Have You Checked Whether Major Studies Have Been Corrected or Retracted?

Is the paper you downloaded still the paper the scholarly record says you should trust?

A study does not become permanently frozen at the moment of publication. Journals can issue corrections. Editors can publish expressions of concern while serious questions are investigated. Articles can be corrected and republished, withdrawn, or retracted. These events may occur months or years after you first downloaded the PDF.

That creates an uncomfortable possibility: a paper central to your literature review may have changed status while the copy sitting in your reference manager looks exactly as it did when you saved it.

The practical question is therefore straightforward: before relying heavily on an important study, have you checked its current publication status?

How do you know whether a research paper has been corrected or retracted?

What can happen to a research paper after publication?

The scholarly record is not static

Publication is a milestone, not an immunity certificate. Errors can be discovered. Authors can identify mistakes. Readers may raise concerns. Journals may investigate data, analyses, authorship, ethics, or other aspects of a publication. The response depends on the nature and seriousness of the problem.

Crossref describes the scholarly record as something that can be updated through corrections, retractions, and related changes, while Crossmark is designed to help readers determine the current status of participating content.

This matters because literature reviews often rely on papers long after those papers were first retrieved.

A correction is not the same as a retraction

Post-publication notices need to be read rather than merely counted. Their meaning differs.

The terminology and procedures can vary among publishers and databases. The important habit is to identify the actual notice and understand what the responsible journal or publisher says changed.

An expression of concern is a warning, not a completed verdict

The U.S. National Library of Medicine describes an expression of concern as an editorial notice concerning the integrity of a published article, generally issued while appropriate investigative procedures are pursued. The eventual outcome may include a retraction, but that is not predetermined.

So an expression of concern should neither be ignored nor casually translated into “this paper has been proven wrong.” It tells you that unresolved concerns are serious enough that the editor has formally alerted readers.

A retraction does not automatically tell you why the paper was retracted

Researchers sometimes treat the word retracted as synonymous with fraud. That is too crude. NLM notes that journals may retract articles for reasons including pervasive error or unsubstantiated or irreproducible data, and PubMed does not itself distinguish retractions arising from honest error from those involving misconduct or plagiarism.

The retraction notice therefore matters. Read it rather than inferring the reason from the status alone.

The publisher's current article page is an important first check

If a study matters substantially to your argument, return to its current publisher record rather than relying only on a locally saved PDF. Publishers commonly link corrections, retractions, editorial notes, or other updates from the article page.

Crossmark provides another mechanism when participating publishers use it. Crossref states that the Crossmark button can show whether content has been corrected, retracted, or otherwise updated, and it can also appear within PDFs so that readers can discover changes after downloading them.

A missing Crossmark indicator does not establish that nothing has changed. Not all content participates in Crossmark, so the publisher record and appropriate bibliographic databases remain important.

PubMed links articles with post-publication notices

For literature indexed in PubMed, the National Library of Medicine creates links between original article citations and related records including retraction notices, errata, expressions of concern, corrected and republished articles, and other linked citations.

PubMed also applies publication types such as Retracted Publication to retracted articles and links them to the associated retraction notices. This makes the current PubMed record more informative than an old citation exported before the status changed.

Other databases and reference systems may surface similar information, but their coverage and update practices should not simply be assumed equivalent.

A correction needs interpretation, not panic

Corrections range widely in consequence. A correction might repair an author affiliation or typographical error with no bearing on the study's findings. Another might change numerical values, figures, analyses, or wording central to your interpretation.

The correct response is therefore not “corrected paper equals unreliable paper.” Read the correction. Identify what changed. Then decide whether the specific claim for which you use the study remains supported.

Crossref's versioning guidance emphasizes linking editorially significant updates such as corrections and retractions to the work they update so readers can identify the current scholarly record.

A retracted study can still be discussed, but its role changes

There are legitimate reasons to cite a retracted paper. You may be studying research integrity, the history of a controversy, the diffusion of an erroneous claim, or the effects of retraction itself. In those cases, the retracted article is the object of analysis.

What should change is the way you use it. A retracted study should not silently remain in a literature review as ordinary supporting evidence after you know its status.

Check the studies that matter most, not only suspicious-looking papers

Retractions and corrections are not always visually obvious in a PDF, particularly an old downloaded copy. Waiting until something “looks suspicious” is therefore a poor detection strategy.

Prioritize status checking for studies that materially support your research rationale, theoretical argument, methods, major empirical claims, or conclusions. The more your interpretation depends on one paper, the more consequential an unnoticed post-publication change becomes.

This is especially important when a conclusion depends heavily on one study, research group, dataset, or method.

A retraction can propagate through later literature

Discovering that a paper has been retracted should prompt another question: which later claims depend on it?

A retracted study may continue to be cited by reviews, theoretical papers, guidelines, or empirical studies published before or after the retraction. If it occupies a central position in the argument you are tracing, examine whether later sources merely cite it or whether their conclusions genuinely depend on its findings.

This is another reason to trace consequential claims back to their original evidence. A long citation chain can make the status of the source at its beginning surprisingly easy to overlook.

Status checking complements critical appraisal rather than replacing it

A paper that has never been corrected or retracted is not automatically trustworthy. Most methodological weaknesses do not result in formal editorial notices. Conversely, a corrected paper is not automatically unusable.

Post-publication status is therefore one dimension of evaluation. You still need to critically evaluate the study itself.

What happens when a foundational paper changes status after you save it?

The key action is not merely detecting the notice. It is determining what the notice changes about the evidence you were using.

Common misconceptions about corrections and retractions

What should you do when an important paper has been updated?

Respond to the type and consequence of the update rather than applying one rule to every notice.

For papers that carry substantial argumentative weight, status checking is a modest task with potentially large consequences. Revisit the publisher page, follow linked notices, and use relevant indexing or metadata systems where available.

Then ask the harder question: if the paper can no longer carry the evidential weight you assigned to it, what is now the strongest evidence supporting the claim?

Have your key papers changed since you first found them?

Questions about corrected and retracted research

Publication status can change after the PDF reaches your library

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.