American Society of Oncology  claims that there can be no de minimis sum.[8]

Non-financial CoIs

Non-financial CoIs can be roughly categorized into three major groups: personal, intellectual, and medical.[9] Personal, religious, or political beliefs, personal experiences, academic competition or rivalry, career advancement or promotion, glory seeking or desire for fame, personal experience with subject of the research, personal relationship with someone who has the disease or condition under study, could serve as typical examples of interests in biomedical research.[10]

One of the recent studies on the requirements of clinical journals for authors’ disclosure of conflicts of interest conclude that “While financial COI disclosure was well defined by the majority of the journals, many did not have clear policies on disclosure of non-financial COI, disclosure of financial COI of family members and institutions of the authors, and effect of disclosed COI or non-disclosure of COI on editorial policies” – conclude the authors of one recent study on requirements of clinical journals for authors’ disclosure of Conflicts of Interest. The study also provides with a list of how non-financial CoIs are currently framed in different scientific journals.[11]

 

Patient’s consent for treatment or research is normally considered to be a process enabling her to make an informed, autonomous and free decision, consequently, the primary goal and the role of health care provider is to help her patient to do so. However, a doctor or researcher can be pressured by institution or principle investigator to get consent as soon as possible, without putting any value into the quality of informed consent process as such. Thus doctor’s/researcher’s primary obligation to respect patients’ autonomy (to provide information to patients in a non-directive manner and to give them time for consideration)  might be compromised by her secondary personal interest to keep the job, to please principal investigator, etc.[12]

CoI on the part of research institutions and RECs

The problem of CoI is relevant not only to individual researchers, but also to research institutions and bodies which evaluate acceptability of research project/program, such as RECs (IRBs). CoIs on the part of RECs members can arise either from their relationship with those whose applications they review (e.g. colleagues or competitors in a professional field), or financial motivation. As for research institutions - their primary goal to manage and oversee research activities might be tempted by the secondary interest, such as financial gain, institutional fame. Institutional CoIs are more complicated to handle as there is no external bodies to which institutions could report or disclose. There is still an ongoing discussion on this issue.

Safeguards

Being in a situation of a COI is not a breach of research ethics or integrity, however it prompts that there is a (hypothetically) higher potential for inappropriate or unethical behavior. The greater the incentives – the greater the likelihood that they would influence the autonomy of researcher’s professional judgement.

As it is evident that human life, both private and professional, is governed by certain kind of interests or motives, and it cannot be interest-free by definition, measures to avoid or minimise/mitigate the risks of CoI in research should be taken. Three major groups of actions how to deal with the CoI can be distinguished: disclosure, management, and prohibition[13].

1) Conflict of interest can be disclosed to:

  • Institution
  • REC/IRB
  • Journal editors/readers
  • Research participants

There is a widespread agreement on the necessity to disclose CoI to the first three groups, while disclosure to research participants remains debatable due to uncertainly of the implications of such an information for lay persons. It is argued that information about financial aspects of research can cause unnecessary anxiety and ungrounded mistrust in research participants.

Another issue related to disclosure of CoI in the process of publication – which information should be disclosure to the editors and which should be made public to the readers? Some information could be related to especially sensitive personal data, e.g. personal relationship with someone who has the disease or condition under study, or personally having certain disease or condition (e.g., obesity, smoking, cancer). It seems questionable if and how this information should be revealed to the third parties such as editors or readers of a journal.

2) Management. In certain situations where the participation of a staff member with CoI is necessary (specific skills, etc.) several measures to mitigate the risk of conflicting interests can be taken, e.g. a staff member without CoI can replace him/her for certain types of action e.g. obtaining consent. Creation of supervising independent ethics or data safety management boards as well as seeking external audit also can serve as measures to manage CoIs.

3) Prohibition. Certain types of relationships between research staff and sponsoring bodies can be even prohibited by national or institutional guidelines. E.g. there is an international consensus that an employee of pharmaceutical company cannot act as an investigator in any clinical drug trial, also investigators are prohibited from serving as paid consultants to the companies during the time of research study.[14]

 

[1] T. Lemmens. Conflict of Interest in Medical research: historical developments // Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008.

[2] What have we learnt from Vioxx? Harlan M Krumholz, Harold H Hines Junior professor of medicine and epidemiology and public health, Joseph S Ross, instructor, Amos H Presler, research associate, David S Egilman. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1779871/

[3] Emanuel E.J, Thompson D.F. The concept of conflict of interest//// Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008.

[5] Research Guidelines. Canadian Medical Association, 2008. https://www.cma.ca/Assets/assets-library/document/en/Final_Toolkit_Research-e.pdf.

[6] Guidelines on the practice of ethics committees in medical research with human participants. Royal Colledge of Physicians, 2007, p. 3.58. https://cdn.shopify.com/s/files/1/0924/4392/files/guidelines-practice-et...

[7] "Any remuneration received from the entity in the twelve months preceding the disclosure and the value of any equity interest in the entity as of the date of disclosure, when aggregated, exceeds $5,000." http://research-ethics.org/topics/conflicts-of-interest/#regulations-and-guidelines.

[8]Ezekiel J. Emanuel,  Dennis T. Thompson. The concept of conflict of interest // Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008. p. 764.

[9] E-Manual. Conflict of Interest.

[10]L.A. Bero, Q. Grundy. Why Having a (Nonfinancial) Interest Is Not a Conflict of Interest. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5176169/pdf/pbio.2001221.pdf.

[11] Shawwa K, Kallas R, Koujanian S, Agarwal A, Neumann I, Alexander P, et al. (2016) Requirements of Clinical Journals for Authors’ Disclosure of Financial and Non-Financial Conflicts of Interest: A Cross Sectional Study. PLoS ONE 11(3): e0152301. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0152301#pone-0152301-t004

[12] Abby R. Rosenberg. „Get the Consent”—Nonfinancial Conflict of Interest in Academic Clinical Research https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5455692/

[13]  Ezekiel J. Emanuel, Dennis T. Thompson. The concept of conflict of interest // Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008. p. 764-65.

[14]  Ezekiel J. Emanuel, Dennis T. Thompson. The concept of conflict of interest // Ezekiel J. Emanuel, Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, David D. Wendler. The Oxford Textbook of Clinical Research Ethics. Oxford University Press, 2008. p. 765.

Conflict of interest

 
1 Start 2 Step 2 3 Complete

The Parliamentary Environment Committee has invited university researcher A to hearing about an invention that would allow condensation water from nuclear power plants to be used as warm water in households and for recreational outdoor pools. Also ministry officials preparing the matter are consulted.

After the committee hearing, researcher A writes a blog post on the university’s website in which she takes a strong stance on the matter. Researcher A also attaches the statement that she has prepared for the Parliamentary Environment Committee hearing.

The blog post is followed by a media incident. As a result, researcher A appears on the morning television as the CEO of her company, telling about the research-based fundamental innovation that enables condensation water from power plants to be utilized in new ways.

Agitated by researcher A’s media appearances and blog posts, researcher B strongly attacks researcher A on his own Facebook page. According to him, researcher A has misinterpreted and misrepresented research results. The fuzz in the media grows even louder.

Researcher B wants to find out whether researcher A has misled the scientific community, the public, and the decision-makers by publicizing misleading or distorted information about research results and overdone the scientific significance of the results and their applicability.

- Which types of conflicts of interest do you identify in this case? Do the conflicts of interest bear integrity-related implications?