Conflicts of Interest in Research: What Should You Declare and When?

What counts as a conflict of interest in research?

You receive consulting fees from a company whose product appears in your study. Your laboratory holds a patent related to the technology you are evaluating. A close collaborator submits a manuscript that you are invited to review. Or perhaps you have a strong professional stake in one side of a scientific dispute.

Do all of these situations count as conflicts of interest? More importantly, which ones must you declare?

The difficulty is that researchers sometimes treat “conflict of interest” as shorthand for financial corruption or scientific wrongdoing. It is broader than that. A conflict can arise whenever a secondary interest or relationship has the potential to compete with your primary responsibility to conduct, evaluate, or communicate research impartially. Whether a particular interest must be disclosed, however, depends partly on the policy governing your role, institution, funder, ethics review, or publication.

Declare relevant interests rather than deciding privately that they do not matter

Conflict-of-interest disclosure is about transparency, not confession

A conflict exists because interests can compete

Researchers have many legitimate interests. They want their projects to succeed, their careers to progress, their collaborators to do well, their inventions to have an impact, and sometimes their investments or consulting work to generate income. Having such interests is not inherently improper.

The concern arises when a secondary interest could interfere with a primary responsibility. In research, that primary responsibility may involve the integrity of study design, participant welfare, objective analysis, accurate reporting, impartial peer review, or another professional duty.

The International Committee of Medical Journal Editors (ICMJE), for example, describes the potential for conflict and bias as arising when professional judgment concerning a primary interest, such as the validity of research, may be influenced by a secondary interest, such as financial gain. Its recommendations also emphasize that perceptions matter because readers need enough information to judge the relevance of an author's relationships and activities for themselves.

This distinction matters. Disclosure is not necessarily an admission that your judgment has actually been compromised. Nor does declaring an interest automatically establish that the interest has produced bias.

Financial interests are important, but they are not the whole story

Financial conflicts are often the easiest to recognize. Depending on the applicable policy, relevant interests may include consulting payments, employment, honoraria, equity ownership, stock options, intellectual property rights, royalties, paid expert testimony, or financial relationships with organizations that may benefit from the research.

Funding itself also requires transparent reporting. In publication, for example, the ICMJE recommends reporting sources of support and explaining the sponsor's role, if any, in study design, data collection, analysis and interpretation, manuscript preparation, and restrictions on publication. The circumstances under which research funding creates a conflict of interest therefore require more than simply asking whether money changed hands.

Conflicts can also be nonfinancial. ICMJE identifies personal relationships or rivalries, academic competition, and intellectual beliefs as examples of interests that may represent or be perceived as conflicts. Depending on the context, close professional relationships, institutional responsibilities, strong personal commitments, or other interests may also be relevant.

Exactly where the line falls is not universal. A journal's disclosure policy may differ from an institution's financial-conflict policy, while a funder may impose specific definitions, thresholds, time periods, and reporting obligations. The distinction between financial and nonfinancial conflicts is therefore useful, but neither category should be treated as a universal checklist that applies identically everywhere.

Actual, potential, and perceived conflicts can all matter

A researcher may be tempted to disclose only an interest that has demonstrably affected a decision. That is too narrow for many research settings.

An actual conflict exists when competing interests are presently interfering with a responsibility. A potential conflict concerns circumstances in which an interest could create a conflict. An apparent or perceived conflict concerns circumstances that could reasonably be seen as compromising independence even if the researcher believes no improper influence has occurred.

Terminology varies among organizations, and not every policy uses these categories in exactly the same way. The practical point is more stable: you should not rely solely on your own confidence that you can remain objective. WHO, for example, requires experts serving in advisory roles to disclose circumstances that could give rise to actual or ostensible conflicts, while ICMJE emphasizes transparent disclosure of relationships and activities that might bias or be seen to bias scholarly work.

What you must disclose depends on the rules that apply to you

There is no single worldwide disclosure form or threshold covering every researcher, discipline, institution, funding agency, and journal. You may simultaneously be subject to several policies.

Some policies use precise financial thresholds

Broad ethical advice should not be confused with a specific regulatory rule. For example, the U.S. Public Health Service financial conflict-of-interest regulations establish detailed requirements for research funded through covered PHS grants and cooperative agreements. Under those rules, investigators disclose specified significant financial interests to their institutions, and the institution determines whether a disclosed interest is related to the funded research and constitutes a financial conflict of interest.

The regulations include monetary thresholds and other definitions that apply within that regulatory framework. They also require covered investigators to make certain disclosures no later than the time of application, update them at least annually during an award, and disclose newly discovered or acquired significant financial interests within 30 days.

Those numbers should not be generalized into a universal research rule. Your institution may use broader standards, another funder may impose different requirements, and a journal may ask you to disclose relationships that do not meet a regulatory financial threshold.

Disclosure and management are separate decisions

One of the most useful habits is to separate two questions: “Should I disclose this?” and “What should happen because of it?”

You normally provide the relevant information first. The appropriate institution, editor, review committee, funder, or other responsible body then determines whether the relationship actually constitutes a conflict under its policy and what response is appropriate. The U.S. Office of Research Integrity similarly notes that final decisions about managing conflicts should be made by the responsible administrators or committees rather than by the researcher alone.

A disclosed conflict does not automatically require abandoning a study, ending a collaboration, or rejecting a manuscript. Depending on its seriousness and the applicable rules, the response may range from disclosure alone to independent oversight, changes in responsibilities, recusal, modification of the research arrangement, or elimination of the conflicting interest. Whether a conflict should be managed rather than eliminated is a separate decision from whether it should be declared.

Disclosure should be treated as an ongoing responsibility

Conflicts do not freeze on the day a project begins. A researcher may acquire stock, enter a consultancy, develop intellectual property, form a new professional relationship, take an institutional role, or encounter another change while research is underway.

That means a declaration made two years ago cannot automatically be assumed to remain complete today. Some systems impose explicit periodic and event-triggered updates. Even where a particular rule does not specify the same timetable, researchers should follow the applicable policy and report newly relevant circumstances when required rather than waiting until publication.

What disclosure looks like across a research project

The important point is not that every consultancy necessarily compromises research. It is that researchers should not silently make the final judgment about the significance of their own competing interests when an applicable disclosure process requires those interests to be reported.

Common mistakes when deciding whether to declare a conflict

Use disclosure as a process, not a one-time statement

The safest practical approach is not to memorize a universal list of conflicts. There isn't one. Instead, identify the roles you occupy and the policies attached to them.

Keep records of what you disclosed, to whom, and when. This becomes particularly useful when a project involves several institutions, funders, collaborators, or journals with different requirements. It also reduces the rather academic problem of discovering, at manuscript submission, that nobody remembers what was declared three years earlier.

What should you check before declaring conflicts of interest?

Frequently asked questions about research conflicts of interest

When in doubt, let the appropriate process evaluate the interest

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.