For editors

What an editor is responsible for

Assessing scope and readiness

• Is the ENT contribution central?

  A central, substantive ENT contribution establishes scope eligibility. Methods and evaluation must support the claims and be appropriate to the study type; clinical deployment or demonstrated patient benefit is not required for every laboratory, mechanistic or early-stage device study. Use the approved Aims and Scope, including audiology, skull-base and functional/reconstructive facial surgery, when screening.

• Can the design support the claim?

  Read the abstract’s conclusion against the Methods. Where a case series claims efficacy, a bench study claims clinical benefit, or a cross-sectional study claims causation, the mismatch is the substantive issue and should be raised at screening rather than left to review.

• Are the required statements present?

  Ethics approval or exemption, informed consent, consent to publication for identifiable material, trial registration, data availability, funding and competing interests. A manuscript missing one of these is held rather than reviewed, and the author is told which is missing.

• Is the reporting complete enough to review?

  Check that the reporting guideline for the design has been followed and its checklist supplied. A reviewer cannot assess a trial with no participant flow or an observational study with no account of confounding.

• Declining at screening

  A manuscript may be declined without external review. Say why in one or two specific sentences: scope, design, or a requirement that cannot be met. A generic decline is the least useful thing an editor can send.

Selecting and managing reviewers

• How many

  Manuscripts that proceed to external peer review are normally evaluated by at least two independent subject-matter experts. Where two reports diverge substantially, seek a third rather than deciding between them alone.

• Matching expertise

  Match the reviewer to what the manuscript actually turns on. A surgical-technique paper with a survival analysis needs someone who can assess the operation and someone who can assess the statistics; one reviewer rarely covers both.

• Independence

  Avoid reviewers at the authors’ institution, recent co-authors, current collaborators, trainees and supervisors. Author-suggested reviewers may be considered, but the choice is the editor’s, and a panel drawn entirely from suggestions is not independent.

• Briefing

  Tell the reviewer what you want assessed and where you have doubts, and give the article type so the expectations are the right ones. Send the reporting checklist with the manuscript.

• Review model and double-blind handling

  Peer review is single-blind by default; double-blind review is available on request. When double-blind review is requested, editors must ensure that author-identifying information is not shared with reviewers. Check the reviewer file before it is sent: acknowledgments, funding text, institutional ethics references and file metadata are the usual places identity survives.

• Keeping it moving

  Follow up an overdue review rather than waiting, and tell the editorial office when a manuscript is stalled so an alternative reviewer can be found. Authors should hear from the journal about a delay before they have to ask about it.

• Handling a poor review

  A review that is abusive, that makes no substantive point, or that appears to be aimed at delaying or appropriating the work is not passed to the author. Set it aside, tell the editorial office, and commission another.

Competing interests, recusal and confidentiality

• When to step aside

  Decline where you are an author or a contributor to the work; where the authors are colleagues at your institution, recent co-authors, trainees or supervisors; where you have a financial interest in the product, device or company studied; or where a personal or professional relationship would make a reader doubt the decision.

• A manuscript you have written

  Authorship of a submitted manuscript is a conflict that requires the assignment to be declined. The manuscript is assigned to another editor, and the author-editor has no access to the reviewers’ identities, the reports, the correspondence or the decision, and takes no part in it.

• Confidentiality

  A submitted manuscript is a confidential document. Do not disclose its existence or contents, do not use its data or ideas before publication, and do not share it with a third party without the editorial office’s agreement. Reviewers’ identities are not disclosed to authors.

• AI tools

  Reviewers must obtain journal permission before using AI to assist with a review. Any permitted use must be disclosed, and confidential manuscript content must not be uploaded where confidentiality cannot be assured. The same restriction applies to an editor: an unpublished manuscript is not uploaded to an external service where its confidentiality cannot be assured.

Making and writing the decision

• Read the reviews, do not count them

  Reviewers advise; the editor decides. A single well-argued objection can outweigh two approving reports, and a recommendation unsupported by reasons should not carry weight simply because it is confident.

• Say what must change

  A revision request distinguishes what is essential from what is optional, and states each essential point in a form the author can act on. “Strengthen the discussion” is not actionable; “report the confidence intervals for the primary outcome and state how the 14 losses to follow-up were handled” is.

• Resolve conflicting reviews for the author

  Where reviewers disagree, say which points the author must address and which they may answer in the response letter. Sending two contradictory reports with no guidance passes the editor’s work to the author.

• Assess the revision against the request

  Check the response letter point by point. A reasoned disagreement is a legitimate response and may be accepted. Where the changes are substantial, return the manuscript to the original reviewers.

• Declining well

  Give the reason. An author who is told what was wrong can improve the work; an author who is told nothing submits the same manuscript elsewhere. Never characterize the authors, and never speculate about their motives.

• Record the reasoning

  The decision letter and the editorial notes are the record of why a manuscript was accepted or declined, and they are what an appeal or a later concern is assessed against.

Research-integrity responsibilities

• Similarity reports

  Read the matches. A similarity report requires editorial judgment and is never an automatic threshold: matched text may be a standard method sentence, an instrument name or a quotation, or it may be plagiarism, and only reading distinguishes them.

• Images and data

  Where a figure appears spliced, duplicated or selectively altered, or where reported numbers are internally inconsistent, request the original unprocessed images or the underlying data before drawing a conclusion.

• Ethics and consent

  Where approval, consent or trial registration is absent, late or inconsistent with the reported dates, ask the authors for the documentation and record the answer.

• Redundant publication

  Where a manuscript appears to overlap an earlier publication by the same authors, ask for both and for an account of the overlap before deciding.

• Concerns about a published article

  Follow the COPE flowcharts through the editorial office. The corresponding author, and where necessary the authors’ institution, is given the opportunity to respond before any notice is published. Outcomes are a correction, an expression of concern or a retraction, as set out in the editorial policies.

Appeals and complaints

• Who considers an appeal

  An appeal is considered by an editor who was not responsible for the decision. If you made the decision under appeal, hand the file to the editorial office rather than answering it yourself.

• What an appeal turns on

  Whether the decision rested on a misunderstanding of the manuscript, on a review that was factually wrong, or on a failure to follow the journal’s policies. A disagreement with the editorial judgment, without new argument or evidence, is not a ground for appeal.

• Complaints about handling

  Complaints about delay, conduct or process go to the editorial office and are handled separately from the scientific decision, so that neither influences the other.

Conduct and communication

• Write to authors and reviewers as colleagues, whatever the decision
• Keep every judgment about a manuscript inside the manuscript — never about the people who wrote it
• Assess the work of trainees, of authors writing in a second language, and of authors from institutions you do not recognize on exactly the terms you assess anyone else’s
• Correspond through the journal’s systems, so the record of a decision is complete
• Keep the editorial board record current — affiliation, specialty and ORCID iD — through the editorial office
• Tell the editorial office when you are unavailable, rather than letting an assignment stand
• Read the editorial policies and the aims and scope as they are published: they are what an author is entitled to rely on