Research Design

Conflicts of Interest

By Dr Pedram Rashidi and Emma Cooney, University of Queensland Centre for Policy Futures   

Conflicts of interest exist in “a situation where an independent observer might reasonably conclude that the professional actions of a person are or may be unduly influenced by other interests.” [1] Perceptions of conflicts of interest are just as important as actual conflicts of interests. [2] 

---

Why is it relevant?

Avoiding and properly disclosing conflicts of interest is essential to maintaining the integrity, credibility, and reliability of research. Conflicts of interest can introduce bias into decision-making processes, potentially compromising research outcomes and, in serious cases, leading to misconduct or corruption. [3] Failing to properly manage actual and perceived conflicts of interest can erode public trust in science, generally, as well as in the individuals and organizations involved in a particular project. In this sense, proactive and responsible management of such conflicts is essential to preserving public trust. [1] 

In Australia, researchers are required to adhere to the Australian Code for the Responsible Conduct of Research, which explicitly mandates ‘transparency in declaring interests.’ [1] Similarly, the International Committee of Medical Journal Editors has characterised a purposeful failure to disclose a conflict of interest as a form of research misconduct. [2] Together, these standards highlight the critical importance of openness and accountability in responsible research practice. 

---

Practical Steps

1. Disclose to your institution all interests that are relevant, or could appear to be relevant, to proposed or ongoing research. This could include financial interests such as direct or indirect payments, company shares or options, royalties, directorships, some scholarships, and operational or infrastructure support. It also includes non-financial interests such as board memberships, personal or social relationships, and/or recent employment with organisations affiliated with industry groups that could stand to benefit from the research. [1]
2. If a conflict of interest exists, your institution will determine the appropriate measures to manage the conflict of interest. Examples of measures that might be taken include requiring the public disclosure of interests when publishing or presenting the research; involving another individual to oversee the research; and requiring the researcher to excuse themselves from any deliberate decision-making, play a reduced role in the research or relinquish financial or other interests. [1]
3. Refer to and comply with your institutional policies. Your institution has a responsibility to maintain policies and procedures for the disclosure and management of interests.  
4. Broader disclosure of conflicts of interest. Conflicts may also need to be disclosed in other instances, such as in funding applications and applications to ethics committees where the conflict impacts the work in question. 

---

Case Study

Case Study 1: A study by researchers at the University of Sydney, published in the Journal of General Internal Medicine, found that many Australian medical researchers failed to disclose payments from pharmaceutical companies. One in four researchers out of 120 trials had at least one undeclared conflict, averaging an undisclosed payment value of almost $9000. The disclosure of these payments is important because pharmaceutical industry funding is commonly associated with a bias towards favourable results for the tested product. [4]  

Case Study 2: In another case, Shahar and Sayers (2018) published an article on abnormal bone growths protruding from the base of younger peoples’ skulls, which the researchers claimed was due to bad posture and phone use. [5] The initial research generated great concern, until closer scrutiny revealed an undeclared conflict of interest by the first author – a chiropractic clinician who provides posture related advice and products. [6]  The paper was later edited to clarify these conflicts, and to redraft sections to emphasise that: genetic predisposition may be a factor; the data was taken retrospectively from a clinician’s database for individuals presenting only with mild symptomology; and “great care should be taken to avoid over generalising these results to an asymptomatic general population”. 

References

Back to menu