Publication Misconduct in Research: The Post-Publication Process

Publication misconduct in research discovered after a paper is already public triggers a defined post-publication pipeline: a reader or institution files a complaint, the journal opens a formal investigation guided by COPE, an expression of concern may be issued while it runs, and a retraction or correction follows if misconduct is confirmed.

Publication misconduct is unethical conduct in the publishing process itself — fabrication, falsification, plagiarism, duplicate submission, or fraudulent (ghost, guest or gift) authorship — as distinct from poor research design or an honest error in the underlying study. Once a flawed paper has already been indexed, cited and built upon, the mechanics of fixing the record are entirely different from catching the same problem at peer review. This article sets out what actually happens once that process starts, who runs it, and how the outcome gets communicated to readers and indexers.

• What counts as publication misconduct once a paper is already public?
• What triggers a post-publication review?
• How does the editorial investigation actually run?
• Retraction, correction, or expression of concern — what is the difference?
• Common questions about post-publication misconduct
• What does this mean for institutions and authors?
• Where the retraction pipeline is heading

What counts as publication misconduct once a paper is already public?

Publication misconduct covers conduct in the writing and submission of a paper, not the underlying experiment: fabrication (inventing data), falsification (manipulating images, figures or results), plagiarism, duplicate or redundant publication, and fraudulent authorship such as ghost, guest or gift authorship. The Royal Society’s publishing ethics policy defines research misconduct as “fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results” — and explicitly excludes “honest error or differences of opinion” from that definition.

That exclusion matters at the post-publication stage. An editor’s first job on receiving a complaint is not to assume guilt but to classify the report: is this a correctable mistake, an authorship dispute, or a potential integrity breach that needs a formal case? Only the third category enters the investigation pipeline described below.

What triggers a post-publication review?

Post-publication scrutiny rarely originates with the journal itself. It is most often triggered externally, then formalised internally once a case is opened.

• Reader or third-party complaints — direct emails to the editor, or public flags on post-publication review platforms such as PubPeer.
• Institutional referral — a university research integrity office notifies the journal after its own internal inquiry.
• Journal screening tools — image-manipulation and paper-mill detection software run against submitted or already-published figures.
• Whistleblowers — co-authors, former colleagues or lab members raising concerns directly.
• Funder or sponsor notice — a grant body flags a discrepancy found during its own audit.

Under the ICMJE Recommendations, editors who receive “convincing evidence” of misconduct have a duty to pursue the matter — including by contacting the authors’ institution — rather than simply declining to act on an anonymous tip.

How does the editorial investigation actually run?

Once a case is opened, the journal follows a sequence set out in COPE’s guidance rather than improvising case by case. The Committee on Publication Ethics (COPE) Retraction Guidelines (2019) and its accompanying flowcharts are the reference most editors use to decide what to do next, and in what order.

1. Initial screening — the editor checks whether the complaint is plausible and falls within the journal’s remit.
2. Author response — the corresponding author is asked to explain, and usually to supply raw data, original images or lab records.
3. Institutional referral — if the explanation is unsatisfactory or the allegation is serious, the editor notifies the authors’ institution, which runs its own inquiry under its research integrity policy; the journal does not adjudicate misconduct itself.
4. Interim notice — while the inquiry is unresolved, the journal may publish an editorial expression of concern attached to the paper.
5. Final notice — depending on the outcome, the journal publishes a correction, a retraction, or lifts the expression of concern with no further action.

This sequencing is the core distinction between research-stage and publication-stage misconduct handling: at the research stage, a funder or institution can simply stop a study; at the publication stage, the journal must communicate every step publicly, because the paper is already part of the citable literature.

Retraction, correction, or expression of concern — what is the difference?

These three notice types are not interchangeable, and mixing them up is one of the most common errors in coverage of publication misconduct. Each has a distinct trigger, a distinct effect on the paper’s citability, and a distinct authority that can issue it.

A retracted paper is not deleted. Standard practice, reflected in NISO RP-45-2022, Communication of Retractions, Removals, and Expressions of Concern (CREC), requires the original text to remain online with a clear, permanent watermark and a linked retraction notice explaining the reason — so the scientific record stays transparent rather than simply erased. Since 2023, Crossref has hosted the freely available Retraction Watch Database, giving publishers, institutions and reference managers a shared, machine-readable source of retraction metadata rather than relying on scattered publisher notices alone.

Common questions about post-publication misconduct

What is the difference between a retraction, a correction, and an expression of concern?

A correction fixes a confirmed error while the paper’s conclusions stand. An expression of concern flags an unresolved investigation without prejudging guilt. A retraction is issued once misconduct or unreliable findings are confirmed, permanently marking the paper as withdrawn from the reliable literature while keeping it accessible.

What triggers a post-publication misconduct investigation?

Investigations are usually triggered by a reader complaint, a post-publication review platform flag, an institutional referral, a whistleblower report, or automated screening for image manipulation or paper-mill patterns. Under the ICMJE Recommendations, editors receiving credible evidence have a duty to pursue it rather than dismiss it.

Does a retracted paper disappear from the internet?

No. Under NISO’s CREC recommended practice, retracted articles must remain accessible with a permanent watermark and a linked retraction notice. Removal is reserved for rare cases involving legal risk, such as defamation or serious safety hazards, not ordinary misconduct findings.

Who decides whether a published paper is retracted?

The journal editor makes the final call, but the decision is informed by the authors’ institution, which runs the substantive misconduct inquiry. COPE’s Retraction Guidelines position the journal as the notice-issuing authority and the institution as the fact-finding authority — the two roles are kept deliberately separate.

What does this mean for institutions and authors?

For research administrators, a post-publication complaint is not a journal-only event. Institutions are expected to run a parallel inquiry, respond to editor requests for data within a defined timeframe, and — if misconduct is confirmed — cooperate on the retraction notice’s wording under COPE guidance. Authors named on a retracted paper should expect it to be discoverable through Crossref’s Retraction Watch Database and CrossMark update badges wherever it is cited or indexed, regardless of original host.

Contributor-role clarity also matters here: disputes over who is accountable for which part of a paper are easier to resolve when contributions were recorded precisely at submission. CASRAI’s authorship resources and research administration guidance cover structuring that accountability before a dispute reaches an editor’s desk.

Where the retraction pipeline is heading

Two shifts are changing post-publication misconduct handling. First, machine-readable retraction metadata — now centralised at Crossref rather than scattered across publisher sites — propagates a retraction to citation databases and discovery layers automatically, closing a gap that once left retracted papers silently cited for years. Second, paper-mill detection is shifting discovery earlier, toward pre-publication screening — but the CREC-based post-publication pipeline remains the backstop for everything that gets through regardless. The mark of a well-functioning system is not the absence of retractions; it is a transparent, standardised trail from complaint to notice that any reader can follow.