CASRAI Dictionary

Tag: conflict of interest

  • Conducting research-security risk assessments: proportionate due diligence

    Research-security policy can sound, to a working researcher, like a demand to treat every collaborator with suspicion and every international partnership as a threat. Understood properly, it is nearly the opposite. The aim of a research-security risk assessment is not to close doors but to open them with eyes open — to ask sensible, proportionate questions before entering a partnership, accepting funding or recruiting a colleague, so that genuine risks are identified and managed while the overwhelming majority of legitimate collaboration proceeds unimpeded. The skill lies in proportion: too little scrutiny leaves real risks unexamined, too much strangles the openness on which research thrives. This article looks at how institutions can build that proportionate assessment into their practice, through the research security domain of the CASRAI Dictionary.

    What a risk assessment is for

    A research-security risk assessment is a structured way of asking, before a commitment is made, whether a proposed partnership, funding source or appointment carries risks that need to be understood and managed. Those risks might include sensitive knowledge or technology being diverted to harmful ends, undisclosed conflicts of interest or commitment, or partnerships with entities whose affiliations warrant closer examination. The purpose is not to reach a verdict of “safe” or “dangerous” but to make an informed decision and proceed with confidence. Most assessments conclude that there is no significant concern, and that is a feature, not a failure — the process exists to distinguish the rare case that needs attention from the many that do not.

    Proportionality is the governing principle

    The single most important principle is proportionality. The depth of due diligence should match the level of risk, and most research carries little. A fundamental, openly published study with a long-standing academic partner needs nothing like the scrutiny appropriate to research in a sensitive area, involving technology with security implications, with a new and unfamiliar partner. Building proportionality into the process — through triage that applies light-touch checks to low-risk activity and reserves deeper examination for the genuinely sensitive — is what keeps research security workable. Without it, either everything receives burdensome scrutiny that researchers come to resent and evade, or nothing does. Proportionate assessment respects both the need for security and the value of open collaboration.

    UK Trusted Research guidance

    Institutions do not have to design this from nothing. In the United Kingdom, the Trusted Research guidance, developed by national security bodies including the agency now responsible for protective security advice and the National Cyber Security Centre, gives researchers and institutions practical help in identifying and managing the risks of international collaboration. Its framing is deliberately constructive: it is about helping researchers collaborate safely and protect their work, not about discouraging international partnership, which it explicitly recognises as essential to research. Trusted Research offers a model of how security guidance can support rather than obstruct, equipping researchers to ask the right questions and make sound judgements while keeping the door to collaboration open. It is a useful touchstone for what proportionate, supportive research security looks like in practice.

    Specific tools: ATAS and disclosure

    Within the wider landscape sit specific mechanisms that an assessment may engage. The Academic Technology Approval Scheme (ATAS) is a UK scheme requiring certain international students and researchers working in specified sensitive subject areas to obtain clearance before commencing their studies or work — a targeted control focused on areas where there is genuine proliferation concern, rather than a blanket restriction. Equally central is the disclosure of conflicts of interest and conflicts of commitment: requiring researchers to declare outside affiliations, funding, appointments and obligations, so that potential conflicts are visible and can be managed. Transparency through disclosure is one of the most effective and least intrusive security tools available, because it surfaces the information needed to assess risk without presuming bad faith. These mechanisms are pieces of a proportionate system, applied where relevant rather than universally.

    The wider policy context

    Research-security risk assessment also responds to expectations set by funders and governments. In the United States, the policy framework known as NSPM-33 set out requirements for research-security programmes at institutions receiving federal funding, including expectations around disclosure and the protection of research. The detail of that framework, and of related concerns such as dual-use research and the cybersecurity of sensitive work, are addressed in our existing coverage of those topics; the point here is that institutional risk assessment is the practical mechanism through which such expectations are met. A research-security programme is, in large part, the capacity to conduct proportionate due diligence consistently and well.

    Embedding assessment in research administration

    For risk assessment to work, it has to be part of the ordinary machinery of research administration rather than a special process invoked in alarm. Integrated into the points where commitments are made — partnership agreements, grant acceptance, recruitment — it becomes a routine, proportionate check rather than a disruptive intervention, and it draws on information institutions already gather. This integration is the concern of our research administration resources: building security-mindedness into normal processes so that it supports good decisions rather than obstructing them. Done well, assessment is largely invisible to the researcher whose work poses no concern, and genuinely helpful to the one whose work does.

    A consistent vocabulary for risk and disclosure

    For risk assessment to function across institutions, funders and partners, the information involved — disclosure categories, risk levels, affiliation and conflict information — must be described consistently, or an assessment in one context will be misread in another. That consistency is what the CASRAI Dictionary provides: a shared vocabulary so that the information underpinning research-security decisions is understood the same way wherever it is recorded. And because the people and contributions involved are part of the research record, they can be described in the same shared framework — the CRediT taxonomy and its full set of contribution roles. Research security, done proportionately, is not a barrier to collaboration but a way of protecting the collaboration that matters — asking the right questions so the lab door can stay open.

  • Conflict-of-interest disclosure: a practical guide for authors

    Conflict-of-interest disclosure is one of the most misunderstood obligations in research publishing. Authors often treat it as an accusation to be avoided — as if declaring an interest were an admission of bias. It is the opposite. Disclosure exists to protect the reader, and a declared interest is a sign of good faith, not of wrongdoing. This guide explains what counts as a competing interest, how the standard disclosure model works, and how this declaration fits alongside the other statements an author now makes. It connects to the author-side declarations described at the author statement and the parallel disclosure covered in the AI-use disclosure guidance.

    What a conflict of interest actually is

    A conflict of interest — many journals prefer the less loaded term competing interest — exists when a secondary interest could, in a reasonable observer’s view, improperly influence the conduct or reporting of research. The key word is could. A conflict is about the potential for influence, not proof that influence occurred. You do not need to believe your judgment was actually affected; you need only recognise that an outside reader, knowing the interest, might reasonably weigh your conclusions differently. That is why disclosure is the remedy: it hands the reader the information they need to make that judgment for themselves.

    Conflicts come in several forms, and all are disclosable:

    • Financial interests: research funding, consultancy or speaker fees, employment, stock or equity, patents and royalties, paid expert testimony, and gifts. Financial interests are the most scrutinised because they are the most measurable.
    • Personal and professional relationships: close personal ties, rivalries, or affiliations that could shape how the work is reported or reviewed.
    • Intellectual or academic commitments: strongly held prior positions, or a role advocating for a particular view, that a reader might reasonably want to know about.
    • Institutional interests: interests held by the author’s employer that bear on the work.

    The standard: the ICMJE disclosure model

    The dominant framework in biomedical and much of STEM publishing comes from the International Committee of Medical Journal Editors (ICMJE), whose recommendations include a standard disclosure approach now adopted very widely. Its design has two features worth understanding.

    First, it asks each author to disclose individually. A competing interest belongs to a person, so every named author completes their own declaration rather than the corresponding author guessing on the others’ behalf. Second, it asks about a defined recent window — the ICMJE form covers interests over the preceding 36 months — rather than leaving “relevant” to the author’s discretion. Bounding the question makes disclosure more consistent and harder to under-report by omission.

    The ICMJE model also asks specifically about the relationship between any interest and the submitted work: money paid to you or your institution, the role of any funder in the study’s design, conduct, or reporting, and so on. The point is to surface not just that an interest exists but how it connects to this particular paper.

    How to approach your own disclosure

    1. Disclose when in doubt. The cost of declaring an interest that turns out to be immaterial is essentially nil; the cost of an undisclosed interest that later surfaces is serious — corrections, expressions of concern, and reputational damage. Asymmetry of risk argues for over-disclosure.
    2. Disclose interests, not just judgments about them. It is not your job to decide whether an interest biased you. State the interest and let editors and readers weigh it.
    3. Cover the defined window for every author. Circulate the disclosure question to all co-authors and collect each person’s declaration against the same time frame.
    4. Name the funder and its role. Funding is a competing interest, and readers are entitled to know whether the funder shaped the design, analysis, or decision to publish. “The funder had no role in…” is itself a meaningful disclosure.
    5. Keep it current. If an interest arises between submission and publication, update the declaration.

    A useful test: imagine the interest being revealed after publication by someone else. If that revelation would embarrass you or undermine trust in the work, it should have been disclosed. Disclose it now.

    Where COI sits among an author’s disclosures

    Competing-interests disclosure is now one of a small family of declarations an author makes at submission, and it helps to see them together. The contribution statement records who did what. The AI-use disclosure records which tools were used and where. The competing-interests declaration records what might bias the reporting. Each answers a different transparency question, and none substitutes for another: a thorough contribution statement does not excuse an undisclosed financial interest, and a clean competing-interests declaration says nothing about contribution. Treat them as a set, completed deliberately rather than copied from a previous paper.

    A note on funding bodies and institutions

    Disclosure obligations do not stop at the journal. Many funders and most institutions operate their own financial conflict-of-interest policies, often with formal thresholds and an annual reporting cycle. These are separate from the publication declaration and can be more demanding. An author with significant financial interests should assume that the institutional disclosure and the journal disclosure are both required, and that they should tell a consistent story.

    Where shared vocabulary fits

    “Conflict of interest”, “competing interest”, “financial interest”, and “funder role” are defined differently across journals, funders, and institutions, and that inconsistency makes a single author’s obligations hard to reconcile. A shared, federated vocabulary that defines these terms precisely — pointing back to ICMJE for the publication standard and to funder and institutional policy for the regulatory layer — is what lets one disclosure serve several systems coherently. Supplying that definitional layer is the role the CASRAI dictionary is designed to play; the relevant terms sit in the compliance-and-regulatory domain.

    Related reading