
A connection is not automatically disqualifying, but silence is not neutral. The editor needs enough detail to judge whether your review can remain independent.
The decision belongs to both reviewer and editor
A conflict of interest is not limited to receiving money from a company whose product appears in a manuscript. Professional relationships, institutional ties, personal relationships, intellectual commitments, litigation, public advocacy, and direct competition can all affect judgment or confidence in the process.
The first decision is yours: some situations make an independent review unrealistic, so you should decline. The second belongs to the editor: uncertain or potentially manageable connections should be disclosed with enough context for the journal to apply its policy. “I can be objective” is not a substitute for disclosure.
The COPE Ethical Guidelines for Peer Reviewers (Version 2, September 2017) tell reviewers to declare all potential competing interests and to raise any they are unsure about. They also set firmer lines. If you are currently employed at the same institution as any author, or have been a recent (for example, within the past three years) mentor, mentee, close collaborator or joint grant holder, COPE says you should not agree to review. Nor should you review a manuscript very similar to one you have in preparation or under consideration at another journal, and you should agree only if you can be unbiased. The ICMJE's responsibilities in the submission and peer review process say reviewers should declare relationships and activities that might bias their evaluation and recuse themselves if a conflict exists.
That puts current institutional ties and recent mentoring, collaboration and joint grants on the decline side, not merely the disclose side. The triage below is consistent with that guidance. Journal policies may differ, for example by setting a different look-back period or naming other disqualifying relationships, and where they do, the journal's policy decides.
A three-question conflict test
Ask these questions before accepting:
- Connection: Do I have a professional, financial, personal, institutional, or intellectual connection to an author, sponsor, product, method, or outcome?
- Effect: Could it influence how I evaluate the work, consciously or otherwise, or create a meaningful personal gain or loss?
- Confidence: If the connection became known after publication, could a reasonable editor, author, or reader question the review's independence?
The third question matters because peer review relies on trust as well as private intention. A reviewer may sincerely believe they can be fair while overlooking how a recent dispute, promotion decision, commercial relationship, or competing manuscript appears to others.
Use this triage:
| Situation | Default response | Why |
|---|---|---|
| Same institution as an author | Decline and disclose | Local relationships and interests are difficult to exclude |
| Current or recent close collaborator | Decline and disclose | Shared work creates loyalty and mutual interests |
| Current or recent mentor, mentee, or supervisor | Decline and disclose | Power and credit relationships impair independence |
| Direct financial interest in outcome | Decline and disclose | The decision may affect personal or organizational gain |
| Active personal dispute or close relationship | Decline and disclose | Strong negative or positive bias is plausible |
| Same broad field with no close connection | Usually eligible | Relevant expertise necessarily overlaps |
| Distant past collaboration | Disclose if still relevant | Recency, closeness, and continuing ties matter |
| Publicly stated scientific position | Consider and disclose | Expertise and intellectual commitment must be separated |
“Default” does not override a journal's rules. If its policy imposes a defined exclusion period or names a disqualifying relationship, follow it.
Conflicts that normally mean decline
Current institutional ties
Working at the same university, hospital, research institute, or company as an author is normally enough to decline. Large organizations may have separate campuses and no daily contact, but editors cannot infer those boundaries from an affiliation. State the shared institution and let the journal apply any exception.
Collaboration, supervision, and grants
Current coauthors, close collaborators, grant partners, supervisors, and trainees have shared reputational or career interests. Recent relationships matter even when a project has ended. COPE gives recent (for example, within the past three years) close collaboration or joint grant holding as relationships where a reviewer should not agree to review. The three-year figure is illustrative rather than a universal limitation period, and journals may set their own.
Describe the relationship rather than merely saying “conflict”:
I coauthored two papers with Dr [Name], most recently in [year], and we remain collaborators on [general project]. I must decline.
This helps an editor distinguish a substantive tie from appearing once among hundreds of contributors in an old consortium paper.
Financial and organizational interests
Decline when the manuscript's outcome could materially affect a company, patent, paid advisory role, legal matter, guideline panel, advocacy organization, or product in which you have a relevant interest. Include interests held through your department when they directly connect to the work, even if payment does not reach you personally.
Do not decide that a payment is too small to matter without checking policy. Give the editor the type of relationship, organization, relevance, and whether it is current.
Direct competition and confidential overlap
Researchers in the same niche are not automatically conflicted. However, you should decline if you are preparing a very similar manuscript, contesting priority, involved in an active dispute, or likely to gain an unfair advantage from seeing the work. Confidential peer review cannot become advance intelligence for your own study.
The complete guide to peer review in medical journals explains the confidentiality duty and why editors need a factual record when misuse is suspected. Prevention begins with reviewers identifying direct competition before acceptance.
Connections to disclose for editorial judgment
Some connections are real but not necessarily disqualifying:
- A distant collaboration that ended years ago
- Membership in the same large professional society or consortium
- Service on a broad committee without close interaction
- A published disagreement conducted professionally
- Work using the same method without a direct priority race
- Recognition of an author from the topic or references
- A public position on the policy question under study
Send a concise disclosure:
Before accepting, I would like to disclose that Dr [Name] and I served on [committee] in [years], without collaborating on research. I do not believe this would prevent an independent review, but I would like you to decide under the journal's policy.
State facts, not a legal argument. Include dates, closeness, whether work is ongoing, and how the connection relates to the manuscript. Then wait for written confirmation.
An editor may proceed because specialized fields have a small expert community. The editor may instead invite someone else, pair you with a reviewer who has complementary independence, or limit the aspects you assess. None of these outcomes is a judgment on your integrity.
Intellectual interests require judgment
Every qualified reviewer has prior views. Publishing a method, treatment approach, or theory does not by itself make someone unable to review related work. The problem arises when identity, advocacy, rivalry, or a material interest makes contrary evidence unusually hard to evaluate fairly.
Ask:
- Have I publicly promised a particular conclusion regardless of new evidence?
- Would acceptance threaten a product, patent, program, or claim central to my work?
- Am I in an active dispute with the authors or their group?
- Can I describe evidence that would change my view?
- Can I evaluate the work without requesting citations or analyses mainly benefiting me?
If the answers raise doubt, disclose or decline. Editors need knowledgeable critics, but not reviewers committed to defeating or promoting a conclusion.
Recognizing authors in anonymous review
Double-anonymous review reduces identity information; it cannot ensure that specialists will never infer authorship. A cohort name, registry number, institutional ethics committee, writing style, preprint, or self-citations may provide clues.
Do not search for confirmation or contact the suspected author. Recognition requires action only when the suspected identity reveals a relevant relationship or interest. Tell the editor:
I believe I may recognize the research group from the cohort and references. If my inference is correct, I had a collaboration with that group ending in [year]. Please advise whether I should continue.
Do not put the guess in comments to authors. If no relevant connection exists, follow the journal's policy and review the work presented.
When a conflict appears after acceptance
Stop reading and contact the editor. Conflicts emerge when an anonymized identity becomes apparent, a new collaboration starts, a patent interest is disclosed in a supplement, or a similar paper enters your own group.
Report:
- What changed or what you learned
- The relationship or interest and relevant dates
- How much of the manuscript you accessed
- Whether any notes or copies exist
- That you will wait for instructions
Do not submit a partial report to justify time already spent. Do not use manuscript information in your own work. If withdrawn, delete copies as instructed and preserve only the administrative correspondence the journal permits.
What editors should document
Senior authors often become editors, and consistent handling matters. For each disclosed connection, record the facts supplied, policy applied, decision, decision-maker, and any management step. Avoid informal exceptions for prominent reviewers.
| Editorial decision | Appropriate when | Record |
|---|---|---|
| Decline reviewer | Independence is materially compromised | Conflict category and reassignment |
| Permit review | Connection is remote and policy allows | Disclosure and editor's rationale |
| Narrow scope | Reviewer can assess one independent component | Assigned component and limits |
| Seek another opinion | Expertise is scarce or views are polarized | Why complementary review is needed |
Reviewer recommendations remain advisory. Our explanation of journal decision letters shows why the editor must integrate reports rather than count votes.
Disclosure wording that works
Use a secure journal field where available. A useful disclosure is short but specific:
I am currently a coinvestigator with Author A on a grant unrelated to this manuscript. The grant runs through [year]. Because the collaboration is active, I must decline.
I served with Author B on a national guideline panel that ended in [year]. We did not collaborate on research. I can review independently, but I am disclosing the connection for your decision.
My laboratory is preparing work on the same intervention and primary outcome in a different population. The overlap creates direct competition, so I must decline.
Avoid vague formulations such as “I know the authors” or “no conflict as far as I am concerned.” Also avoid forwarding invitation details to colleagues while asking whether they think you are conflicted; the editor is the proper recipient.
Before accepting, complete this check:
- I considered every author, not only the corresponding author
- I checked current institution, collaborations, supervision, and grants
- I considered financial, organizational, legal, and personal interests
- I considered active competition and similar unpublished work
- I applied the journal's stated policy
- I disclosed uncertainty with dates and context
- I waited for the editor's decision
The goal is not a reviewer with no prior knowledge or opinions. It is a review whose relevant connections are visible, assessed, and compatible with independent judgment.
Frequently asked questions
Is working at the same institution always a reviewer conflict of interest?
It is generally a strong reason to decline, even at a large institution, because employment links and local relationships may not be visible to the editor. Journal policies can be more specific. Disclose the shared institution immediately and do not review unless the editor explicitly confirms that its policy permits it.
How recent must a collaboration be to create a conflict?
There is no universal interval. COPE gives recent close collaboration or joint grant holding within the past three years as examples, but journals may use different periods and context matters. State the nature and date of the collaboration so the editor can apply the journal's policy rather than relying on a private cutoff.
Is scientific competition itself a conflict of interest?
Working in the same field is not automatically a conflict; journals need reviewers with relevant expertise. Decline when direct competition, access to similar unpublished work, a priority dispute, or a strong personal stake could impair objectivity or create a risk of using confidential information. Disclose uncertainty to the editor.
What if a reviewer recognizes an author in double-anonymous review?
Recognition alone does not necessarily require withdrawal. Tell the editor if recognition reveals a relationship, rivalry, or interest that could affect the review. Do not try to confirm the identity, contact the suspected author, or mention the guess in comments to authors.
Should reviewer conflict disclosures be sent to the authors?
Follow the journal's process. Potential conflicts are disclosed to the editor, through a secure journal field where available, and the editor decides whether the reviewer should serve. The editor, not the authors, is the proper recipient.