How Authorship Disputes Get Resolved

When authors disagree about who should be credited on a paper, the first question most guidance sidesteps is a procedural one: who actually has the authority to settle it, and how does that process actually run? This guide covers the institutional-process side specifically — jurisdiction, the escalation ladder from informal talk to formal adjudication, what evidence gets weighed, realistic timelines, and what outcomes (and appeals) look like. It does not cover author-order conventions or prevention strategies, which CASRAI covers in Resolving Authorship Order Disputes, and it does not cover the separate question of when a dispute crosses into formal research misconduct, which is covered in When Does an Authorship Dispute Cross the Line Into Research Misconduct? Most authorship disputes never reach that threshold — this is about how the ordinary, non-misconduct kind actually gets worked out.

Who actually has jurisdiction

The consistent position across standards bodies is that authorship qualification is the author group’s own responsibility first, and the institution’s responsibility if the group can’t agree — not the journal’s. The International Committee of Medical Journal Editors (ICMJE) states this directly in its authorship recommendations: “it is the collective responsibility of the authors, not the journal to which the work is submitted, to determine that all people named as authors meet all four criteria; it is not the role of journal editors to determine who qualifies or does not qualify for authorship.” The Committee on Publication Ethics (COPE) takes the same position from the editor’s side of the desk: its authorship guidance and case resources direct editors not to adjudicate authorship disputes themselves, but to pause processing and refer an unresolved dispute back to the authors’ institution.

That leaves a clear jurisdictional order for most disputes that don’t involve an allegation of misconduct: the author group first, a neutral mediator second if direct discussion stalls, and the institution (or institutions, in a multi-site collaboration) as the formal decision-making venue if neither of those resolves it. Journals and publishers sit outside this chain as record-keepers, not adjudicators — they can pause a submission, decline to proceed, or (post-publication) issue an Expression of Concern while an institutional process runs, but they don’t make the authorship decision themselves.

The escalation ladder most institutional processes follow

UK Research Integrity Office (UKRIO)’s Model Authorship Dispute Procedure (published September 2025, commissioned by UK Research and Innovation) is one of the few publicly documented, step-by-step institutional models available, and it lays out a three-stage structure that mirrors what most university research-integrity offices describe in less formal terms:

Stage 1: Direct, informal discussion

Resolution is attempted at the most local level first — the person who raised the concern (UKRIO’s model calls this person the “Organiser”) invites the other contributors to an informal discussion, ideally with the points of dispute listed in advance and any existing authorship agreement or contribution record on hand. The model’s advice is to aim for a decision during the meeting itself, or to agree a clear short timeline (UKRIO suggests around 10 working days) for a follow-up decision if not.

Stage 2: Informal resolution with a neutral mediator

If direct discussion doesn’t resolve it, or there’s a real or perceived power imbalance (a common factor in student–supervisor disputes), the next step is a neutral, non-decision-making Mediator — often sourced through HR or a trusted senior colleague with no stake in the outcome. The mediator’s role is to facilitate the conversation, not to decide it; UKRIO’s model recommends escalating to formal adjudication if this stage hasn’t produced resolution within roughly 30 working days.

Stage 3: Formal investigation and adjudication

Where informal routes fail or aren’t suitable, a designated decision-maker — UKRIO’s model calls this the “Responsible Person,” typically the institution’s research integrity lead or another senior, conflict-free official — takes over. This person collects evidence from all parties, may interview contributors, and issues a binding decision. Formal adjudication is explicitly meant to be reserved for genuine authorship disagreements, not as a venue for unrelated interpersonal conflict — UKRIO’s model is explicit that this stage “should not be used frivolously to address relationship tensions, personality clashes, or conflicts or differences of academic opinion.”

A second real-world model: NIH’s Intramural Research Program

The NIH Office of Intramural Research (OIR) publishes its own authorship dispute resolution process, structured similarly in principle but with different institutional machinery. NIH’s policy splits the process into an informal stage and a formal adjudication stage:

Direct dialogue — the parties discuss their perspectives directly, guided by NIH’s Guidelines for the Conduct of Research in the Intramural Research Program.
Ombudsman mediation — parties can bring the dispute to NIH’s Office of the Ombudsman for confidential, informal mediation.
Voluntary Peer Panel — as an alternative informal-stage option, parties can agree in writing to present their case to a panel of three NIH scientists with relevant subject-matter expertise and to be bound by the panel’s decision.
Scientific Director adjudication — if none of the above resolves the dispute, or the parties decline the Peer Panel route, the institute or center’s Scientific Director makes a binding decision.

The structural difference worth noting: NIH offers a peer-panel option (subject-matter experts, jointly agreed in advance to be binding) that UKRIO’s generic institutional model doesn’t include, reflecting NIH’s specific capacity to convene scientist panels internally. Both models converge on the same underlying shape — informal resolution attempted first, a neutral third party available before anything formal, and a single senior official empowered to make a binding call if that fails.

Who’s who in a formal process

The party raising the concern — initiates discussion and, if needed, formally requests the institutional process.
Research contributors — everyone whose contribution is in question, given the opportunity to submit evidence and respond.
An impartial party — someone who can accompany a contributor who feels vulnerable or facing a power imbalance, without arguing their case for them.
A mediator — neutral, facilitates discussion, does not decide the outcome.
A responsible person / adjudicator — the institution’s research-integrity lead, a senior official, or (at NIH) the Scientific Director — the only role in the process actually empowered to decide the dispute.

What evidence actually gets weighed

Formal adjudication isn’t a debate about impressions — it runs on documentation. Per UKRIO’s model, a Responsible Person typically expects to see:

Any existing authorship agreement (an authorship strategy agreement or CRediT statement drafted at project outset)
A written summary of the dispute and a record of prior informal-resolution attempts
Documents tracking contribution over time — lab notebooks, project management records, correspondence, meeting minutes
A list of contributors with their roles and how each is claimed to meet the institution’s or journal’s authorship criteria
The manuscript or output itself, at whatever stage it’s in (draft, submitted, under review, or published)
The desired outcome each party is seeking, and their rationale

This is precisely why documenting contributions using a structured framework like the 14 CRediT roles — ideally with the lead/equal/supporting qualifier — matters well beyond the byline itself: it’s the exact kind of contemporaneous record an adjudicator weighs most heavily, because it was created before the dispute existed, not written to justify a position after the fact.

Realistic timelines

Institutions vary, but UKRIO’s published model gives concrete, citable benchmarks that are broadly representative of how these processes are structured elsewhere:

Informal discussion (Stage 1): a decision at the meeting itself, or an agreed follow-up within roughly 10 working days
Informal mediation (Stage 2): escalate to formal process if unresolved after roughly 30 working days from the first mediated meeting
Formal process, initial review: the Responsible Person allocates up to 10 working days to review materials and decide whether formal adjudication is the right route (versus, e.g., referral to an HR or misconduct process instead)
Formal adjudication: a binding decision within 20 working days of receiving all materials from the parties

End to end, a dispute that runs the full ladder — informal discussion, mediation, and full formal adjudication — can reasonably take two to three months under a documented model like this one, longer if the case is complex, cross-institutional, or a party is slow to engage. NIH’s OIR process does not publish equivalent public timeline figures for its informal-mediation or Scientific Director stages.

Outcomes, and whether a decision can be appealed

The realistic set of outcomes from formal adjudication includes: a revised author list or order, no change to the original list, a documented resolution the parties agree to going forward, or — if the evidence raises concerns beyond a straightforward authorship disagreement — referral to a separate HR grievance process or a formal research-misconduct investigation instead. UKRIO’s model is explicit on this last point: if the Responsible Person’s initial review turns up evidence suggestive of misconduct (harassment, discrimination, undisclosed conflicts of interest, or breaches of research integrity), Stage 3 as an authorship process is deemed the wrong venue, and the matter is redirected to the institution’s appropriate procedure. That’s the same threshold question covered in more depth in CASRAI’s misconduct-threshold guide — the fact-fabrication/falsification/plagiarism line, not the authorship question itself, is what triggers that redirect.

On appeals specifically: UKRIO’s published model states plainly that the Responsible Person’s decision “is final and should be accepted by all parties,” and that no appeals process is built into the model itself. That’s a design choice for that particular model, not a universal rule — individual institutions adapting it, or running their own separate policy, may route a further challenge through a standard grievance or disciplinary-appeal mechanism instead of a dedicated authorship appeal. The practical takeaway: don’t assume an appeal route exists by default. Check the specific institution’s research-integrity or authorship policy (not just publisher guidance) before assuming a formal adjudication decision can be revisited.

If the dispute surfaces after publication

The jurisdictional order doesn’t change once a paper is published, but the mechanics do. Per UKRIO’s guidance on the publisher’s role, a journal is not the decision-maker even after publication — it will typically ask the authors and their institution(s) whether a resolution process is already underway, and if the dispute is unresolved or not being addressed in a reasonable time, the journal will likely publish an Expression of Concern flagging the unresolved authorship question while the institutional investigation runs. Amending a byline after publication — a corrigendum — normally still requires all parties’ agreement or a clear institutional determination behind it; publishers won’t make that call unilaterally.

Preventing disputes before they need any of this

None of the above is quick, and every stage exists specifically because prevention didn’t happen. CASRAI’s guide to resolving authorship order disputes covers the practical prevention side in depth — drafting an authorship agreement at project kickoff, using CRediT roles as a running record rather than a submission-time afterthought, and revisiting the agreement at real project milestones. The institutional process in this guide is the fallback for when that didn’t happen, or wasn’t enough — not a substitute for it.

Frequently asked questions

Can a journal editor resolve an authorship dispute?

No. ICMJE and COPE both place that responsibility with the author group first and the institution second — editors can pause, query, or decline to proceed with a submission, and can refer an unresolved dispute to the authors’ institution, but they don’t adjudicate who should be listed as an author.

Who has the final say in an authorship dispute?

Within a documented institutional model like UKRIO’s, a designated senior official — a research-integrity lead or other conflict-free “Responsible Person” — makes the binding decision at the formal stage. At NIH’s Intramural Research Program, that role sits with the institute or center’s Scientific Director. In both cases, it’s an institutional role, not the journal and not the disputing parties themselves once the matter reaches formal adjudication.

How long does an authorship dispute take to resolve?

It depends heavily on which stage resolves it. Under UKRIO’s published model, informal discussion is meant to conclude within about 10 working days, informal mediation within about 30 working days if it’s going to work at all, and formal adjudication within 20 working days of the adjudicator receiving all the evidence — meaning a dispute that runs the full ladder can reasonably take two to three months.

Can a formal authorship decision be appealed?

Not automatically. UKRIO’s model treats the Responsible Person’s decision as final with no built-in appeal step. Whether a further appeal is available depends on the specific institution’s own policy — check that policy directly rather than assuming either that an appeal exists or that it doesn’t.

What happens if the dispute involves more than a disagreement about credit?

If the evidence raised during a formal review suggests something like harassment, discrimination, an undisclosed conflict of interest, or fabricated/falsified contribution records, institutional models redirect the case out of the ordinary authorship-dispute process and into the relevant HR or research-misconduct procedure instead. See CASRAI’s guide on when an authorship dispute crosses into research misconduct for that specific threshold.

Related CASRAI resources

Topic cluster

CRediT & Authorship Attribution — the full cluster hub for CRediT taxonomy, roles, and authorship-attribution content.

How Authorship Disputes Get Resolved: The Institutional Process