Peer Reviewer Ethics: Conflicts of Interest, Confidentiality, and Timeliness

Reviewer Ethics

Every peer review invitation you accept carries three promises, whether or not anyone spells them out. You are promising that the manuscript stays confidential, that your judgement is independent, and that your report arrives on time. Peer reviewer ethics is simply the practice of keeping those three promises — and knowing what to do the moment you realise you cannot.

This guide covers the four categories of conflict of interest, the exact three-step response when one applies, the five confidentiality obligations that begin the second you accept, how to handle a deadline you can no longer meet, and the four situations where declining is the right professional call.

Why reviewer ethics matters

The peer review system runs on three assumptions that hold at the same time:

The manuscript is confidential. It is unpublished research, protected until the paper appears in print.
The reviewer is independent. There is no undisclosed relationship with the authors that could bias the assessment.
The review is delivered on time. Authors and editors are waiting, and delays have direct consequences for both.

When any one of these breaks, the failure is not private. It reaches the authors whose careers depend on a timely decision, the journal whose reputation rests on a defensible process, and the scientific record itself.

These are not bureaucratic rules. They are the minimum conditions for peer review to function with integrity.

Conflicts of interest in peer review

A peer review conflict of interest is any relationship that might compromise — or appear to compromise — the independence of a review. The appearance matters as much as the reality, because peer review is a trust system, and trust is damaged by what looks wrong just as surely as by what is wrong.

There are four categories.

How to handle a conflict of interest — three steps

The response is the same in every case, and the order matters.

Step 1 — Identify. Ask whether any conflict category applies. Do this before reading the manuscript, not after. Once you have read an unpublished paper, you cannot unread it, and a conflict declared at that point is a much messier problem for everyone.

Step 2 — Declare. Contact the editor immediately and explain the conflict plainly. You do not need a long justification. Editors handle these disclosures routinely — it is an ordinary part of the job, not an accusation or an admission.

Step 3 — Follow. Accept the editor’s decision. They may ask you to decline, they may ask you to proceed with the conflict on record, or they may transfer the manuscript to another reviewer. That call belongs to the editor, not to you.

The principle: when in doubt, declare it. Editors would rather know.

Confidentiality obligations in peer review

Accepting a review invitation creates five specific obligations. Peer review confidentiality is not a general attitude of discretion — it is this list.

Do not share the manuscript with colleagues, students, or lab members without the editor’s explicit permission. If you want to involve a junior colleague as a training exercise, ask first — most editors will agree, and most require the co-reviewer to be named.
Do not use the ideas, data, or methods in your own work before publication. This is the most serious breach in the entire list, because it converts privileged access into personal advantage.
Do not cite the paper until it appears in print.
Treat it as unpublished — protect it exactly the way you would protect your own pre-submission work.
Keep the review itself confidential, in most systems, including most open review formats. Where a journal publishes reviewer reports, that decision is the journal’s to make and announce, not yours.

These obligations do not expire when you submit your report. They run until the paper is published — and if the paper is rejected and never published, they do not expire at all.

Timeliness — the ethics of the deadline

Late reviews are the most common ethical failure in peer review, and the easiest to dismiss as harmless. They are not harmless. A late review leaves authors in limbo, leaves editors unable to decide, and stretches the journal’s timelines for every other paper in the queue.

Here is what to do at each stage:

Stage

Before accepting

After accepting, if things change

At the deadline

After the deadline

The right action

Check the deadline honestly. Can it be met given your current commitments — not your ideal week?

Contact the editor immediately. Not at the deadline — the moment the problem becomes apparent.

Submit what is ready, or request a short extension with a specific date. Partial is better than silence.

Contact the editor immediately. Silence is the worst possible outcome for everyone in the process.

The through-line is that declining early is always more professional than accepting and delivering late. Editors can replace a reviewer who declines in two days. They cannot easily replace one who goes quiet for six weeks.

When to decline a peer review invitation

Four situations make declining the right and professional call.

An undisclosed conflict. Any conflict-of-interest category applies and has not yet been declared to the editor.
The paper is outside your specific expertise. The manuscript requires expertise in particular methods or subject matter you do not hold. General familiarity with the field is not the same thing. A reviewer who cannot evaluate the statistics in a paper should say so rather than reviewing around them.
You cannot meet the deadline. Assess this before you accept, not after. An honest look at the next three weeks is worth more than good intentions.
Duplicate review. You have already reviewed the same manuscript for another journal. Most journals prohibit this outright, and all of them require it to be declared. If you have seen the paper before, tell the editor — your prior view of an earlier version is exactly the kind of context they need.

Timing turns a decline into a professional act or an unprofessional one. Declining within 48 hours is a professional act. Declining after two weeks, or not at all, is not.

The three principles, in short

Declare. Protect. Decline. Respond. Submit. Early.

Declare every conflict — before reading the manuscript.
Protect the manuscript — it is unpublished and confidential.
Respond on time — and immediately if circumstances change.

Reviewers who follow these three principles become the reviewers editors return to. That is not a coincidence — reliability under an ethical standard is the whole of a reviewer’s professional reputation.

Frequently asked questions

Is being in the same field as the authors a conflict of interest? No. Shared field is why you were invited. It becomes a conflict when you work on the same specific research question, compete for the same funding, or have a public prior disagreement with the authors — the professional category above.

Can I ask a PhD student to help with a review? Only with the editor’s explicit permission, and the student should be named. Involving a co-reviewer without asking breaches the first confidentiality obligation, even when your intent is purely educational.

What if I discover a conflict after I have started reading? Stop and contact the editor immediately. Declaring late is far better than not declaring, and the editor needs to know how much of the manuscript you have seen.

Do I have to explain why I am declining? No, but a one-line reason helps the editor. “Outside my methods expertise” or “conflict — former collaborator” takes five seconds and makes their next invitation better targeted.

How long do confidentiality obligations last? Until the paper is published. If it is never published, they do not end.